url
stringlengths 52
54
| repository_url
stringclasses 1
value | labels_url
stringlengths 66
68
| comments_url
stringlengths 61
63
| events_url
stringlengths 59
61
| html_url
stringlengths 42
44
| id
int64 203M
2.43B
| node_id
stringlengths 18
24
| number
int64 1
141
| title
stringlengths 12
122
| user
dict | labels
sequencelengths 0
0
| state
stringclasses 2
values | locked
bool 1
class | assignee
null | assignees
sequencelengths 0
0
| milestone
null | comments
sequencelengths 0
30
| created_at
timestamp[s] | updated_at
timestamp[s] | closed_at
timestamp[s] | author_association
stringclasses 2
values | active_lock_reason
null | draft
bool 0
classes | pull_request
dict | body
stringlengths 0
17.4k
⌀ | reactions
dict | timeline_url
stringlengths 61
63
| performed_via_github_app
null | state_reason
stringclasses 1
value |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
https://api.github.com/repos/usnistgov/frvt/issues/141 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/141/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/141/comments | https://api.github.com/repos/usnistgov/frvt/issues/141/events | https://github.com/usnistgov/frvt/issues/141 | 2,432,289,049 | I_kwDOBIKFiM6Q-ckZ | 141 | Possibility to relax Ubuntu 20.04 minor version requirement | {
"login": "alexkrz",
"id": 38694722,
"node_id": "MDQ6VXNlcjM4Njk0NzIy",
"avatar_url": "https://avatars.githubusercontent.com/u/38694722?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/alexkrz",
"html_url": "https://github.com/alexkrz",
"followers_url": "https://api.github.com/users/alexkrz/followers",
"following_url": "https://api.github.com/users/alexkrz/following{/other_user}",
"gists_url": "https://api.github.com/users/alexkrz/gists{/gist_id}",
"starred_url": "https://api.github.com/users/alexkrz/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/alexkrz/subscriptions",
"organizations_url": "https://api.github.com/users/alexkrz/orgs",
"repos_url": "https://api.github.com/users/alexkrz/repos",
"events_url": "https://api.github.com/users/alexkrz/events{/privacy}",
"received_events_url": "https://api.github.com/users/alexkrz/received_events",
"type": "User",
"site_admin": false
} | [] | open | false | null | [] | null | [] | 2024-07-26T13:50:07 | 2024-07-26T13:50:07 | null | NONE | null | null | null | Hi,
I am wondering whether it is necessary to specifically check for Ubuntu version 20.04.3 LTS.
The currently supported version is Ubuntu 20.04.6 and contains lots of important security updates.
Since the minor version changes should not have a relevant impact on the included packages, I would advocate to allow all Ubuntu 20.04.x versions.
I will create a PR and hope that the PR can be accepted.
Kind regards | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/141/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/141/timeline | null | null |
https://api.github.com/repos/usnistgov/frvt/issues/139 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/139/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/139/comments | https://api.github.com/repos/usnistgov/frvt/issues/139/events | https://github.com/usnistgov/frvt/issues/139 | 2,408,085,118 | I_kwDOBIKFiM6PiHZ- | 139 | incorrect path to some images for age-estimation | {
"login": "xerus",
"id": 2479721,
"node_id": "MDQ6VXNlcjI0Nzk3MjE=",
"avatar_url": "https://avatars.githubusercontent.com/u/2479721?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/xerus",
"html_url": "https://github.com/xerus",
"followers_url": "https://api.github.com/users/xerus/followers",
"following_url": "https://api.github.com/users/xerus/following{/other_user}",
"gists_url": "https://api.github.com/users/xerus/gists{/gist_id}",
"starred_url": "https://api.github.com/users/xerus/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/xerus/subscriptions",
"organizations_url": "https://api.github.com/users/xerus/orgs",
"repos_url": "https://api.github.com/users/xerus/repos",
"events_url": "https://api.github.com/users/xerus/events{/privacy}",
"received_events_url": "https://api.github.com/users/xerus/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [] | 2024-07-15T07:46:38 | 2024-07-15T20:03:52 | 2024-07-15T20:03:52 | CONTRIBUTOR | null | null | null | Hi, the commit 18038d652b950f919f53957b9ddda8d34fd7d04b add folder "videos" to all image paths in the singleMedia.txt file, however some images are in the `common/images/face/` folder
eg:
`../common/images/face/videos/small.ppm` should be `../common/images/face/small.ppm` | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/139/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/139/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/138 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/138/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/138/comments | https://api.github.com/repos/usnistgov/frvt/issues/138/events | https://github.com/usnistgov/frvt/issues/138 | 2,386,294,354 | I_kwDOBIKFiM6OO_ZS | 138 | Unexpected path of `video1` for `run_validate_ae.sh` | {
"login": "paplhjak",
"id": 52961799,
"node_id": "MDQ6VXNlcjUyOTYxNzk5",
"avatar_url": "https://avatars.githubusercontent.com/u/52961799?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/paplhjak",
"html_url": "https://github.com/paplhjak",
"followers_url": "https://api.github.com/users/paplhjak/followers",
"following_url": "https://api.github.com/users/paplhjak/following{/other_user}",
"gists_url": "https://api.github.com/users/paplhjak/gists{/gist_id}",
"starred_url": "https://api.github.com/users/paplhjak/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/paplhjak/subscriptions",
"organizations_url": "https://api.github.com/users/paplhjak/orgs",
"repos_url": "https://api.github.com/users/paplhjak/repos",
"events_url": "https://api.github.com/users/paplhjak/events{/privacy}",
"received_events_url": "https://api.github.com/users/paplhjak/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"We've updated the file with the correct path. Thank you."
] | 2024-07-02T13:40:03 | 2024-07-02T18:27:45 | 2024-07-02T18:27:45 | NONE | null | null | null | When running `./run_validate_ae.sh`, the execution fails with:
![image](https://github.com/usnistgov/frvt/assets/52961799/ba30b30c-6a18-4efd-ad18-88b48591c287)
This can be fixed by moving the `video1` folder down by one directory, however, this should not be expected of the user. | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/138/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/138/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/137 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/137/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/137/comments | https://api.github.com/repos/usnistgov/frvt/issues/137/events | https://github.com/usnistgov/frvt/issues/137 | 2,381,506,997 | I_kwDOBIKFiM6N8um1 | 137 | Disparity of median enrollment time when running on single core. | {
"login": "rishidd11",
"id": 85778019,
"node_id": "MDQ6VXNlcjg1Nzc4MDE5",
"avatar_url": "https://avatars.githubusercontent.com/u/85778019?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/rishidd11",
"html_url": "https://github.com/rishidd11",
"followers_url": "https://api.github.com/users/rishidd11/followers",
"following_url": "https://api.github.com/users/rishidd11/following{/other_user}",
"gists_url": "https://api.github.com/users/rishidd11/gists{/gist_id}",
"starred_url": "https://api.github.com/users/rishidd11/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/rishidd11/subscriptions",
"organizations_url": "https://api.github.com/users/rishidd11/orgs",
"repos_url": "https://api.github.com/users/rishidd11/repos",
"events_url": "https://api.github.com/users/rishidd11/events{/privacy}",
"received_events_url": "https://api.github.com/users/rishidd11/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"During our timing test, we only run with a single process (and limit the number of CPUs to one). We run and time the template generation call over many iterations, then take the median of those timing values."
] | 2024-06-29T05:28:04 | 2024-07-01T11:22:00 | 2024-07-01T11:21:59 | NONE | null | null | null | The NIST document specifies that the median enrollment time must be under 1.5 seconds on a single core, tested on an Intel(R) Xeon(R) Gold 6140 CPU @ 2.30GHz. In the run_testdriver.sh script, the num_forks argument is used to run validate11.cpp.
When I run the script on a single core using the command taskset --cpu-list 1 run_validate11.sh, the enrollment time for multiple processes (i.e., num_forks = 4) increases fourfold compared to a single process. However, there is no increase in time for multiple processes when I run the script without the taskset command.
How does NIST internally check the median enrollment time? Do they use the taskset command to limit the number of CPUs to one? | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/137/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/137/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/136 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/136/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/136/comments | https://api.github.com/repos/usnistgov/frvt/issues/136/events | https://github.com/usnistgov/frvt/issues/136 | 2,352,111,098 | I_kwDOBIKFiM6MMl36 | 136 | Suggest to split and zip "common/images/face" to release files | {
"login": "ImNotMT",
"id": 13098079,
"node_id": "MDQ6VXNlcjEzMDk4MDc5",
"avatar_url": "https://avatars.githubusercontent.com/u/13098079?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ImNotMT",
"html_url": "https://github.com/ImNotMT",
"followers_url": "https://api.github.com/users/ImNotMT/followers",
"following_url": "https://api.github.com/users/ImNotMT/following{/other_user}",
"gists_url": "https://api.github.com/users/ImNotMT/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ImNotMT/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ImNotMT/subscriptions",
"organizations_url": "https://api.github.com/users/ImNotMT/orgs",
"repos_url": "https://api.github.com/users/ImNotMT/repos",
"events_url": "https://api.github.com/users/ImNotMT/events{/privacy}",
"received_events_url": "https://api.github.com/users/ImNotMT/received_events",
"type": "User",
"site_admin": false
} | [] | open | false | null | [] | null | [
"`git clone --depth 1` may lower the size of files",
"thank you for the approach. but for my trying, still have 7.7G. the images and video size is 5.7G, cannot avoid that huge .git folder. ",
"I think basically caused by 4.7G videos uploaded recently by this commit 7ca40ad12d548232e88aecb167934bac1e659f29."
] | 2024-06-13T21:39:27 | 2024-06-14T09:25:08 | null | NONE | null | null | null | now git clone the frvt repo would create 10G+ .git files (files in frvt repo not counting yet), while the scripts and source code files only a few KBs.
it is not a good idea to git clone the whole frvt repo while you are mixing the data files (no version control/history required) and source files in git. | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/136/reactions",
"total_count": 3,
"+1": 3,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/136/timeline | null | null |
https://api.github.com/repos/usnistgov/frvt/issues/135 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/135/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/135/comments | https://api.github.com/repos/usnistgov/frvt/issues/135/events | https://github.com/usnistgov/frvt/issues/135 | 2,335,215,176 | I_kwDOBIKFiM6LMI5I | 135 | Stuck at warming up FE Model | {
"login": "SSicaaJL",
"id": 43140909,
"node_id": "MDQ6VXNlcjQzMTQwOTA5",
"avatar_url": "https://avatars.githubusercontent.com/u/43140909?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/SSicaaJL",
"html_url": "https://github.com/SSicaaJL",
"followers_url": "https://api.github.com/users/SSicaaJL/followers",
"following_url": "https://api.github.com/users/SSicaaJL/following{/other_user}",
"gists_url": "https://api.github.com/users/SSicaaJL/gists{/gist_id}",
"starred_url": "https://api.github.com/users/SSicaaJL/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/SSicaaJL/subscriptions",
"organizations_url": "https://api.github.com/users/SSicaaJL/orgs",
"repos_url": "https://api.github.com/users/SSicaaJL/repos",
"events_url": "https://api.github.com/users/SSicaaJL/events{/privacy}",
"received_events_url": "https://api.github.com/users/SSicaaJL/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [] | 2024-06-05T08:16:06 | 2024-06-14T02:59:48 | 2024-06-14T02:59:33 | NONE | null | null | null | null | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/135/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/135/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/134 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/134/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/134/comments | https://api.github.com/repos/usnistgov/frvt/issues/134/events | https://github.com/usnistgov/frvt/issues/134 | 2,314,622,451 | I_kwDOBIKFiM6J9lXz | 134 | Doubt regarding NIST Public Key | {
"login": "ElisonSherton",
"id": 45599762,
"node_id": "MDQ6VXNlcjQ1NTk5NzYy",
"avatar_url": "https://avatars.githubusercontent.com/u/45599762?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ElisonSherton",
"html_url": "https://github.com/ElisonSherton",
"followers_url": "https://api.github.com/users/ElisonSherton/followers",
"following_url": "https://api.github.com/users/ElisonSherton/following{/other_user}",
"gists_url": "https://api.github.com/users/ElisonSherton/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ElisonSherton/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ElisonSherton/subscriptions",
"organizations_url": "https://api.github.com/users/ElisonSherton/orgs",
"repos_url": "https://api.github.com/users/ElisonSherton/repos",
"events_url": "https://api.github.com/users/ElisonSherton/events{/privacy}",
"received_events_url": "https://api.github.com/users/ElisonSherton/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"If you are saying the string 856B9B28 does not match the string 5A18F9... then that is expected\r\n\r\nIf you are saying that something else doesn't match. then email us again.\r\n________________________________\r\nFrom: Vinayak Nayak ***@***.***>\r\nSent: Friday, May 24, 2024 3:23 AM\r\nTo: usnistgov/frvt ***@***.***>\r\nCc: Subscribed ***@***.***>\r\nSubject: [usnistgov/frvt] Doubt regarding NIST Public Key (Issue #134)\r\n\r\n\r\nHi!\r\n\r\nI am trying to encrypt my final submission file using the outlines mentioned here<https://www.nist.gov/system/files/documents/2021/10/25/nist_encryption.pdf>.\r\n\r\nAs mentioned in section 2.3, I try to import the NIST public key for FRTE provided here<https://www.nist.gov/system/files/documents/2021/10/25/nist_encryption.pdf> in the second row.\r\n\r\nHowever the key that I get is slightly different. The fingerprint obtained after importing the key is the same as the one mentioned in submission document but the key is different as can be seen below.\r\n\r\nkey_mismatch.png (view on web)<https://github.com/usnistgov/frvt/assets/45599762/48e32547-5bc7-4be4-8e41-0e0f980431d4>\r\n\r\nIs this acceptable for the submission or am I missing something when importing the key?\r\n\r\nThanks,\r\nVinayak.\r\n\r\n—\r\nReply to this email directly, view it on GitHub<https://github.com/usnistgov/frvt/issues/134>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ADVZFNO35GUQN6DQLHTUSJ3ZD3TGVAVCNFSM6AAAAABIHA4MCSVHI2DSMVQWIX3LMV43ASLTON2WKOZSGMYTINRSGI2DKMI>.\r\nYou are receiving this because you are subscribed to this thread.Message ID: ***@***.***>\r\n",
"Yeah that's what I meant. Thanks for the confirmation!"
] | 2024-05-24T07:22:58 | 2024-05-27T11:10:33 | 2024-05-27T11:10:33 | CONTRIBUTOR | null | null | null | Hi!
I am trying to encrypt my final submission file using the outlines [mentioned here](https://www.nist.gov/system/files/documents/2021/10/25/nist_encryption.pdf).
As mentioned in section 2.3, I try to import the NIST public key for FRTE [provided here](https://www.nist.gov/system/files/documents/2021/10/25/nist_encryption.pdf) in the second row.
However the key that I get is slightly different. The fingerprint obtained after importing the key is the same as the one mentioned in submission document but the key is different as can be seen below.
![key_mismatch](https://github.com/usnistgov/frvt/assets/45599762/48e32547-5bc7-4be4-8e41-0e0f980431d4)
Is this acceptable for the submission or am I missing something when importing the key?
Thanks,
Vinayak.
| {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/134/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/134/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/133 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/133/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/133/comments | https://api.github.com/repos/usnistgov/frvt/issues/133/events | https://github.com/usnistgov/frvt/issues/133 | 2,293,159,893 | I_kwDOBIKFiM6IrtfV | 133 | Failure-to-enroll/-to-extract | {
"login": "FlatErik90",
"id": 19189799,
"node_id": "MDQ6VXNlcjE5MTg5Nzk5",
"avatar_url": "https://avatars.githubusercontent.com/u/19189799?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/FlatErik90",
"html_url": "https://github.com/FlatErik90",
"followers_url": "https://api.github.com/users/FlatErik90/followers",
"following_url": "https://api.github.com/users/FlatErik90/following{/other_user}",
"gists_url": "https://api.github.com/users/FlatErik90/gists{/gist_id}",
"starred_url": "https://api.github.com/users/FlatErik90/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/FlatErik90/subscriptions",
"organizations_url": "https://api.github.com/users/FlatErik90/orgs",
"repos_url": "https://api.github.com/users/FlatErik90/repos",
"events_url": "https://api.github.com/users/FlatErik90/events{/privacy}",
"received_events_url": "https://api.github.com/users/FlatErik90/received_events",
"type": "User",
"site_admin": false
} | [] | open | false | null | [] | null | [
"Page 40 in which report (?). In our evaluations a failure to convert an image to a template will elevate 1:1 FNMR or 1:N FNIR if the template is used in a mated comparison. It will not cause increase 1:1 FMR or 1:N FPIR. The effect is the same as if the template had been produced but it always produces a low similarity score.\r\n\r\n\r\n\r\n\r\n\r\n\r\n\r\nFrom: Erik ***@***.***>\r\nDate: Monday, May 13, 2024 at 11:34 AM\r\nTo: usnistgov/frvt ***@***.***>\r\nCc: Subscribed ***@***.***>\r\nSubject: [usnistgov/frvt] Failure-to-enroll/-to-extract (Issue #133)\r\n\r\nHi,\r\n\r\nI have a question regarding the evaluation and reporting of failure-to-enroll and failure-to-extract. Your report on page 40 mentions that failure-to-extract (the inability to produce a template for search) is not reported. However, the table header in the results section reads \"Failure to extract features.\" Could you clarify if this actually refers to failure-to-enroll, which is the inability to produce a template for enrollment? Additionally, are the same evaluation metrics used for both Face and Iris in IREX10, including the failures mentioned above?\r\n\r\n—\r\nReply to this email directly, view it on GitHub<https://github.com/usnistgov/frvt/issues/133>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ADVZFNITMKXX2P4N3YTSIY3ZCDMRZAVCNFSM6AAAAABHUOQJ5GVHI2DSMVQWIX3LMV43ASLTON2WKOZSGI4TGMJVHE4DSMY>.\r\nYou are receiving this because you are subscribed to this thread.Message ID: ***@***.***>\r\n",
"I came across this issue before. The wording is a bit misleading, and it suggests it isn't counted.\nThink of it logically: Of course it is counted in the results. Otherwise, it is in your interest to fail to extract any marginally difficult images.\n\nIt is counted and will be considered the worst score, so it is in your interest to actually extract it usually. You get given enough time to try various methods of extraction.",
"@paddygr : I'm referring to the latest FRTE 1:N report, specifically section 3.5, which states, \"We do not report FTX...\". Given this, I assume the numbers reported in Table 15 under the column \"failure to extract features\" are only FTE – the proportion of templates that could not be generated during enrollment. My follow-up question concerns the IREX10 submissions: Does the FTE Rate reported on the table on the website also refer to the proportion of template generation during enrollment?"
] | 2024-05-13T15:34:28 | 2024-05-14T07:51:09 | null | NONE | null | null | null | Hi,
I have a question regarding the evaluation and reporting of failure-to-enroll and failure-to-extract. Your report on page 40 mentions that failure-to-extract (the inability to produce a template for search) is not reported. However, the table header in the results section reads "Failure to extract features." Could you clarify if this actually refers to failure-to-enroll, which is the inability to produce a template for enrollment? Additionally, are the same evaluation metrics used for both Face and Iris in IREX10, including the failures mentioned above? | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/133/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/133/timeline | null | null |
https://api.github.com/repos/usnistgov/frvt/issues/132 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/132/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/132/comments | https://api.github.com/repos/usnistgov/frvt/issues/132/events | https://github.com/usnistgov/frvt/issues/132 | 2,248,143,235 | I_kwDOBIKFiM6F__GD | 132 | Matching templates ends with signal 11 | {
"login": "angelabr",
"id": 17623199,
"node_id": "MDQ6VXNlcjE3NjIzMTk5",
"avatar_url": "https://avatars.githubusercontent.com/u/17623199?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/angelabr",
"html_url": "https://github.com/angelabr",
"followers_url": "https://api.github.com/users/angelabr/followers",
"following_url": "https://api.github.com/users/angelabr/following{/other_user}",
"gists_url": "https://api.github.com/users/angelabr/gists{/gist_id}",
"starred_url": "https://api.github.com/users/angelabr/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/angelabr/subscriptions",
"organizations_url": "https://api.github.com/users/angelabr/orgs",
"repos_url": "https://api.github.com/users/angelabr/repos",
"events_url": "https://api.github.com/users/angelabr/events{/privacy}",
"received_events_url": "https://api.github.com/users/angelabr/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [] | 2024-04-17T12:12:33 | 2024-04-18T07:59:58 | 2024-04-18T07:59:58 | NONE | null | null | null | null | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/132/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/132/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/130 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/130/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/130/comments | https://api.github.com/repos/usnistgov/frvt/issues/130/events | https://github.com/usnistgov/frvt/issues/130 | 2,173,814,957 | I_kwDOBIKFiM6Bkcit | 130 | Linking external so libraries using CMAKE_INSTALL_RPATH. | {
"login": "rishidd11",
"id": 85778019,
"node_id": "MDQ6VXNlcjg1Nzc4MDE5",
"avatar_url": "https://avatars.githubusercontent.com/u/85778019?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/rishidd11",
"html_url": "https://github.com/rishidd11",
"followers_url": "https://api.github.com/users/rishidd11/followers",
"following_url": "https://api.github.com/users/rishidd11/following{/other_user}",
"gists_url": "https://api.github.com/users/rishidd11/gists{/gist_id}",
"starred_url": "https://api.github.com/users/rishidd11/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/rishidd11/subscriptions",
"organizations_url": "https://api.github.com/users/rishidd11/orgs",
"repos_url": "https://api.github.com/users/rishidd11/repos",
"events_url": "https://api.github.com/users/rishidd11/events{/privacy}",
"received_events_url": "https://api.github.com/users/rishidd11/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"This thread may be helpful to you - https://github.com/usnistgov/frvt/issues/128",
"I followed the thread but still unable to link the libfrvt_1N_null_000.so file with the dependencies as relative path. Following is the folder structure and the cmake file:\r\n[CMakeLists.txt](https://github.com/usnistgov/frvt/files/14534503/CMakeLists.txt)\r\n![nist1](https://github.com/usnistgov/frvt/assets/85778019/bae289f4-0fcb-44d8-84da-b74a8087efcb)\r\n\r\nThe onnx and opencv \".so\" files are in onnxruntime/build/Linux/Release/ and opencv/build/lib/ folders\r\nWhat changes should I make in CMakeLists.txt in order to add relative path of onnx and opencv so files?\r\nIn the thread you mentioned, he moves all the opencv .so files in lib folder and sets the as the LDD_Library_path is set to frvt/1N/lib. I also did the same but couldn't achieve relative linking.\r\n\r\n",
"I use onnxruntime and opencv. This is my `CMakeLists.txt`. You can refer to it.\r\n```\r\nproject(nullImpl)\r\ncmake_minimum_required(VERSION 2.8)\r\n\r\n# Find OpenCV package\r\n# set(CMAKE_MODULE_PATH ${CMAKE_MODULE_PATH} ${CMAKE_CURRENT_SOURCE_DIR}/../../lib/opencv-4.9.0/cmake)\r\nset(OpenCV_DIR ${CMAKE_CURRENT_SOURCE_DIR}/../../lib/opencv-4.9.0/lib/cmake/opencv4)\r\nfind_package(OpenCV REQUIRED)\r\n\r\nset (CMAKE_CXX_FLAGS \"${CMAKE_CXX_FLAGS} -std=c++17\")\r\n\r\n# Link onnxruntime\r\nset(ONNX_LIB_DIR ${CMAKE_CURRENT_SOURCE_DIR}/../../lib/onnxruntime-linux-x64-1.17.1/lib)\r\n\r\nset(SOURCES nullimplfrvt11.cpp fd.cpp)\r\nadd_executable(nullimplfrvt11 ${SOURCES})\r\ntarget_link_libraries(nullimplfrvt11 ${ONNX_LIB_DIR}/libonnxruntime.so \r\n \t\t\t ${ONNX_LIB_DIR}/libonnxruntime.so.1.17.1)\r\ntarget_link_libraries(nullimplfrvt11 ${OpenCV_LIBS})\r\n\r\n# Include directories\r\ninclude_directories(${OpenCV_INCLUDE_DIRS})\r\ninclude_directories(${CMAKE_CURRENT_SOURCE_DIR}/../include \r\n ${CMAKE_CURRENT_SOURCE_DIR}/../../../common/src/include\r\n ${CMAKE_CURRENT_SOURCE_DIR}/../../lib/onnxruntime-linux-x64-1.17.1/include\r\n ${CMAKE_CURRENT_SOURCE_DIR}/../../lib/opencv-4.9.0/include)\r\n\r\n# Configure built shared libraries in top-level lib directory\r\nset (CMAKE_LIBRARY_OUTPUT_DIRECTORY ${CMAKE_CURRENT_SOURCE_DIR}/../../lib)\r\n\r\n# Build the shared libraries\r\nadd_library (frvt_11_VVTK_001 SHARED nullimplfrvt11.cpp fd.cpp)\r\n\r\n# Link Torch to the shared library\r\ntarget_link_libraries(frvt_11_VVTK_001 ${ONNX_LIB_DIR}/libonnxruntime.so \r\n \t\t\t ${ONNX_LIB_DIR}/libonnxruntime.so.1.17.1)\r\ntarget_link_libraries(frvt_11_VVTK_001 ${OpenCV_LIBS})\r\n```"
] | 2024-03-07T13:02:02 | 2024-03-15T06:27:33 | 2024-03-07T14:41:55 | NONE | null | null | null | I want to link the "libfrvt_1N_null_000.so" (present in "./lib" folder) with the opencv and onnxruntime libraries which are present in "./lib/opencv" and "./lib/onnxruntime" folder respectively. How can I link the main .so file with the opencv and onnx dependencies such that relative path is added using CMake. I tried using CMAKE_INSTALL_RPATH but I am still unable to link it as relative path.
Thankyou | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/130/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/130/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/129 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/129/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/129/comments | https://api.github.com/repos/usnistgov/frvt/issues/129/events | https://github.com/usnistgov/frvt/issues/129 | 2,173,760,716 | I_kwDOBIKFiM6BkPTM | 129 | FRTE Latest Report Query | {
"login": "ElisonSherton",
"id": 45599762,
"node_id": "MDQ6VXNlcjQ1NTk5NzYy",
"avatar_url": "https://avatars.githubusercontent.com/u/45599762?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ElisonSherton",
"html_url": "https://github.com/ElisonSherton",
"followers_url": "https://api.github.com/users/ElisonSherton/followers",
"following_url": "https://api.github.com/users/ElisonSherton/following{/other_user}",
"gists_url": "https://api.github.com/users/ElisonSherton/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ElisonSherton/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ElisonSherton/subscriptions",
"organizations_url": "https://api.github.com/users/ElisonSherton/orgs",
"repos_url": "https://api.github.com/users/ElisonSherton/repos",
"events_url": "https://api.github.com/users/ElisonSherton/events{/privacy}",
"received_events_url": "https://api.github.com/users/ElisonSherton/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"Hi @mlngan \r\n\r\nIf you can provide some clarification here, it would be really great!\r\n\r\nThanks,\r\nVinayak.",
"K > 1 is not exercised currently. It will be in the future.\r\n\r\n\r\nFrom: Vinayak Nayak ***@***.***>\r\nDate: Thursday, March 7, 2024 at 7:34 AM\r\nTo: usnistgov/frvt ***@***.***>\r\nCc: Subscribed ***@***.***>\r\nSubject: [usnistgov/frvt] FRTE Latest Report Query (Issue #129)\r\n\r\nIn the latest report here<https://pages.nist.gov/frvt/reports/11/frvt_11_report.pdf>, under section 3.2, the test design principle for different types of comparison is provided.\r\n\r\nIn all the tracks (namely visa, mugshot, visa-border, visa-border non-frontal yaw, kiosk-border, border-border, wild, child-exploitation), the following two statements are present:\r\n\r\nThe number of images used to make a template is one.\r\nThe number of templates used to make each comparison score is two corresponding to simple one-to-one verification\r\n\r\nIn the API document<https://pages.nist.gov/frvt/api/FRVT_ongoing_11_api.pdf> on line 82, we can see the following table\r\n\r\nimage.png (view on web)<https://github.com/usnistgov/frvt/assets/45599762/cac32890-5639-4b75-92a4-f15485c69542>\r\n\r\nMy question is then when will multiple images (i.e. K > 1) be sent to createFaceTemplate API for making an enrolment or verification template?\r\n\r\nThanks,\r\nVinayak.\r\n\r\n—\r\nReply to this email directly, view it on GitHub<https://github.com/usnistgov/frvt/issues/129>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ADVZFNOEFLOKP6YEM3DK3MLYXBNBFAVCNFSM6AAAAABEK5CTMKVHI2DSMVQWIX3LMV43ASLTON2WKOZSGE3TGNZWGA3TCNQ>.\r\nYou are receiving this because you are subscribed to this thread.Message ID: ***@***.***>\r\n"
] | 2024-03-07T12:32:59 | 2024-03-14T09:57:10 | 2024-03-14T09:57:10 | CONTRIBUTOR | null | null | null | In the [latest report here](https://pages.nist.gov/frvt/reports/11/frvt_11_report.pdf), under section 3.2, the test design principle for different types of comparison is provided.
In all the tracks (namely visa, mugshot, visa-border, visa-border non-frontal yaw, kiosk-border, border-border, wild, child-exploitation), the following two statements are present:
> The number of images used to make a template is one.
> The number of templates used to make each comparison score is two corresponding to simple one-to-one verification
In the [API document](https://pages.nist.gov/frvt/api/FRVT_ongoing_11_api.pdf) on line 82, we can see the following table
![image](https://github.com/usnistgov/frvt/assets/45599762/cac32890-5639-4b75-92a4-f15485c69542)
My question is then when will multiple images (i.e. K > 1) be sent to `createFaceTemplate` API for making an enrolment or verification template?
Thanks,
Vinayak. | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/129/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/129/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/128 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/128/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/128/comments | https://api.github.com/repos/usnistgov/frvt/issues/128/events | https://github.com/usnistgov/frvt/issues/128 | 2,166,091,595 | I_kwDOBIKFiM6BG-9L | 128 | Can we build a library in the target environment? | {
"login": "ElisonSherton",
"id": 45599762,
"node_id": "MDQ6VXNlcjQ1NTk5NzYy",
"avatar_url": "https://avatars.githubusercontent.com/u/45599762?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ElisonSherton",
"html_url": "https://github.com/ElisonSherton",
"followers_url": "https://api.github.com/users/ElisonSherton/followers",
"following_url": "https://api.github.com/users/ElisonSherton/following{/other_user}",
"gists_url": "https://api.github.com/users/ElisonSherton/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ElisonSherton/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ElisonSherton/subscriptions",
"organizations_url": "https://api.github.com/users/ElisonSherton/orgs",
"repos_url": "https://api.github.com/users/ElisonSherton/repos",
"events_url": "https://api.github.com/users/ElisonSherton/events{/privacy}",
"received_events_url": "https://api.github.com/users/ElisonSherton/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"Can you use a relative RPATH instead? You can also change your RPATH to '.' or whatever is relevant with `chrpath`",
"Hi @xsacha \r\n\r\nThanks for the prompt response.\r\n\r\nSo, here is how my lib is structured\r\n\r\n![image](https://github.com/usnistgov/frvt/assets/45599762/09e77854-a7bb-4f90-90a1-a320ecd3ca65)\r\n\r\nI have tried the following options in my CMakeLists.txt file in the src/nullImpl folder\r\n\r\n```\r\n# Option 1\r\nset (CMAKE_SHARED_LINKER_FLAGS \"-Wl,-rpath-link,$ORIGIN/opencv_42/lib\")\r\n\r\n# Option 2\r\nset (CMAKE_SHARED_LINKER_FLAGS \"-Wl,-rpath,$ORIGIN/opencv_42/lib\")\r\n\r\n# Option 3\r\nset(ocv_name opencv_42)\r\nset(CMAKE_SHARED_LINKER_FLAGS \"-Wl,-rpath-link,${CMAKE_CURRENT_SOURCE_DIR}/../../lib/${ocv_name}/lib/\")\r\n\r\n#Option 4\r\nset(ocv_name opencv_42)\r\nset(CMAKE_SHARED_LINKER_FLAGS \"-Wl,-rpath,${CMAKE_CURRENT_SOURCE_DIR}/../../lib/${ocv_name}/lib/\")\r\n```\r\n\r\nWhen I do chrpath, I can see that the rpath is set appropriately as expected\r\n\r\n![image](https://github.com/usnistgov/frvt/assets/45599762/5c9a18cb-d951-48ca-af9a-cb257554adb6)\r\n\r\nAlso the libraries for libopencv_dnn, libopencv_imgproc etc. are present in the proper locations as seen below\r\n\r\n![image](https://github.com/usnistgov/frvt/assets/45599762/d2aa0647-28a4-46d8-a0d8-79073e457690)\r\n\r\nHowever when executing the `run_validate_11.sh` file, these libraries are not fetched...\r\n\r\nCould you suggest how can I debug this issue further? I have tried several things but nothing has succeeded so far...\r\n\r\n",
"I would say your chrpath is showing that the RPATH is misconfigured. For example if you move the library to another path, does it still work? It looks like a full path in the output.\nYou can also change the rpath to a relative path.\n\nTry this: move the libraries and bin to a different path as the validation would do and then check if it works. If it works, no problem.",
"Thanks for the comment, indeed it was misconfigured.\r\n\r\nNow I think it is configured correctly as I can see that it is showing origin correctly in the rpath of the exeutable.\r\n\r\n![image](https://github.com/usnistgov/frvt/assets/45599762/e86beede-b5e2-405c-92bb-ed4af184bd25)\r\n\r\nHowever, still some opencv libraries are not being picked up despite their presence in the libs folder as shown in the following folder structure\r\n\r\n![image](https://github.com/usnistgov/frvt/assets/45599762/06247b35-2ec9-4c6c-8ed4-72829f1f2fb9)\r\n\r\n![image](https://github.com/usnistgov/frvt/assets/45599762/ec9f6676-07da-47fa-916d-941834205343)",
"Hi @xsacha ,\r\n\r\nIf I do an ldd to check if all the libraries are linked properly against the submission library, I get the following output\r\n\r\n![image](https://github.com/usnistgov/frvt/assets/45599762/a4d3eafe-1772-4dcf-9bc1-24fa88a89f25)\r\n\r\nIt is linking with all the libraries properly except `libopencv_imgproc.so`, any ideas why would that be? The library exists in the same path as `libopencv_dnn.so` and `libopencv_core.so` which are there in the same directory as `libopencv_imgproc.so`.\r\n\r\nAny idea on what might be going wrong? This same thing works fine when I test on my local machine but on the VM built using ISO provided by NIST, I encounter this issue...\r\n\r\nThanks,\r\nVinayak.",
"Solved this issue. After inspecting which files are not linked to my submission library I found that the opencv libraries were not linked and this was happening because the LD_LIBRARY_PATH was set only to frvt/11/lib.\r\n\r\nKeeping all the opencv libraries in lib folder in a flat format helped solve this issue.",
"@ElisonSherton Can you give some hint on how you changed the CMakeLists.txt since I am facing the same issue as well. My folder structure inside lib folder is as follows:\r\nlib:\r\n libfrvt_nullimpl.so\r\n opencv_folder\r\n onnxruntime_folder\r\nopencv and onnxruntime \".so\" files are inside the opencv_folder/build/lib and onnxruntime_folder/build/Linux/Release/ folders respectively.\r\n\r\nAlso did you set the ldd_library_path to frvt/11/lib and moved all the opencv libraries to that folder? If so then how did you edit the cmakelists.txt?",
"Hey @rishidd11,\r\n\r\nIf you're facing the same issue as me, then this may help you.\r\n\r\nLD_LIBRARY_PATH is already set by NIST to point to the `frvt/11/lib` folder [as seen here](https://github.com/usnistgov/frvt/blob/b6b23da12b8d814d96a63e69179133e6972c20e2/11/run_validate_11.sh#L23), so I did not manually change it or make any changes to CMakeLists.txt for linking purposes. I do assume that libs are stored in the `frvt/11/lib` location and accordingly set my paths.\r\n\r\nInstead from the opencv_folder/build/lib (as in your case), I took all of the .so files and kept them inside the `frvt/11/lib` folders. That made my code work as expected.\r\n\r\nHope this helps.\r\n\r\nThanks,\r\nVinayak."
] | 2024-03-04T07:05:51 | 2024-03-09T09:23:24 | 2024-03-07T09:15:42 | CONTRIBUTOR | null | null | null | Hi there.
I have observed that when I link a library (opencvlibs) which is built on my source system, the submission package passes all the checks and gives meaningful results.
However when I try to run the `run_validate11.sh` file on a VM created using iso image [provided here on line 89](https://pages.nist.gov/frvt/api/FRVT_common.pdf), the linker gives a problem
as follows
```
/usr/bin/ld: warning: libopencv_dnn.so.408, needed by ../../../lib/libfrvt_11_company_000.so, not found (try using -rpath or -rpath-link)
```
Even after providing -rpath as follows before linking the target with the opencv libraries, the same issue prevails
```
# Create a variable for opencv package name
set(ocv_name opencv_48)
set(CMAKE_SHARED_LINKER_FLAGS "-Wl,-rpath,${CMAKE_CURRENT_SOURCE_DIR}/../../lib/${ocv_name}/lib/")
```
Can someone guide me how to overcome this issue? Can we build opencv in the target environment on the fly while initializing the environment? This would take a lot of time since compiling and building opencv is a really difficult task. However I was wondering if there is a simpler solution that can help me in this situation.
Would be grateful for any help, thanks a lot! | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/128/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/128/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/127 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/127/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/127/comments | https://api.github.com/repos/usnistgov/frvt/issues/127/events | https://github.com/usnistgov/frvt/issues/127 | 2,162,918,560 | I_kwDOBIKFiM6A64Sg | 127 | free(): invalid pointer after trying to link torch library | {
"login": "deeperlearner",
"id": 30494216,
"node_id": "MDQ6VXNlcjMwNDk0MjE2",
"avatar_url": "https://avatars.githubusercontent.com/u/30494216?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/deeperlearner",
"html_url": "https://github.com/deeperlearner",
"followers_url": "https://api.github.com/users/deeperlearner/followers",
"following_url": "https://api.github.com/users/deeperlearner/following{/other_user}",
"gists_url": "https://api.github.com/users/deeperlearner/gists{/gist_id}",
"starred_url": "https://api.github.com/users/deeperlearner/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/deeperlearner/subscriptions",
"organizations_url": "https://api.github.com/users/deeperlearner/orgs",
"repos_url": "https://api.github.com/users/deeperlearner/repos",
"events_url": "https://api.github.com/users/deeperlearner/events{/privacy}",
"received_events_url": "https://api.github.com/users/deeperlearner/received_events",
"type": "User",
"site_admin": false
} | [] | open | false | null | [] | null | [] | 2024-03-01T08:54:15 | 2024-03-01T08:57:13 | null | NONE | null | null | null | I encounter the same issue as #49 . I simply add the following line in `frvt/11/src/nullImpl/CMakeLists.txt`:
```
target_link_libraries(frvt_11_null_001 ${TORCH_LIBRARIES})
```
Then, running `./scripts/build_null_impl.sh` and `./run_validate_11.sh` gives
```
Checking for hard-coded config directory free(): invalid pointer
PID 33486 exited due to signal 6
free(): invalid pointer
```
`./run_validate_11.sh` run successfully without that line.
How can I link libtorch correctly in the `CMakeLists.txt`? Thanks for your help! | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/127/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/127/timeline | null | null |
https://api.github.com/repos/usnistgov/frvt/issues/126 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/126/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/126/comments | https://api.github.com/repos/usnistgov/frvt/issues/126/events | https://github.com/usnistgov/frvt/issues/126 | 2,158,438,999 | I_kwDOBIKFiM6ApypX | 126 | ONNX Runtime Multithreading Issue Preventing NIST FRVT 11 Submission | {
"login": "AwaisPF",
"id": 97581451,
"node_id": "U_kgDOBdD5iw",
"avatar_url": "https://avatars.githubusercontent.com/u/97581451?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/AwaisPF",
"html_url": "https://github.com/AwaisPF",
"followers_url": "https://api.github.com/users/AwaisPF/followers",
"following_url": "https://api.github.com/users/AwaisPF/following{/other_user}",
"gists_url": "https://api.github.com/users/AwaisPF/gists{/gist_id}",
"starred_url": "https://api.github.com/users/AwaisPF/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/AwaisPF/subscriptions",
"organizations_url": "https://api.github.com/users/AwaisPF/orgs",
"repos_url": "https://api.github.com/users/AwaisPF/repos",
"events_url": "https://api.github.com/users/AwaisPF/events{/privacy}",
"received_events_url": "https://api.github.com/users/AwaisPF/received_events",
"type": "User",
"site_admin": false
} | [] | open | false | null | [] | null | [
"Did you close the docker and re-open one?",
"there is no usage of docker here. It ran directly on host machine as recommended."
] | 2024-02-28T08:55:35 | 2024-03-25T08:42:22 | null | NONE | null | null | null | ## Reporting a Multithreading Issue in ONNX Runtime to NIST FRVT GitHub Repository
**Body:**
I am encountering an issue with ONNX Runtime that prevents my submission to the NIST FRVT 11 evaluation.
**Describe the problem:**
* I am using ONNX Runtime for inference in my submission package.
* I have attempted to restrict the number of threads used by ONNX Runtime to 1 using the provided GUID.
* However, when my submission package is executed, the NIST FRVT script detects 17 threads instead of the expected 2.
* This discrepancy becomes apparent when using the `top -H` command, which reveals 17 threads are created, with only 1 in a running state and the remaining 16 in a sleeping state.
* Due to this multithreading behavior, my package crashes when executed by NIST FRVT on multiple cores, violating their single-threaded requirement.
**Here is screen shot of htop of same package**
<img width="937" alt="Screenshot 2024-02-28 at 7 03 14 PM" src="https://github.com/usnistgov/frvt/assets/97581451/945394ba-3486-4a44-9b44-924692796293">
* Only one thread is running but why there are so many threads launched
**Possible impact:**
* This issue prevents my package from being successfully evaluated in the NIST FRVT 11 program.
* I kindly request the NIST FRVT developers to investigate this multithreading issue in ONNX Runtime and provide any potential solutions or workarounds.
* Additionally, I would appreciate guidance on overcoming this limitation and ensuring successful submission compliance with the single-threaded requirement.
**Additional information:**
* **Version of ONNX Runtime==1.10.0**
**Closing:**
Thank you for your time and consideration. I look forward to your response and assistance in resolving this issue. | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/126/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/126/timeline | null | null |
https://api.github.com/repos/usnistgov/frvt/issues/125 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/125/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/125/comments | https://api.github.com/repos/usnistgov/frvt/issues/125/events | https://github.com/usnistgov/frvt/issues/125 | 2,155,982,443 | I_kwDOBIKFiM6Aga5r | 125 | Error in compiling with libtorch | {
"login": "deeperlearner",
"id": 30494216,
"node_id": "MDQ6VXNlcjMwNDk0MjE2",
"avatar_url": "https://avatars.githubusercontent.com/u/30494216?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/deeperlearner",
"html_url": "https://github.com/deeperlearner",
"followers_url": "https://api.github.com/users/deeperlearner/followers",
"following_url": "https://api.github.com/users/deeperlearner/following{/other_user}",
"gists_url": "https://api.github.com/users/deeperlearner/gists{/gist_id}",
"starred_url": "https://api.github.com/users/deeperlearner/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/deeperlearner/subscriptions",
"organizations_url": "https://api.github.com/users/deeperlearner/orgs",
"repos_url": "https://api.github.com/users/deeperlearner/repos",
"events_url": "https://api.github.com/users/deeperlearner/events{/privacy}",
"received_events_url": "https://api.github.com/users/deeperlearner/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"revise the content of `./src/nullImpl/CMakeLists.txt` solved the issue.\r\n```\r\nproject(nullImpl)\r\ncmake_minimum_required(VERSION 2.8)\r\n\r\nfind_package(Torch REQUIRED)\r\nset(CMAKE_CXX_FLAGS \"${CMAKE_CXX_FLAGS} ${TORCH_CXX_FLAGS}\")\r\n\r\nadd_executable(nullImpl nullimplfrvt11.cpp)\r\ntarget_link_libraries(nullImpl ${TORCH_LIBRARIES})\r\nset_property(TARGET nullImpl PROPERTY CXX_STANDARD 17)\r\n\r\nset (CMAKE_CXX_FLAGS \"${CMAKE_CXX_FLAGS} -std=c++17\")\r\ninclude_directories (${CMAKE_CURRENT_SOURCE_DIR}/../include ${CMAKE_CURRENT_SOURCE_DIR}/../../../common/src/include ${CMAKE_CURRENT_SOURCE_DIR}/../../lib/libtorch/include)\r\n\r\n# Configure built shared libraries in top-level lib directory\r\nset (CMAKE_LIBRARY_OUTPUT_DIRECTORY ${CMAKE_CURRENT_SOURCE_DIR}/../../lib)\r\n\r\n# Build the shared libraries\r\nadd_library (frvt_11_VVTK_001 SHARED nullimplfrvt11.cpp)\r\ntarget_link_libraries(frvt_11_VVTK_001 ${TORCH_LIBRARIES})\r\n```"
] | 2024-02-27T08:33:21 | 2024-03-01T02:08:37 | 2024-03-01T02:08:37 | NONE | null | null | null | I download libtorch by `wget https://download.pytorch.org/libtorch/nightly/cpu/libtorch-win-shared-with-deps-latest.zip`.
I add the following code inside `nullimplfrvt11.cpp`.
```C++
#include <torch/script.h>
ReturnStatus
NullImplFRVT11::initialize(const std::string &configDir)
{
torch::jit::script::Module module;
try {
std::string path = configDir + "script_model.pt";
module = torch::jit::load(path);
}
catch (const c10::Error& e) {
std::cerr << "error loading the model" << e.msg() << "\n";
return ReturnStatus(ReturnCode::ConfigError);
}
std::cout << "ok\n";
return ReturnStatus(ReturnCode::Success);
}
```
Running `./script/build_null_impl.sh` was fine and generated `libfrvt_11_<>_001.so`.
But when I run `./run_validate_11.sh`, it has the following error:
```
/usr/bin/ld: /train/frvt/11/lib/libfrvt_11_VVTK_001.so: undefined reference to `c10::ListType::get(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, c10::Type::SingletonOrSharedTypePtr<c10::Type>)'
/usr/bin/ld: /train/frvt/11/lib/libfrvt_11_VVTK_001.so: undefined reference to `torch::jit::load(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, c10::optional<c10::Device>, bool)'
/usr/bin/ld: /train/frvt/11/lib/libfrvt_11_VVTK_001.so: undefined reference to `c10::StringType::get()'
/usr/bin/ld: /train/frvt/11/lib/libfrvt_11_VVTK_001.so: undefined reference to `c10::AnyType::get()'
/usr/bin/ld: /train/frvt/11/lib/libfrvt_11_VVTK_001.so: undefined reference to `c10::UndefinedTensorImpl::_singleton'
```
I have followed [pytorch/pytorch#13541](https://github.com/pytorch/pytorch/issues/13541) and [pytorch/pytorch#17492](https://github.com/pytorch/pytorch/issues/17492). But still no luck. Any suggestions or something I've missed? Thanks for your help. | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/125/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/125/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/124 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/124/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/124/comments | https://api.github.com/repos/usnistgov/frvt/issues/124/events | https://github.com/usnistgov/frvt/issues/124 | 2,143,429,571 | I_kwDOBIKFiM5_wiPD | 124 | Submission of Dynamically linked library(.so file) to NIST. | {
"login": "rishidd1",
"id": 160201268,
"node_id": "U_kgDOCYx6NA",
"avatar_url": "https://avatars.githubusercontent.com/u/160201268?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/rishidd1",
"html_url": "https://github.com/rishidd1",
"followers_url": "https://api.github.com/users/rishidd1/followers",
"following_url": "https://api.github.com/users/rishidd1/following{/other_user}",
"gists_url": "https://api.github.com/users/rishidd1/gists{/gist_id}",
"starred_url": "https://api.github.com/users/rishidd1/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/rishidd1/subscriptions",
"organizations_url": "https://api.github.com/users/rishidd1/orgs",
"repos_url": "https://api.github.com/users/rishidd1/repos",
"events_url": "https://api.github.com/users/rishidd1/events{/privacy}",
"received_events_url": "https://api.github.com/users/rishidd1/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"Yes, you may include supplemental .so libraries in the same ./lib folder where your core library is located."
] | 2024-02-20T02:51:39 | 2024-02-20T13:43:16 | 2024-02-20T13:43:16 | NONE | null | null | null | Hi I have used onnxruntime and opencv libraries in my face recognition implementation. NIST asks vendors to submit a single .so file (libfrvt_<track>_<provider>_<sequence>.so) (core implementation) as a part of tar package. The only library that will be explicitly linked to test driver is the core library. Currently I am working in docker environment that has opencv and onnxruntime dependencies and generating the .so file. However, I am unable to run the testdriver script outside the docker environment as no dependencies are there.
Can I provide supplemental .so libraries (containing the dependencies) along with the core .so file? If yes then how? | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/124/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/124/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/123 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/123/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/123/comments | https://api.github.com/repos/usnistgov/frvt/issues/123/events | https://github.com/usnistgov/frvt/issues/123 | 2,138,299,151 | I_kwDOBIKFiM5_c9sP | 123 | Conversion of uint8_t to float for template | {
"login": "rishidd1",
"id": 160201268,
"node_id": "U_kgDOCYx6NA",
"avatar_url": "https://avatars.githubusercontent.com/u/160201268?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/rishidd1",
"html_url": "https://github.com/rishidd1",
"followers_url": "https://api.github.com/users/rishidd1/followers",
"following_url": "https://api.github.com/users/rishidd1/following{/other_user}",
"gists_url": "https://api.github.com/users/rishidd1/gists{/gist_id}",
"starred_url": "https://api.github.com/users/rishidd1/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/rishidd1/subscriptions",
"organizations_url": "https://api.github.com/users/rishidd1/orgs",
"repos_url": "https://api.github.com/users/rishidd1/repos",
"events_url": "https://api.github.com/users/rishidd1/events{/privacy}",
"received_events_url": "https://api.github.com/users/rishidd1/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"Many developers use floating point internally, but we use a generic byte stream to allow any type, including a constructed type. One solution is to use casting, for example.\r\n \r\n// Cast float to uint8_t during template creation\r\nconst uint8_t* bytes = reinterpret_cast<const uint8_t*>(fv.data());\r\n \r\n// Cast uint8_t back to float during match\r\nfloat *featureVector = (float *)enrollTemplate.data();\r\n \r\nCan you elaborate on what you mean by the values are incorrect when you use reinterpret_cast?",
"Thanks! Above solution worked for me!"
] | 2024-02-16T10:26:26 | 2024-02-21T13:07:41 | 2024-02-21T13:07:41 | NONE | null | null | null | Hi! I have a doubt regarding the conversion of unint8_t to float for the template vector. In the validate1N.cpp file templ(template) variable datatype is declared as vector<uint8_t>. The datatype of embeddings from the recognizer model is of data type float. How can I obtain a conversion of uint8 to float. Also, if I use reinterpret cast for conversion, the values I get are incorrect. Why havent NIST utilized data type of templ as float or double instead of uint8_t?
| {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/123/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/123/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/120 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/120/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/120/comments | https://api.github.com/repos/usnistgov/frvt/issues/120/events | https://github.com/usnistgov/frvt/issues/120 | 2,009,033,903 | I_kwDOBIKFiM53v2yv | 120 | CMake version | {
"login": "shritor",
"id": 149577702,
"node_id": "U_kgDOCOpf5g",
"avatar_url": "https://avatars.githubusercontent.com/u/149577702?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/shritor",
"html_url": "https://github.com/shritor",
"followers_url": "https://api.github.com/users/shritor/followers",
"following_url": "https://api.github.com/users/shritor/following{/other_user}",
"gists_url": "https://api.github.com/users/shritor/gists{/gist_id}",
"starred_url": "https://api.github.com/users/shritor/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/shritor/subscriptions",
"organizations_url": "https://api.github.com/users/shritor/orgs",
"repos_url": "https://api.github.com/users/shritor/repos",
"events_url": "https://api.github.com/users/shritor/events{/privacy}",
"received_events_url": "https://api.github.com/users/shritor/received_events",
"type": "User",
"site_admin": false
} | [] | open | false | null | [] | null | [] | 2023-11-24T03:19:28 | 2023-11-24T03:19:28 | null | NONE | null | null | null | my Lib c++ is used CMake ver 3.27. and when submit lib, your system support it ? | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/120/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/120/timeline | null | null |
https://api.github.com/repos/usnistgov/frvt/issues/119 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/119/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/119/comments | https://api.github.com/repos/usnistgov/frvt/issues/119/events | https://github.com/usnistgov/frvt/issues/119 | 2,006,157,128 | I_kwDOBIKFiM53k4dI | 119 | Error when runnung the run_validate_11.sh. [ERROR] Match validation failed | {
"login": "LionLion888",
"id": 133086055,
"node_id": "U_kgDOB-67Zw",
"avatar_url": "https://avatars.githubusercontent.com/u/133086055?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/LionLion888",
"html_url": "https://github.com/LionLion888",
"followers_url": "https://api.github.com/users/LionLion888/followers",
"following_url": "https://api.github.com/users/LionLion888/following{/other_user}",
"gists_url": "https://api.github.com/users/LionLion888/gists{/gist_id}",
"starred_url": "https://api.github.com/users/LionLion888/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/LionLion888/subscriptions",
"organizations_url": "https://api.github.com/users/LionLion888/orgs",
"repos_url": "https://api.github.com/users/LionLion888/repos",
"events_url": "https://api.github.com/users/LionLion888/events{/privacy}",
"received_events_url": "https://api.github.com/users/LionLion888/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"[Update] Problem solved!\r\n\r\nBecause I was using virtual machine to run the FRVT validationand the error is due to the small number of system processor and not enough of space of the memory capacity.\r\n\r\nIf increace the number of system processor and memory capacity, the problem can be solved and successfully go through all the validation process."
] | 2023-11-22T11:39:42 | 2023-11-24T05:56:04 | 2023-11-24T05:56:04 | NONE | null | null | null | Hi there,
I'm trying to pack up the submit code. I can run the null implementation and pack up the null testing code successfully.
But if I try to pack up my own submitted code, there will have some error like follows:
Could anyone figure out the problem?
```
[100%] Built target validate11
[SUCCESS] Built executable in /home/pj/frvt_2023_adaface_submit/11/bin.
------------------------------
Running 1:1 validation
------------------------------
Checking for hard-coded config directory [SUCCESS]
Creating Enrollment Templates (Single Process) [SUCCESS]
Creating Enrollment Templates on Multiple Images per Subject (Single Process) [SUCCESS]
Creating Enrollment Templates (Multiple Processes) PID 4474 exited due to signal 9
PID 4473 exited due to signal 9
[SUCCESS]
Creating Verification Templates (Multiple Processes) PID 4491 exited due to signal 9
PID 4493 exited due to signal 9
[SUCCESS]
Matching Templates (Multiple Processes) [ERROR] Failed to open stream for validation/templates/332.template.
[ERROR] Unable to retrieve template from file : validation/templates/332.template
[ERROR] Failed to open stream for validation/templates/657.template.
[ERROR] Unable to retrieve template from file : validation/templates/657.template
PID 4513 exited due to signal 15
[ERROR] Failed to open stream for validation/templates/985.template.
[ERROR] Unable to retrieve template from file : validation/templates/985.template
PID 4514 exited due to signal 15
PID 4515 exited due to signal 15
[ERROR] Match validation failed
```
| {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/119/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/119/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/117 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/117/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/117/comments | https://api.github.com/repos/usnistgov/frvt/issues/117/events | https://github.com/usnistgov/frvt/issues/117 | 2,001,307,066 | I_kwDOBIKFiM53SYW6 | 117 | Segmentation fault when submit | {
"login": "minhsaco99",
"id": 63111142,
"node_id": "MDQ6VXNlcjYzMTExMTQy",
"avatar_url": "https://avatars.githubusercontent.com/u/63111142?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/minhsaco99",
"html_url": "https://github.com/minhsaco99",
"followers_url": "https://api.github.com/users/minhsaco99/followers",
"following_url": "https://api.github.com/users/minhsaco99/following{/other_user}",
"gists_url": "https://api.github.com/users/minhsaco99/gists{/gist_id}",
"starred_url": "https://api.github.com/users/minhsaco99/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/minhsaco99/subscriptions",
"organizations_url": "https://api.github.com/users/minhsaco99/orgs",
"repos_url": "https://api.github.com/users/minhsaco99/repos",
"events_url": "https://api.github.com/users/minhsaco99/events{/privacy}",
"received_events_url": "https://api.github.com/users/minhsaco99/received_events",
"type": "User",
"site_admin": false
} | [] | open | false | null | [] | null | [] | 2023-11-20T03:46:54 | 2023-11-20T03:46:54 | null | NONE | null | null | null | I successfully run my app with run_validate_11.sh but when submit, the organization run raise segmentation fault with some off angle visa images. I try to get the organization providing visa image and successfully process it. I don't know why this error happen. Can someone help me ? Here are the bug information
```
0x00007f7a02bd82c0` in void __gnu_cxx::new_allocator<float>::construct<float, float const&>(float*, float const&) () from techainer_001/config/../lib/libfrvt_11_techainer_001.so
(gdb) where
#0 0x00007f7a02bd82c0 in void __gnu_cxx::new_allocator<float>::construct<float, float const&>(float*, float const&) () from techainer_001/config/../lib/libfrvt_11_techainer_001.so
#1 0x00007f7a02bd530c in void std::allocator_traits<std::allocator<float> >::construct<float, float const&>(std::allocator<float>&, float*, float const&) () from techainer_001/config/../lib/libfrvt_11_techainer_001.so
#2 0x00007f7a02bd5432 in void std::vector<float, std::allocator<float> >::_M_realloc_insert<float const&>(__gnu_cxx::__normal_iterator<float*, std::vector<float, std::allocator<float> > >, float const&) () from techainer_001/config/../lib/libfrvt_11_techainer_001.so
#3 0x00007f7a02bd082a in float& std::vector<float, std::allocator<float> >::emplace_back<float const&>(float const&) () from techainer_001/config/../lib/libfrvt_11_techainer_001.so
#4 0x00007f7a02be02a1 in FRVT_11::TechainerImplFRVT11::createFaceTemplate(std::vector<FRVT::Image, std::allocator<FRVT::Image> > const&, FRVT::TemplateRole, std::vector<unsigned char, std::allocator<unsigned char> >&, std::vector<FRVT::EyePair, std::allocator<FRVT::EyePair> >&) () from techainer_001/config/../lib/libfrvt_11_techainer_001.so
#5 0x000055d95fe841d2 in FRVT_11::TemplateProcessor::processSingleTemplate(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > > const&, std::vector<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::allocator<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > > const&, std::vector<FRVT::Image, std::allocator<FRVT::Image> > const&) ()
#6 0x000055d95fe87a5b in FRVT_11::TemplateProcessor::processLine(unsigned long, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) ()
#7 0x00007f7a03282038 in BiometricEvaluation::MPI::CSVProcessor::processWorkPackage(BiometricEvaluation::MPI::WorkPackage&) () from techainer_001/testdriver/libbiomeval.so.12
#8 0x00007f7a0327d30f in BiometricEvaluation::MPI::Receiver::PackageWorker::workerMain() () from techainer_001/testdriver/libbiomeval.so.12
#9 0x00007f7a03206b2f in BiometricEvaluation::Process::ForkWorkerController::start(bool) () from techainer_001/testdriver/libbiomeval.so.12
#10 0x00007f7a03206744 in BiometricEvaluation::Process::ForkManager::startWorker(std::shared_ptr<BiometricEvaluation::Process::WorkerController>, bool, bool) () from techainer_001/testdriver/libbiomeval.so.12
#11 0x00007f7a0327f2b1 in BiometricEvaluation::MPI::Receiver::startWorkers() () from techainer_001/testdriver/libbiomeval.so.12
#12 0x00007f7a0327f74a in BiometricEvaluation::MPI::Receiver::start() () from techainer_001/testdriver/libbiomeval.so.12
#13 0x00007f7a032739bb in BiometricEvaluation::MPI::Runtime::start(BiometricEvaluation::MPI::Distributor&, BiometricEvaluation::MPI::Receiver&) () from techainer_001/testdriver/libbiomeval.so.12
#14 0x000055d95fe79ecc in main ()
``` | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/117/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/117/timeline | null | null |
https://api.github.com/repos/usnistgov/frvt/issues/114 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/114/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/114/comments | https://api.github.com/repos/usnistgov/frvt/issues/114/events | https://github.com/usnistgov/frvt/issues/114 | 1,972,084,889 | I_kwDOBIKFiM51i6CZ | 114 | why is Finalization function's time limited 40000 sec ? | {
"login": "shritor",
"id": 149577702,
"node_id": "U_kgDOCOpf5g",
"avatar_url": "https://avatars.githubusercontent.com/u/149577702?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/shritor",
"html_url": "https://github.com/shritor",
"followers_url": "https://api.github.com/users/shritor/followers",
"following_url": "https://api.github.com/users/shritor/following{/other_user}",
"gists_url": "https://api.github.com/users/shritor/gists{/gist_id}",
"starred_url": "https://api.github.com/users/shritor/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/shritor/subscriptions",
"organizations_url": "https://api.github.com/users/shritor/orgs",
"repos_url": "https://api.github.com/users/shritor/repos",
"events_url": "https://api.github.com/users/shritor/events{/privacy}",
"received_events_url": "https://api.github.com/users/shritor/received_events",
"type": "User",
"site_admin": false
} | [] | open | false | null | [] | null | [
"Time is limited because a developers might choose to do an expensive computation instead of a copy.\r\n\r\n\r\nFrom: shritor ***@***.***>\r\nDate: Wednesday, November 1, 2023 at 6:17 AM\r\nTo: usnistgov/frvt ***@***.***>\r\nCc: Subscribed ***@***.***>\r\nSubject: [usnistgov/frvt] why is Finalization function's time limited 40000 sec ? (Issue #114)\r\n\r\nenroll function in validate1N.cpp call createTemplate from list ppm in input file then write into edb file and manifest.\r\nmy createTemplate function, I created Feature Vector.\r\nand I used initializeIdentification func to read edb and manifest, init search algorithms when Search Initialization\r\nI think finalization is used to re-write edb and manifest.\r\n\r\n—\r\nReply to this email directly, view it on GitHub<https://github.com/usnistgov/frvt/issues/114>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ADVZFNLPKKDKPW63BJYY4ELYCIOTXAVCNFSM6AAAAAA6ZB3KI2VHI2DSMVQWIX3LMV43ASLTON2WKOZRHE3TEMBYGQ4DQOI>.\r\nYou are receiving this because you are subscribed to this thread.Message ID: ***@***.***>\r\n",
"@paddygr \r\n![Screenshot from 2024-06-05 11-27-59](https://github.com/usnistgov/frvt/assets/149577702/e75a8fac-96e7-4e37-941c-26c2c73ec71a)\r\nCan you tell me why there is no result of 3M and 12M vector ? alttekglobal_000",
"Accuracy is very poor; it’s not a good use of resources to run at large N.\r\n\r\nFrom: shritor ***@***.***>\r\nDate: Sunday, June 9, 2024 at 9:48 PM\r\nTo: usnistgov/frvt ***@***.***>\r\nCc: Grother, Patrick J. (Fed) ***@***.***>, Mention ***@***.***>\r\nSubject: Re: [usnistgov/frvt] why is Finalization function's time limited 40000 sec ? (Issue #114)\r\n\r\n@paddygr<https://github.com/paddygr>\r\nScreenshot.from.2024-06-05.11-27-59.png (view on web)<https://github.com/usnistgov/frvt/assets/149577702/e75a8fac-96e7-4e37-941c-26c2c73ec71a>\r\nCan you tell me why there is no result of 3M and 12M vector ? alttekglobal_000\r\n\r\n—\r\nReply to this email directly, view it on GitHub<https://github.com/usnistgov/frvt/issues/114#issuecomment-2157033832>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ADVZFNN2KAPINBRMOVJKLEDZGUAVBAVCNFSM6AAAAABJBMK5BKVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCNJXGAZTGOBTGI>.\r\nYou are receiving this because you were mentioned.Message ID: ***@***.***>\r\n"
] | 2023-11-01T10:17:18 | 2024-06-10T11:40:26 | null | NONE | null | null | null | enroll function in validate1N.cpp call createTemplate from list ppm in input file then write into edb file and manifest.
My createTemplate function, I created Feature Vector.
I used initializeIdentification func to read edb and manifest, init search algorithms when Search Initialization
I think finalization is used to re-write edb and manifest.
| {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/114/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/114/timeline | null | null |
https://api.github.com/repos/usnistgov/frvt/issues/113 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/113/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/113/comments | https://api.github.com/repos/usnistgov/frvt/issues/113/events | https://github.com/usnistgov/frvt/issues/113 | 1,956,324,514 | I_kwDOBIKFiM50mySi | 113 | [FRVT 1:1] Can I edit the header file? | {
"login": "yhbembedding",
"id": 55545444,
"node_id": "MDQ6VXNlcjU1NTQ1NDQ0",
"avatar_url": "https://avatars.githubusercontent.com/u/55545444?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/yhbembedding",
"html_url": "https://github.com/yhbembedding",
"followers_url": "https://api.github.com/users/yhbembedding/followers",
"following_url": "https://api.github.com/users/yhbembedding/following{/other_user}",
"gists_url": "https://api.github.com/users/yhbembedding/gists{/gist_id}",
"starred_url": "https://api.github.com/users/yhbembedding/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/yhbembedding/subscriptions",
"organizations_url": "https://api.github.com/users/yhbembedding/orgs",
"repos_url": "https://api.github.com/users/yhbembedding/repos",
"events_url": "https://api.github.com/users/yhbembedding/events{/privacy}",
"received_events_url": "https://api.github.com/users/yhbembedding/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"You cannot modify any of the source code that we have published. Please consider the example provided in https://github.com/usnistgov/frvt/blob/master/11/src/nullImpl/nullimplfrvt11.cpp on how to convert between uint8_t and float.",
"Thanks for your answer, in file cpp, matching is random. Can I modify it.\r\n![Screenshot from 2023-10-24 08-15-49](https://github.com/usnistgov/frvt/assets/55545444/cc8014f3-000b-4354-b6f3-d7199a5f53ae)\r\nWhat are editable files?, util.h, util.cpp,validate11.cpp ?",
"You implement whatever match you want, in your compiled library. You must not change the API. You cannot edit those files.\r\n\r\nFrom: Hoàng Bá Ý ***@***.***>\r\nDate: Monday, October 23, 2023 at 9:20 PM\r\nTo: usnistgov/frvt ***@***.***>\r\nCc: Subscribed ***@***.***>\r\nSubject: Re: [usnistgov/frvt] [FRVT 1:1] Can I edit the header file? (Issue #113)\r\n\r\nThanks for your answer, in file cpp, matching is random. Can I modify it.\r\n[Image removed by sender. Screenshot from 2023-10-24 08-15-49]<https://user-images.githubusercontent.com/55545444/277520535-cc8014f3-000b-4354-b6f3-d7199a5f53ae.png>\r\nWhat are editable files?, util.h, util.cpp,validate11.cpp ?\r\n\r\n—\r\nReply to this email directly, view it on GitHub<https://github.com/usnistgov/frvt/issues/113#issuecomment-1776312895>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ADVZFNJ4MBWQCZMGHTDRHNDYA4JVNAVCNFSM6AAAAAA6LOCC42VHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTONZWGMYTEOBZGU>.\r\nYou are receiving this because you are subscribed to this thread.Message ID: ***@***.***>\r\n",
"![Screenshot from 2023-11-30 09-19-34](https://github.com/usnistgov/frvt/assets/149577702/45ea4c75-e23d-4a82-b1d4-a8ac9a0c50e7)\r\nsome members ? can I add some attributes in nullimplfrvt11.h @mlngan @paddygr @yhbembedding "
] | 2023-10-23T04:22:10 | 2023-11-30T02:26:34 | 2023-10-23T16:01:31 | NONE | null | null | null | nullimplfrvt11.h
I fix it from uint8_t to float.
![Screenshot from 2023-10-23 11-17-58](https://github.com/usnistgov/frvt/assets/55545444/463cad6c-95ac-4e06-a5f6-96f09b1065db)
| {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/113/reactions",
"total_count": 2,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 1,
"heart": 0,
"rocket": 0,
"eyes": 1
} | https://api.github.com/repos/usnistgov/frvt/issues/113/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/112 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/112/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/112/comments | https://api.github.com/repos/usnistgov/frvt/issues/112/events | https://github.com/usnistgov/frvt/issues/112 | 1,948,921,040 | I_kwDOBIKFiM50KizQ | 112 | What is include in the 1:N finalization time limited? | {
"login": "LionLion888",
"id": 133086055,
"node_id": "U_kgDOB-67Zw",
"avatar_url": "https://avatars.githubusercontent.com/u/133086055?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/LionLion888",
"html_url": "https://github.com/LionLion888",
"followers_url": "https://api.github.com/users/LionLion888/followers",
"following_url": "https://api.github.com/users/LionLion888/following{/other_user}",
"gists_url": "https://api.github.com/users/LionLion888/gists{/gist_id}",
"starred_url": "https://api.github.com/users/LionLion888/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/LionLion888/subscriptions",
"organizations_url": "https://api.github.com/users/LionLion888/orgs",
"repos_url": "https://api.github.com/users/LionLion888/repos",
"events_url": "https://api.github.com/users/LionLion888/events{/privacy}",
"received_events_url": "https://api.github.com/users/LionLion888/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"Answer is B. Finalization takes N templates and produces a searchable data structure.\r\n\r\nFrom: LionLion888 ***@***.***>\r\nDate: Wednesday, October 18, 2023 at 2:13 AM\r\nTo: usnistgov/frvt ***@***.***>\r\nCc: Subscribed ***@***.***>\r\nSubject: [usnistgov/frvt] What is include in the 1:N finalization time limited? (Issue #112)\r\n\r\nHi there,\r\n\r\nIn the FRVT 1:N time limits (5.1. Time limits<https://pages.nist.gov/frvt/api/FRVT_IREX_ongoing_1N_api.pdf>), I'm wondering to know what is included in the 1:N finalization time limited (mark with red box)?\r\n[https://user-images.githubusercontent.com/133086055/276111081-9875eba9-52a5-449e-97a1-2a8d797343dc.png]<https://user-images.githubusercontent.com/133086055/276111081-9875eba9-52a5-449e-97a1-2a8d797343dc.png>\r\n\r\nIn detail, which following is 1:N finalization time limited (mark with red box) represented?\r\n\r\nA. Feature extraction execution time (mark with blue box) + Enrolled database execution time (mark with green box)\r\n\r\nB. Only include enrolled database execution time (mark with green box)\r\n[https://user-images.githubusercontent.com/133086055/276111122-6932e77a-9136-4429-9bad-5faacad643c9.png]<https://user-images.githubusercontent.com/133086055/276111122-6932e77a-9136-4429-9bad-5faacad643c9.png>\r\n\r\n—\r\nReply to this email directly, view it on GitHub<https://github.com/usnistgov/frvt/issues/112>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ADVZFNOPF3GG4OCX44TCTYLX75XOZAVCNFSM6AAAAAA6E6D7K2VHI2DSMVQWIX3LMV43ASLTON2WKOZRHE2DQOJSGEYDIMA>.\r\nYou are receiving this because you are subscribed to this thread.Message ID: ***@***.***>\r\n",
"Hi paddygr,\r\n\r\nI got it! Thanks for your response."
] | 2023-10-18T06:12:46 | 2023-10-19T13:28:08 | 2023-10-19T13:28:08 | NONE | null | null | null | Hi there,
In the FRVT 1:N time limits ([5.1. Time limits](https://pages.nist.gov/frvt/api/FRVT_IREX_ongoing_1N_api.pdf)), I'm wondering to know what is included in the **1:N finalization time limited** (mark with red box)?
<img src="https://github.com/usnistgov/frvt/assets/133086055/9875eba9-52a5-449e-97a1-2a8d797343dc" width="70%" >
In detail, which following is 1:N finalization time limited (mark with red box) represented?
A. Feature extraction execution time (mark with blue box) + Enrolled database execution time (mark with green box)
B. Only include enrolled database execution time (mark with green box)
<img src="https://github.com/usnistgov/frvt/assets/133086055/6932e77a-9136-4429-9bad-5faacad643c9" width="70%" >
| {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/112/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/112/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/111 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/111/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/111/comments | https://api.github.com/repos/usnistgov/frvt/issues/111/events | https://github.com/usnistgov/frvt/issues/111 | 1,925,590,017 | I_kwDOBIKFiM5yxiwB | 111 | Time Limits clarification for creation of M Templates from a single image | {
"login": "ElisonSherton",
"id": 45599762,
"node_id": "MDQ6VXNlcjQ1NTk5NzYy",
"avatar_url": "https://avatars.githubusercontent.com/u/45599762?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ElisonSherton",
"html_url": "https://github.com/ElisonSherton",
"followers_url": "https://api.github.com/users/ElisonSherton/followers",
"following_url": "https://api.github.com/users/ElisonSherton/following{/other_user}",
"gists_url": "https://api.github.com/users/ElisonSherton/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ElisonSherton/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ElisonSherton/subscriptions",
"organizations_url": "https://api.github.com/users/ElisonSherton/orgs",
"repos_url": "https://api.github.com/users/ElisonSherton/repos",
"events_url": "https://api.github.com/users/ElisonSherton/events{/privacy}",
"received_events_url": "https://api.github.com/users/ElisonSherton/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"All the images you are talking about have just a single face even if you detect thousands.\r\nYou get 1500ms per image (which is one face).",
"In `4.4.4` of the API document, we noticed the following\r\n\r\n![image](https://github.com/usnistgov/frvt/assets/45599762/f9e6e16f-3680-4e2d-8fdb-33d418a7f4ec)\r\n\r\nSo, we do have images which have more than one face in them right?\r\n\r\nThanks,\r\nVinayak.",
"Yeah there is just one foreground face",
"Hi @xsacha \r\n\r\nAgreed that there is one foreground face, but since we're not aware of which face is foreground or background, we can return M templates and then as mentioned in points 3 and 4 of section 4.4.4, matching could be done across all sets of enrollment and verification templates right?\r\n\r\nFor that we need to send multiple face templates per image and for that I was asking will we get more time or do we still have to abide to 1500 ms for getting the templates for all the faces from a single image?\r\n\r\nThanks,\r\nVinayak.",
"We are only conducting formal timing runs on single-person (single template) images. While we are not enforcing timing limits for multi-person (multiple templates) images, please implement something sensible (e.g., 1500 ms * the number of faces detected would be OK).",
"Thanks for the clarification @mlngan \r\n\r\nThis really helps!"
] | 2023-10-04T07:37:45 | 2023-10-06T12:31:54 | 2023-10-05T17:11:34 | CONTRIBUTOR | null | null | null | Hi everyone!
In **Table 6** from [this document](https://pages.nist.gov/frvt/api/FRVT_ongoing_11_api.pdf), we can see that there is a single image and if there are M faces which our algorithm detects, then we will return M templates in the output of `createFaceTemplate` API call.
However as per **Table 1** from the same document, we can see that feature extraction should only take 1500 ms per image. And only if there are `K` images of a single person, the time limit would be scaled up by that factor.
Am I correct in understanding following scenarios?
- We detect 5 faces in an image. Do we get 1500 ms to perform feature extraction of all the five faces **OR** do we get 1500 $\times$ 5 = 7500 ms
- We are given 5 images and we detect 1 face in each image. Will we get 1500 $\times$ 5 = 7500 ms to do the feature extraction?
Thanks,
Vinayak. | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/111/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/111/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/110 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/110/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/110/comments | https://api.github.com/repos/usnistgov/frvt/issues/110/events | https://github.com/usnistgov/frvt/issues/110 | 1,913,517,055 | I_kwDOBIKFiM5yDfP_ | 110 | Inaccurate Feature Length Calculation Due to This Err | {
"login": "mkwml",
"id": 34429706,
"node_id": "MDQ6VXNlcjM0NDI5NzA2",
"avatar_url": "https://avatars.githubusercontent.com/u/34429706?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/mkwml",
"html_url": "https://github.com/mkwml",
"followers_url": "https://api.github.com/users/mkwml/followers",
"following_url": "https://api.github.com/users/mkwml/following{/other_user}",
"gists_url": "https://api.github.com/users/mkwml/gists{/gist_id}",
"starred_url": "https://api.github.com/users/mkwml/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/mkwml/subscriptions",
"organizations_url": "https://api.github.com/users/mkwml/orgs",
"repos_url": "https://api.github.com/users/mkwml/repos",
"events_url": "https://api.github.com/users/mkwml/events{/privacy}",
"received_events_url": "https://api.github.com/users/mkwml/received_events",
"type": "User",
"site_admin": false
} | [] | open | false | null | [] | null | [
"Can you please provide more details/clarification on the error that you're encountering?",
"Hello, @mkwml do you know which line is causing the errors? I am facing the same issue currently as well."
] | 2023-09-26T13:28:19 | 2024-06-19T03:27:02 | null | NONE | null | null | null | https://github.com/usnistgov/frvt/blob/ec6188c89acffe71c68cdd41c7ffa2cb64a33a7e/1N/src/testdriver/validate1N.cpp#L62C5-L62C5
The current code contains an error that leads to inaccurate calculation of feature length, affecting the overall accuracy of the finall result. | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/110/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/110/timeline | null | null |
https://api.github.com/repos/usnistgov/frvt/issues/109 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/109/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/109/comments | https://api.github.com/repos/usnistgov/frvt/issues/109/events | https://github.com/usnistgov/frvt/issues/109 | 1,909,844,663 | I_kwDOBIKFiM5x1eq3 | 109 | Additional supporting library compilation on different hardware. | {
"login": "jaisanant0",
"id": 21287452,
"node_id": "MDQ6VXNlcjIxMjg3NDUy",
"avatar_url": "https://avatars.githubusercontent.com/u/21287452?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jaisanant0",
"html_url": "https://github.com/jaisanant0",
"followers_url": "https://api.github.com/users/jaisanant0/followers",
"following_url": "https://api.github.com/users/jaisanant0/following{/other_user}",
"gists_url": "https://api.github.com/users/jaisanant0/gists{/gist_id}",
"starred_url": "https://api.github.com/users/jaisanant0/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/jaisanant0/subscriptions",
"organizations_url": "https://api.github.com/users/jaisanant0/orgs",
"repos_url": "https://api.github.com/users/jaisanant0/repos",
"events_url": "https://api.github.com/users/jaisanant0/events{/privacy}",
"received_events_url": "https://api.github.com/users/jaisanant0/received_events",
"type": "User",
"site_admin": false
} | [] | open | false | null | [] | null | [] | 2023-09-23T12:02:30 | 2023-09-23T12:02:30 | null | NONE | null | null | null | Hi
My package has been built successfully and it needs additional libraries which i have put in lib/ directory.
I have doubt that do we need to compile additional library on the hardware mentioned by NIST like INTEL XEON since package and supporting library is compile on different x86 hardware but with same .iso image.
| {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/109/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/109/timeline | null | null |
https://api.github.com/repos/usnistgov/frvt/issues/108 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/108/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/108/comments | https://api.github.com/repos/usnistgov/frvt/issues/108/events | https://github.com/usnistgov/frvt/issues/108 | 1,880,322,409 | I_kwDOBIKFiM5wE3Fp | 108 | run_validate_ae.sh throws warnings from the sanityCheck | {
"login": "golunovas",
"id": 19951039,
"node_id": "MDQ6VXNlcjE5OTUxMDM5",
"avatar_url": "https://avatars.githubusercontent.com/u/19951039?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/golunovas",
"html_url": "https://github.com/golunovas",
"followers_url": "https://api.github.com/users/golunovas/followers",
"following_url": "https://api.github.com/users/golunovas/following{/other_user}",
"gists_url": "https://api.github.com/users/golunovas/gists{/gist_id}",
"starred_url": "https://api.github.com/users/golunovas/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/golunovas/subscriptions",
"organizations_url": "https://api.github.com/users/golunovas/orgs",
"repos_url": "https://api.github.com/users/golunovas/repos",
"events_url": "https://api.github.com/users/golunovas/events{/privacy}",
"received_events_url": "https://api.github.com/users/golunovas/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"It has been corrected, please update the file and thank you."
] | 2023-09-04T13:43:32 | 2023-09-05T01:41:01 | 2023-09-05T01:41:01 | NONE | null | null | null | During the validation process I'm getting the following warnings:
```
Sanity checking validation output
validation/estimateAge_1.log -
[WARNING] The following entries in .log generated non-successful return codes:
sed: can't read validation/.log: No such file or directory
[DONE]
validation/estimateAge_2.log - [DONE]
validation/verifyAge.log -
[WARNING] The following entries in .log generated non-successful return codes:
sed: can't read validation/.log: No such file or directory
[DONE]
```
It looks like `$action` variable is not defined here: https://github.com/usnistgov/frvt/blob/master/age-estimation/run_validate_ae.sh#L50 and the same issue is present here: https://github.com/usnistgov/frvt/blob/master/age-estimation/run_validate_ae.sh#L42 | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/108/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/108/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/107 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/107/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/107/comments | https://api.github.com/repos/usnistgov/frvt/issues/107/events | https://github.com/usnistgov/frvt/issues/107 | 1,866,135,479 | I_kwDOBIKFiM5vOve3 | 107 | Does 1:1 Face test on images rotated to 90 or 180 degree? | {
"login": "JackLin-Authme",
"id": 66769915,
"node_id": "MDQ6VXNlcjY2NzY5OTE1",
"avatar_url": "https://avatars.githubusercontent.com/u/66769915?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/JackLin-Authme",
"html_url": "https://github.com/JackLin-Authme",
"followers_url": "https://api.github.com/users/JackLin-Authme/followers",
"following_url": "https://api.github.com/users/JackLin-Authme/following{/other_user}",
"gists_url": "https://api.github.com/users/JackLin-Authme/gists{/gist_id}",
"starred_url": "https://api.github.com/users/JackLin-Authme/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/JackLin-Authme/subscriptions",
"organizations_url": "https://api.github.com/users/JackLin-Authme/orgs",
"repos_url": "https://api.github.com/users/JackLin-Authme/repos",
"events_url": "https://api.github.com/users/JackLin-Authme/events{/privacy}",
"received_events_url": "https://api.github.com/users/JackLin-Authme/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"We do not intend on testing with rotated images in our actual test. HOWEVER, these edge cases were put into the validation package to test that your algorithm does not crash in the event such unexpected cases are encountered.",
"@mlngan and what about 1:N? Does it have rotated (90, 180) images in validation datasets?",
"Yes. 1:1 and 1:N share a number of images in the validation set.",
"> Yes. 1:1 and 1:N share a number of images in the validation set.\r\n\r\nThanks. And what about the _evaluation_ 1:N dataset? Should we expect rotated images there?",
"We do not intend on testing with rotated images in the actual 1:N test."
] | 2023-08-25T01:58:32 | 2023-08-28T16:47:57 | 2023-08-25T12:21:05 | NONE | null | null | null | I saw 2 cases in common/faces folder, rotated_90.ppm and rotated_180.ppm.
Does it mean that there are some rotated image will be test in 1:1 protocol? | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/107/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/107/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/106 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/106/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/106/comments | https://api.github.com/repos/usnistgov/frvt/issues/106/events | https://github.com/usnistgov/frvt/issues/106 | 1,863,111,525 | I_kwDOBIKFiM5vDNNl | 106 | misssed "faceunknown" mark at enroll_1N_multiface.txt | {
"login": "SungjunKim1",
"id": 37128354,
"node_id": "MDQ6VXNlcjM3MTI4MzU0",
"avatar_url": "https://avatars.githubusercontent.com/u/37128354?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/SungjunKim1",
"html_url": "https://github.com/SungjunKim1",
"followers_url": "https://api.github.com/users/SungjunKim1/followers",
"following_url": "https://api.github.com/users/SungjunKim1/following{/other_user}",
"gists_url": "https://api.github.com/users/SungjunKim1/gists{/gist_id}",
"starred_url": "https://api.github.com/users/SungjunKim1/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/SungjunKim1/subscriptions",
"organizations_url": "https://api.github.com/users/SungjunKim1/orgs",
"repos_url": "https://api.github.com/users/SungjunKim1/repos",
"events_url": "https://api.github.com/users/SungjunKim1/events{/privacy}",
"received_events_url": "https://api.github.com/users/SungjunKim1/received_events",
"type": "User",
"site_admin": false
} | [] | open | false | null | [] | null | [] | 2023-08-23T11:08:41 | 2023-08-23T11:08:41 | null | NONE | null | null | null | in frvt/1N/input/face/enroll_1N_multiface.txt,
```
21 ../common/images/face/S008-02-t10_01.ppm faceunknown ../common/images/face/S008-04-t10_01.ppm faceunknown ../common/images/face/S008-06-t10_01.ppm
```
It's not important... but take note
Thanks
| {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/106/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/106/timeline | null | null |
https://api.github.com/repos/usnistgov/frvt/issues/105 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/105/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/105/comments | https://api.github.com/repos/usnistgov/frvt/issues/105/events | https://github.com/usnistgov/frvt/issues/105 | 1,862,845,833 | I_kwDOBIKFiM5vCMWJ | 105 | `<` wrong syntax in utils.h | {
"login": "JackLin-Authme",
"id": 66769915,
"node_id": "MDQ6VXNlcjY2NzY5OTE1",
"avatar_url": "https://avatars.githubusercontent.com/u/66769915?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/JackLin-Authme",
"html_url": "https://github.com/JackLin-Authme",
"followers_url": "https://api.github.com/users/JackLin-Authme/followers",
"following_url": "https://api.github.com/users/JackLin-Authme/following{/other_user}",
"gists_url": "https://api.github.com/users/JackLin-Authme/gists{/gist_id}",
"starred_url": "https://api.github.com/users/JackLin-Authme/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/JackLin-Authme/subscriptions",
"organizations_url": "https://api.github.com/users/JackLin-Authme/orgs",
"repos_url": "https://api.github.com/users/JackLin-Authme/repos",
"events_url": "https://api.github.com/users/JackLin-Authme/events{/privacy}",
"received_events_url": "https://api.github.com/users/JackLin-Authme/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"fixed the issue and file is updated, thank you."
] | 2023-08-23T08:35:50 | 2023-08-23T10:59:21 | 2023-08-23T10:59:21 | NONE | null | null | null | https://github.com/usnistgov/frvt/blob/b73285b59ab6b6b6b00a359ccd87bf06f7cba049/common/src/include/util.h#L122
https://github.com/usnistgov/frvt/blob/b73285b59ab6b6b6b00a359ccd87bf06f7cba049/common/src/include/util.h#L123
There are unexpected characters `<` in util.h leading fail.
<img width="1020" alt="截圖 2023-08-23 16 38 28" src="https://github.com/usnistgov/frvt/assets/66769915/690ed0f7-0630-4b73-bfd3-db14716e5e83"> | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/105/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/105/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/104 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/104/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/104/comments | https://api.github.com/repos/usnistgov/frvt/issues/104/events | https://github.com/usnistgov/frvt/issues/104 | 1,862,340,829 | I_kwDOBIKFiM5vARDd | 104 | Undefined enums | {
"login": "hiroosaito",
"id": 142949434,
"node_id": "U_kgDOCIU8Og",
"avatar_url": "https://avatars.githubusercontent.com/u/142949434?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/hiroosaito",
"html_url": "https://github.com/hiroosaito",
"followers_url": "https://api.github.com/users/hiroosaito/followers",
"following_url": "https://api.github.com/users/hiroosaito/following{/other_user}",
"gists_url": "https://api.github.com/users/hiroosaito/gists{/gist_id}",
"starred_url": "https://api.github.com/users/hiroosaito/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/hiroosaito/subscriptions",
"organizations_url": "https://api.github.com/users/hiroosaito/orgs",
"repos_url": "https://api.github.com/users/hiroosaito/repos",
"events_url": "https://api.github.com/users/hiroosaito/events{/privacy}",
"received_events_url": "https://api.github.com/users/hiroosaito/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"Updated the file, please try again. Thank you.",
"It still doesn't work. There are two \"<\" symbols in lines 122-123 that need to be removed.",
"file is fixed, try again thank you."
] | 2023-08-22T23:41:37 | 2023-08-23T11:01:23 | 2023-08-23T00:26:25 | NONE | null | null | null | The enums Action::EstimateAge and Action::VerifyAge in "common/src/util.cpp" are not defined in "common/src/include/util.h", and which causes a compile error.
Could you take care of the issue? | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/104/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/104/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/103 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/103/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/103/comments | https://api.github.com/repos/usnistgov/frvt/issues/103/events | https://github.com/usnistgov/frvt/issues/103 | 1,849,480,777 | I_kwDOBIKFiM5uPNZJ | 103 | Is there any memory limit in 1:N Identification? | {
"login": "PeggyWang888",
"id": 25610609,
"node_id": "MDQ6VXNlcjI1NjEwNjA5",
"avatar_url": "https://avatars.githubusercontent.com/u/25610609?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/PeggyWang888",
"html_url": "https://github.com/PeggyWang888",
"followers_url": "https://api.github.com/users/PeggyWang888/followers",
"following_url": "https://api.github.com/users/PeggyWang888/following{/other_user}",
"gists_url": "https://api.github.com/users/PeggyWang888/gists{/gist_id}",
"starred_url": "https://api.github.com/users/PeggyWang888/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/PeggyWang888/subscriptions",
"organizations_url": "https://api.github.com/users/PeggyWang888/orgs",
"repos_url": "https://api.github.com/users/PeggyWang888/repos",
"events_url": "https://api.github.com/users/PeggyWang888/events{/privacy}",
"received_events_url": "https://api.github.com/users/PeggyWang888/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"Per Section 4 of the [General Evaluation Specifications](https://pages.nist.gov/frvt/api/FRVT_common.pdf), each evaluation machine has at least 128 GB of memory. We anticipate that 16 processes can be run without time slicing, though NIST will handle all multiprocessing work via fork(). Hope this information helps."
] | 2023-08-14T10:22:32 | 2023-08-14T11:57:59 | 2023-08-14T11:57:58 | NONE | null | null | null | I am working on FRVT 1:N, and my query is : is there any memory limit in 1:N Identification? | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/103/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/103/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/102 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/102/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/102/comments | https://api.github.com/repos/usnistgov/frvt/issues/102/events | https://github.com/usnistgov/frvt/issues/102 | 1,841,017,579 | I_kwDOBIKFiM5tu7Lr | 102 | What kind of vector is in the template? 1:N | {
"login": "LionLion888",
"id": 133086055,
"node_id": "U_kgDOB-67Zw",
"avatar_url": "https://avatars.githubusercontent.com/u/133086055?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/LionLion888",
"html_url": "https://github.com/LionLion888",
"followers_url": "https://api.github.com/users/LionLion888/followers",
"following_url": "https://api.github.com/users/LionLion888/following{/other_user}",
"gists_url": "https://api.github.com/users/LionLion888/gists{/gist_id}",
"starred_url": "https://api.github.com/users/LionLion888/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/LionLion888/subscriptions",
"organizations_url": "https://api.github.com/users/LionLion888/orgs",
"repos_url": "https://api.github.com/users/LionLion888/repos",
"events_url": "https://api.github.com/users/LionLion888/events{/privacy}",
"received_events_url": "https://api.github.com/users/LionLion888/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"createTemplate produces feature data from one image. The function takes one image, so there is nothing to match against, so there is no recognition, no similarity score.\r\n\r\nFrom: LionLion888 ***@***.***>\r\nDate: Tuesday, August 8, 2023 at 6:18 AM\r\nTo: usnistgov/frvt ***@***.***>\r\nCc: Subscribed ***@***.***>\r\nSubject: [usnistgov/frvt] What kind of vector is in the template? 1:N (Issue #102)\r\n\r\nHi there,\r\n\r\nWhen createTemplate(TemplateRole=Enrollment_1N), the implementation will return a template to the calling application.\r\n\r\nI am wondering to know what kind of vector is in the template? Is it a feature from the face detection or it can be a similarity score produce by the face recognition as well.\r\n\r\n[Image removed by sender.]<https://camo.githubusercontent.com/3ca920d1ec7bce0560c019b1465b782d8c384f2cdecb238e67c77556b444f8bd/68747470733a2f2f696d6775722e636f6d2f505a58624333692e706e67>\r\n\r\nBesides, if the vector in the template can be a similarity score, does that mean the Template Generation will deal with face detection and face recognition, also the total of the face detection and face recognition execution time need to within 1.5 sec.\r\n\r\n[Image removed by sender.]<https://camo.githubusercontent.com/dcf20b311188703d4941006761e76f94684dae44375932d1d61d1542e027b360/68747470733a2f2f696d6775722e636f6d2f58496f595637382e706e67>\r\n\r\n—\r\nReply to this email directly, view it on GitHub<https://github.com/usnistgov/frvt/issues/102>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ADVZFNNNN4T2HF7RABZ5BC3XUIHANANCNFSM6AAAAAA3IHVDJA>.\r\nYou are receiving this because you are subscribed to this thread.Message ID: ***@***.***>\r\n"
] | 2023-08-08T10:18:33 | 2023-11-23T09:03:53 | 2023-11-23T09:03:53 | NONE | null | null | null | Hi there,
When **createTemplate(TemplateRole=Enrollment_1N)**, the implementation will return a template to the calling application.
I am wondering to know what kind of vector is in the template? Is it a **feature** from the face detection or it can be a **similarity score** produce by the face recognition as well.
<img src=https://imgur.com/PZXbC3i.png width=60% />
Besides, if the vector in the template can be a similarity score, does that mean the **Template Generation** will deal with face detection and face recognition, also the total of the face detection and face recognition execution time need to within 1.5 sec.
<img src=https://imgur.com/XIoYV78.png width=60% /> | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/102/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/102/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/101 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/101/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/101/comments | https://api.github.com/repos/usnistgov/frvt/issues/101/events | https://github.com/usnistgov/frvt/issues/101 | 1,839,088,820 | I_kwDOBIKFiM5tnkS0 | 101 | How many images does search dataset have? 1:N | {
"login": "LionLion888",
"id": 133086055,
"node_id": "U_kgDOB-67Zw",
"avatar_url": "https://avatars.githubusercontent.com/u/133086055?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/LionLion888",
"html_url": "https://github.com/LionLion888",
"followers_url": "https://api.github.com/users/LionLion888/followers",
"following_url": "https://api.github.com/users/LionLion888/following{/other_user}",
"gists_url": "https://api.github.com/users/LionLion888/gists{/gist_id}",
"starred_url": "https://api.github.com/users/LionLion888/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/LionLion888/subscriptions",
"organizations_url": "https://api.github.com/users/LionLion888/orgs",
"repos_url": "https://api.github.com/users/LionLion888/repos",
"events_url": "https://api.github.com/users/LionLion888/events{/privacy}",
"received_events_url": "https://api.github.com/users/LionLion888/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"The search/probe could be any of the following\r\n1. one image of one person (Function A)\r\n2. multiple images of one person (Function A)\r\n3. one image with multiple people in it (Function B)\r\n\r\n**Function A:** \r\nFRVT::ReturnStatus\r\n createFaceTemplate(\r\n const std::vector<FRVT::Image> &faces,\r\n FRVT::TemplateRole role,\r\n std::vector<uint8_t> &templ,\r\n std::vector<FRVT::EyePair> &eyeCoordinates)\r\n\r\n**Function B:** \r\nFRVT::ReturnStatus\r\n createFaceTemplate(\r\n const FRVT::Image &image,\r\n FRVT::TemplateRole role,\r\n std::vector<std::vector<uint8_t>> &templ,\r\n std::vector<FRVT::EyePair> &eyeCoordinates)\r\n",
"Thanks for your reply!",
"Look at the N values at the top of the columns here:\r\nhttps://pages.nist.gov/frvt/html/frvt1N.html\r\n\r\n\r\n\r\nFrom: LionLion888 ***@***.***>\r\nDate: Tuesday, August 8, 2023 at 2:16 AM\r\nTo: usnistgov/frvt ***@***.***>\r\nCc: Subscribed ***@***.***>\r\nSubject: Re: [usnistgov/frvt] How many images does search dataset have? 1:N (Issue #101)\r\n\r\nThanks for your reply!\r\n\r\n—\r\nReply to this email directly, view it on GitHub<https://github.com/usnistgov/frvt/issues/101#issuecomment-1668971736>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ADVZFNP5QKGF34KEOZMPDNTXUHKSRANCNFSM6AAAAAA3GXIAFQ>.\r\nYou are receiving this because you are subscribed to this thread.Message ID: ***@***.***>\r\n"
] | 2023-08-07T09:46:44 | 2023-08-08T11:05:34 | 2023-08-07T13:26:59 | NONE | null | null | null | Hi,
In the 1:N identify project, the enrollment dataset will get 1 million images as input data, if face detection failed, it shall still return a blank template and continue to go through the next enroll image.
For the search dataset, I'm wondering there would be how many images? In the following picture, the search photo seems to be only one image. (Wanted to make sure it wold be **one person with one image** or **one person with multi images**.)
![image](https://github.com/usnistgov/frvt/assets/133086055/90187b80-9501-41f1-b93e-0ef969a7b301)
| {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/101/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/101/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/100 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/100/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/100/comments | https://api.github.com/repos/usnistgov/frvt/issues/100/events | https://github.com/usnistgov/frvt/issues/100 | 1,809,758,807 | I_kwDOBIKFiM5r3rpX | 100 | Python Memory Leaks | {
"login": "AwaisPF",
"id": 97581451,
"node_id": "U_kgDOBdD5iw",
"avatar_url": "https://avatars.githubusercontent.com/u/97581451?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/AwaisPF",
"html_url": "https://github.com/AwaisPF",
"followers_url": "https://api.github.com/users/AwaisPF/followers",
"following_url": "https://api.github.com/users/AwaisPF/following{/other_user}",
"gists_url": "https://api.github.com/users/AwaisPF/gists{/gist_id}",
"starred_url": "https://api.github.com/users/AwaisPF/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/AwaisPF/subscriptions",
"organizations_url": "https://api.github.com/users/AwaisPF/orgs",
"repos_url": "https://api.github.com/users/AwaisPF/repos",
"events_url": "https://api.github.com/users/AwaisPF/events{/privacy}",
"received_events_url": "https://api.github.com/users/AwaisPF/received_events",
"type": "User",
"site_admin": false
} | [] | open | false | null | [] | null | [] | 2023-07-18T11:42:50 | 2023-07-18T11:42:50 | null | NONE | null | null | null | Hi,
I write here for a solution to memory leaks I applied for frvt 1:1 submission but got a memory leak error I have also tested with Valgrind for memory leaks checking there are so many memory leaks detected like `invalid read of size 4` or maybe that is not the memory leak can any provide guidance about to check where exactly memory leaks occur or second approach would be to convert the model into c++ which should be better debugging memory leaks or converting python code into c++
by the way, I am using retinaface library for face detection and insightFace for feature extraction.
here is Valgrind output slice
`==140416== by 0x49F424B: Py_InitializeEx (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x1091ED: main (in /additional_drive/nist/frvt_onnx_sequential/11/test_mem)
==140416== Address 0x5387020 is 1,520 bytes inside a block of size 2,208 free'd
==140416== at 0x483CA3F: free (in /usr/lib/x86_64-linux-gnu/valgrind/vgpreload_memcheck-amd64-linux.so)
==140416== by 0x4ACD397: ??? (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x4AC9E1F: PyDict_SetDefault (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x4A62DFF: PyUnicode_InternInPlace (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x4A62E90: PyUnicode_InternFromString (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x4AFC4C1: PyDescr_NewMethod (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x48E34AA: PyType_Ready (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x4AB8E04: _PyTypes_Init (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x49F3A6A: ??? (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x49F41A2: Py_InitializeFromConfig (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x49F424B: Py_InitializeEx (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x1091ED: main (in /additional_drive/nist/frvt_onnx_sequential/11/test_mem)
==140416== Block was alloc'd at
...
...
...
==140416== Block was alloc'd at
==140416== at 0x483B7F3: malloc (in /usr/lib/x86_64-linux-gnu/valgrind/vgpreload_memcheck-amd64-linux.so)
==140416== by 0x4AB69E7: ??? (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x4B20CB7: PyGrammar_AddAccelerators (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x4B23184: PyParser_New (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x4B26DD2: ??? (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x49E79DA: PyParser_ASTFromStringObject (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x49E8387: PyRun_StringFlags (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x4A2BADF: ??? (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x4ABB546: ??? (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x48D0D6C: ??? (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x48D246C: _PyEval_EvalFrameDefault (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x48DC06A: ??? (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416==
==140416== ERROR SUMMARY: 898 errors from 107 contexts (suppressed: 0 from 0)` | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/100/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/100/timeline | null | null |
https://api.github.com/repos/usnistgov/frvt/issues/98 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/98/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/98/comments | https://api.github.com/repos/usnistgov/frvt/issues/98/events | https://github.com/usnistgov/frvt/issues/98 | 1,769,626,555 | I_kwDOBIKFiM5pelu7 | 98 | valgrind is detecting memory leaks in python code | {
"login": "AwaisPF",
"id": 97581451,
"node_id": "U_kgDOBdD5iw",
"avatar_url": "https://avatars.githubusercontent.com/u/97581451?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/AwaisPF",
"html_url": "https://github.com/AwaisPF",
"followers_url": "https://api.github.com/users/AwaisPF/followers",
"following_url": "https://api.github.com/users/AwaisPF/following{/other_user}",
"gists_url": "https://api.github.com/users/AwaisPF/gists{/gist_id}",
"starred_url": "https://api.github.com/users/AwaisPF/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/AwaisPF/subscriptions",
"organizations_url": "https://api.github.com/users/AwaisPF/orgs",
"repos_url": "https://api.github.com/users/AwaisPF/repos",
"events_url": "https://api.github.com/users/AwaisPF/events{/privacy}",
"received_events_url": "https://api.github.com/users/AwaisPF/received_events",
"type": "User",
"site_admin": false
} | [] | open | false | null | [] | null | [
"@mlngan can you please help me to solve this issue"
] | 2023-06-22T12:58:57 | 2023-06-22T13:19:39 | null | NONE | null | null | null | Even i ran simple cpp file with valgrind having belllow code
```
#include "Python.h"
#include "iostream"
using namespace std;
int main(){
setenv("PYTHONPATH",".",1);
Py_Initialize();
cout <<"Hello\n";
Py_Finalize();
}
```
got bellow errors on valgring.
```
==140416== by 0x49F424B: Py_InitializeEx (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x1091ED: main (in /additional_drive/nist/frvt_onnx_sequential/11/test_mem)
==140416== Address 0x5387020 is 1,520 bytes inside a block of size 2,208 free'd
==140416== at 0x483CA3F: free (in /usr/lib/x86_64-linux-gnu/valgrind/vgpreload_memcheck-amd64-linux.so)
==140416== by 0x4ACD397: ??? (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x4AC9E1F: PyDict_SetDefault (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x4A62DFF: PyUnicode_InternInPlace (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x4A62E90: PyUnicode_InternFromString (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x4AFC4C1: PyDescr_NewMethod (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x48E34AA: PyType_Ready (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x4AB8E04: _PyTypes_Init (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x49F3A6A: ??? (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x49F41A2: Py_InitializeFromConfig (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x49F424B: Py_InitializeEx (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x1091ED: main (in /additional_drive/nist/frvt_onnx_sequential/11/test_mem)
==140416== Block was alloc'd at
...
...
...
==140416== Block was alloc'd at
==140416== at 0x483B7F3: malloc (in /usr/lib/x86_64-linux-gnu/valgrind/vgpreload_memcheck-amd64-linux.so)
==140416== by 0x4AB69E7: ??? (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x4B20CB7: PyGrammar_AddAccelerators (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x4B23184: PyParser_New (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x4B26DD2: ??? (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x49E79DA: PyParser_ASTFromStringObject (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x49E8387: PyRun_StringFlags (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x4A2BADF: ??? (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x4ABB546: ??? (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x48D0D6C: ??? (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x48D246C: _PyEval_EvalFrameDefault (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416== by 0x48DC06A: ??? (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==140416==
==140416== ERROR SUMMARY: 898 errors from 107 contexts (suppressed: 0 from 0)
```
but without vargind there are no errors please suggest me do I need to resolve these errors as nist suggest and replied that I have to fix memory leakage issue
BTW I am applieng for NIST_11 challenge every thing is working fine but memory leakage is the problem and I have on google and found out that python by default does that and it own garbage collector frees memory and may that is why valgrind unable to detect that. | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/98/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/98/timeline | null | null |
https://api.github.com/repos/usnistgov/frvt/issues/97 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/97/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/97/comments | https://api.github.com/repos/usnistgov/frvt/issues/97/events | https://github.com/usnistgov/frvt/issues/97 | 1,718,812,500 | I_kwDOBIKFiM5mcv9U | 97 | Adding open source libraries for core library | {
"login": "ghost",
"id": 10137,
"node_id": "MDQ6VXNlcjEwMTM3",
"avatar_url": "https://avatars.githubusercontent.com/u/10137?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ghost",
"html_url": "https://github.com/ghost",
"followers_url": "https://api.github.com/users/ghost/followers",
"following_url": "https://api.github.com/users/ghost/following{/other_user}",
"gists_url": "https://api.github.com/users/ghost/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ghost/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ghost/subscriptions",
"organizations_url": "https://api.github.com/users/ghost/orgs",
"repos_url": "https://api.github.com/users/ghost/repos",
"events_url": "https://api.github.com/users/ghost/events{/privacy}",
"received_events_url": "https://api.github.com/users/ghost/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"You can try using rpath=$ORIGIN when you build your library, but you need to link your submission library against any 3rd party libraries that are required. For example, add the following to your CMakeLists.txt:\r\n\r\nset (CMAKE_SHARED_LINKER_FLAGS \"-Wl,-rpath,$ORIGIN\")\r\nadd_library (frvt_11_null_001 SHARED nullimplfrvt11.cpp)\r\ntarget_link_libraries(frvt_11_null_001 thirdpartylib1.so thirdpartylib2.so thirdpartylib3.so)",
"@mlngan \r\n Thank you for the above note. We are able to build the shared library using the above method and are able to run the demo code in this repo (but the catch is that we are **modifying** the **CMake build** system) both for the shared library generation and for the evaluation code. \r\n\r\nIn this case, can you let know how the final submission library would look like **libfrvt_11_null_001.so** and this is the library sent to NIST for FRVT evaluation. When we compile the library with dependecy libraries and use this library for evaluation without linking internal dep libs, the problem arises for undefined internal reference functions.\r\n\r\ni.e. **w/o** these lines in **CMAKE** for **validate_1N.cpp**, we get **undefined reference** to \"**fcn1 from lib1**\", etc. as the **compiler cannot find** these internal dep fcns. \r\n\r\n```\r\ninclude_directories($INCLUDE_DIR_LIB1, $INCLUDE_DIR_LIB2, $INCLUDE_DIR_LIB3) \r\ntarget_link_libraries($LIB1, $LIB2, $LIB3)\r\n```\r\n\r\nThe final .tar.gz file contains these folders from run_validate_1N.sh ([Link](https://github.com/usnistgov/frvt/blob/master/1N/run_validate_1N.sh), line 111) -> **doc, config, lib, validation**\r\n\r\n`tar -zcf $libstring.tar.gz ./doc ./config ./lib ./validation`\r\n\r\nPlease comment or elaborate on how the final submission would look like and if we are to add the include directories can we put it in a folder **include** and it to the above 4 folders **doc, config, lib, validation**\r\nwhile adding the $LIB1.so to **lib** folder. \r\n\r\nFRNinja123",
"You should be putting it in the lib directory (where your tested library is) for simplification as that would ensure it gets put in the same place on the test system too.\r\nIf you're linking against the adjacent library (and looking in the current directory), it will also find it there.",
"@xsacha and @mlngan \r\n Thank you for your support. We think we have the hang of it. This has more to do with how CMake Build systems works. Kindly close/delete the issue as this is solved. \r\n\r\nFRNinja123"
] | 2023-05-22T03:10:40 | 2023-05-24T07:17:07 | 2023-05-22T12:59:44 | NONE | null | null | null | Hi @kayee-nist
We have few open source libraries which we would like to add to our core library. It seems NIST only accepts 1 core library in .so format which includes 4 folders config, doc, etc
Currently We add these lines to NIST demo code in this repo and it works.
```
include_directories($INCLUDE_DIR_LIB1, $INCLUDE_DIR_LIB2, $INCLUDE_DIR_LIB3, ...)
link_libraries(/usr/local/lib/liblib1.so, /usr/local/lib/liblib2.so, /usr/local/lib/liblib3.so, ... )
```
The strict requirement has been to **NOT** modify the CMake build system and to **NOT** modify any include files. Addidng dep packages is really hard with these constraints. Hence, we are asking if a better solution is possible.
But would like to know if there is a better way to submit our proposal than to hard code these library functions (fcn1 from lib1, fcn2 from lib2, fcn3 from lib3, etc) and deps as a single libImplfrvt1N.cpp file.
FRNinja123 | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/97/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/97/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/96 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/96/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/96/comments | https://api.github.com/repos/usnistgov/frvt/issues/96/events | https://github.com/usnistgov/frvt/issues/96 | 1,698,387,170 | I_kwDOBIKFiM5lO1Ti | 96 | The twins project exist old type "Multiface" | {
"login": "yyfrzx",
"id": 18046952,
"node_id": "MDQ6VXNlcjE4MDQ2OTUy",
"avatar_url": "https://avatars.githubusercontent.com/u/18046952?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/yyfrzx",
"html_url": "https://github.com/yyfrzx",
"followers_url": "https://api.github.com/users/yyfrzx/followers",
"following_url": "https://api.github.com/users/yyfrzx/following{/other_user}",
"gists_url": "https://api.github.com/users/yyfrzx/gists{/gist_id}",
"starred_url": "https://api.github.com/users/yyfrzx/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/yyfrzx/subscriptions",
"organizations_url": "https://api.github.com/users/yyfrzx/orgs",
"repos_url": "https://api.github.com/users/yyfrzx/repos",
"events_url": "https://api.github.com/users/yyfrzx/events{/privacy}",
"received_events_url": "https://api.github.com/users/yyfrzx/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"The Multiface structure has been removed from the null implementation."
] | 2023-05-06T02:31:31 | 2023-05-08T11:42:49 | 2023-05-08T11:42:49 | NONE | null | null | null | In the twins project, the program compilation fails due to the old type "Multiface" in the API parameters, please help to solve this problem. | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/96/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/96/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/95 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/95/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/95/comments | https://api.github.com/repos/usnistgov/frvt/issues/95/events | https://github.com/usnistgov/frvt/issues/95 | 1,692,123,572 | I_kwDOBIKFiM5k28G0 | 95 | Error about frvt11 test | {
"login": "qqqtankqqq",
"id": 15229781,
"node_id": "MDQ6VXNlcjE1MjI5Nzgx",
"avatar_url": "https://avatars.githubusercontent.com/u/15229781?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/qqqtankqqq",
"html_url": "https://github.com/qqqtankqqq",
"followers_url": "https://api.github.com/users/qqqtankqqq/followers",
"following_url": "https://api.github.com/users/qqqtankqqq/following{/other_user}",
"gists_url": "https://api.github.com/users/qqqtankqqq/gists{/gist_id}",
"starred_url": "https://api.github.com/users/qqqtankqqq/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/qqqtankqqq/subscriptions",
"organizations_url": "https://api.github.com/users/qqqtankqqq/orgs",
"repos_url": "https://api.github.com/users/qqqtankqqq/repos",
"events_url": "https://api.github.com/users/qqqtankqqq/events{/privacy}",
"received_events_url": "https://api.github.com/users/qqqtankqqq/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"Apologies for this - a fixed has been pushed."
] | 2023-05-02T09:56:15 | 2023-05-03T00:46:18 | 2023-05-02T12:01:22 | NONE | null | null | null | I prepared my environment and tried to test frvt11 ,and got this error.
```
_~/frvt0501/11/src/testdriver/validate11.cpp:96:29: error: ‘using element_type = class FRVT_11::Interface’ {aka ‘class FRVT_11::Interface’} has no member named ‘createTemplate’; did you mean ‘createFaceTemplate’?
96 | auto ret = implPtr->createTemplate(faces, role, templ, eyes);
| ^~~~~~~~~~~~~~
| createFaceTemplate
```
I found ```src/nullImpl//nullimplfrvt11.h``` function name changed to ```createFaceTemplate``` and ```createIrisTemplate```
how should I solve it ?
| {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/95/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/95/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/94 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/94/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/94/comments | https://api.github.com/repos/usnistgov/frvt/issues/94/events | https://github.com/usnistgov/frvt/issues/94 | 1,660,505,168 | I_kwDOBIKFiM5i-UxQ | 94 | Questions about the latest code structure changes for the FRVT | {
"login": "yyfrzx",
"id": 18046952,
"node_id": "MDQ6VXNlcjE4MDQ2OTUy",
"avatar_url": "https://avatars.githubusercontent.com/u/18046952?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/yyfrzx",
"html_url": "https://github.com/yyfrzx",
"followers_url": "https://api.github.com/users/yyfrzx/followers",
"following_url": "https://api.github.com/users/yyfrzx/following{/other_user}",
"gists_url": "https://api.github.com/users/yyfrzx/gists{/gist_id}",
"starred_url": "https://api.github.com/users/yyfrzx/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/yyfrzx/subscriptions",
"organizations_url": "https://api.github.com/users/yyfrzx/orgs",
"repos_url": "https://api.github.com/users/yyfrzx/repos",
"events_url": "https://api.github.com/users/yyfrzx/events{/privacy}",
"received_events_url": "https://api.github.com/users/yyfrzx/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [] | 2023-04-10T09:23:51 | 2023-04-11T00:39:31 | 2023-04-11T00:39:31 | NONE | null | null | null | Hi
I have a question about the latest master code:
Is the latest master branch code structure of FRVT project changed as follows?
yzx@7750:~/frvt-master-0410/1N$ tree .
.
└── src
└── include
└── frvt1N.h
2 directories, 1 file
yzx@7750:~/frvt-master-0410/1N$
**BTW:**
On February 15th the code structure looked like this:
yzx@7750:~/frvt-master-0215/1N$ tree .
.
├── CMakeLists.txt
├── input
│ ├── enroll_1N_multiface.txt
│ ├── enroll_1N_short.txt
│ ├── enroll_1N.txt
│ ├── insert.txt
│ ├── search_1N_multiperson.txt
│ └── search_1N.txt
├── README.md
├── run_validate_1N.sh
├── scripts
│ ├── build_null_impl.sh
│ ├── compile_and_link.sh
│ └── run_testdriver.sh
└── src
├── include
│ └── frvt1N.h
├── nullImpl
│ ├── CMakeLists.txt
│ ├── nullimplfrvt1N.cpp
│ └── nullimplfrvt1N.h
└── testdriver
├── CMakeLists.txt
└── validate1N.cpp
6 directories, 18 files
yzx@7750:~/frvt-master-0215/1N$
| {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/94/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/94/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/93 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/93/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/93/comments | https://api.github.com/repos/usnistgov/frvt/issues/93/events | https://github.com/usnistgov/frvt/issues/93 | 1,643,062,129 | I_kwDOBIKFiM5h7yNx | 93 | There are two questions about whether the templateID will be duplicated in the input folder in the 1:N test | {
"login": "yyfrzx",
"id": 18046952,
"node_id": "MDQ6VXNlcjE4MDQ2OTUy",
"avatar_url": "https://avatars.githubusercontent.com/u/18046952?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/yyfrzx",
"html_url": "https://github.com/yyfrzx",
"followers_url": "https://api.github.com/users/yyfrzx/followers",
"following_url": "https://api.github.com/users/yyfrzx/following{/other_user}",
"gists_url": "https://api.github.com/users/yyfrzx/gists{/gist_id}",
"starred_url": "https://api.github.com/users/yyfrzx/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/yyfrzx/subscriptions",
"organizations_url": "https://api.github.com/users/yyfrzx/orgs",
"repos_url": "https://api.github.com/users/yyfrzx/repos",
"events_url": "https://api.github.com/users/yyfrzx/events{/privacy}",
"received_events_url": "https://api.github.com/users/yyfrzx/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"1. No. The enrollment template IDs will always be unique.\r\n2. No. Template IDs that get inserted will not coincide with what's already enrolled in the gallery. ",
"Thanks for your quick reply!"
] | 2023-03-28T01:35:47 | 2023-03-29T02:24:26 | 2023-03-28T12:55:37 | NONE | null | null | null | Hi,
We are preparing a NIST 1:N submission and are facing some issues:
1. Will the templateID of the enroll_1N_multiface.txt under the input folder be duplicated in the real test?
For example:
It currently looks like this:
1 .. /common/images/S001-01-t10_01.ppm UNKNOWN .. /common/images/S001-03-t10_01.ppm UNKNOWN .. /common/images/S001-05-t10_01.ppm UNKNOWN
21 .. /common/images/S008-02-t10_01.ppm UNKNOWN .. /common/images/S008-04-t10_01.ppm UNKNOWN .. /common/images/S008-06-t10_01.ppm
45 .. /common/images/S023-01-t10_01.ppm UNKNOWN .. /common/images/S023-03-t10_01.ppm UNKNOWN
Will this happen?:
1 .. /common/images/S001-01-t10_01.ppm UNKNOWN .. /common/images/S001-03-t10_01.ppm UNKNOWN .. /common/images/S001-05-t10_01.ppm UNKNOWN
21 .. /common/images/S008-02-t10_01.ppm UNKNOWN .. /common/images/S008-04-t10_01.ppm UNKNOWN .. /common/images/S008-06-t10_01.ppm
21 .. /common/images/S001-07-t10_01.ppm UNKNOWN .. /common/images/S001-08-t10_02.ppm UNKNOWN .. /common/images/S001-09-t10_01.ppm UNKNOWN
45 .. /common/images/S023-01-t10_01.ppm UNKNOWN .. /common/images/S023-03-t10_01.ppm UNKNOWN
2. Will the templateID in the insert.txt under the input folder coincide with the templateID in the enroll_1N.txt?
Thank you.
Looking forward to your reply. | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/93/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/93/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/92 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/92/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/92/comments | https://api.github.com/repos/usnistgov/frvt/issues/92/events | https://github.com/usnistgov/frvt/issues/92 | 1,598,805,216 | I_kwDOBIKFiM5fS9Tg | 92 | run_validate_11.sh script gets stuck on first test case | {
"login": "pawani2v",
"id": 112162847,
"node_id": "U_kgDOBq94Hw",
"avatar_url": "https://avatars.githubusercontent.com/u/112162847?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/pawani2v",
"html_url": "https://github.com/pawani2v",
"followers_url": "https://api.github.com/users/pawani2v/followers",
"following_url": "https://api.github.com/users/pawani2v/following{/other_user}",
"gists_url": "https://api.github.com/users/pawani2v/gists{/gist_id}",
"starred_url": "https://api.github.com/users/pawani2v/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/pawani2v/subscriptions",
"organizations_url": "https://api.github.com/users/pawani2v/orgs",
"repos_url": "https://api.github.com/users/pawani2v/repos",
"events_url": "https://api.github.com/users/pawani2v/events{/privacy}",
"received_events_url": "https://api.github.com/users/pawani2v/received_events",
"type": "User",
"site_admin": false
} | [] | open | false | null | [] | null | [
"Same thing, but for run_validate_1N script. Any update here?",
"@RomanShnurov No updates, i haven't been able to resolve or find what is the issue.",
"@mlngan can you please help us out.",
"Can you please try and run the provided null implementation and let us know if you encounter the same issue?",
"@mlngan \r\nI didn't encounter the problem when run the provided null implementation.\r\nBut I don't get how my code and @pawani2v 's code can affect the script if we see that our programs terminate successfully.",
"Sounds like a genuine problem in your implementation.\r\n\r\nCheck permissions of the folder.\r\nAlso make sure your app is resilient to a moved config directory, which is what that first test is.",
"@xsacha, thanks for reply!\r\nBut how permissions of the folder can help in our situation?\r\nAnd what is the configuration folder? The folder where my models(and other config things) are supposed to be stored?",
"From what I remember, the test makes the folder unreadable (permissions) and tests that you cope with this situation.\r\n\r\nThe folder you use should be the one that is provided to the API. You shouldn't hardcode the name. You should also respond with an appropriate error if you cannot access it. That's what this test is.\r\n\r\nIf your app crashed or otherwise failed due to not checking the folder, you might need to manually reset the permissions first.",
"@xsacha thanks for responding, i don't think that the config folder is a issue since i am able to load models ( i can see it but putting print in my code). Also `CreateTemplate` runs for all test methods (again can confirm since i put print in it), and as i mentioned in the original question i have put a line `std::cout << \"Exiting with status \" << exitStatus << std::endl;` in `validate11.cpp` before return and it prints but then after it their is no response.",
"@mlngan These are the error i get when i run the `provided null implementation` in virtual machine created using image provided by NIST, also it the same error i get on my machine.\r\n\r\n![image](https://user-images.githubusercontent.com/112162847/222353427-7c165705-de84-4ff3-a6ee-d0e262af247a.png)\r\n![image](https://user-images.githubusercontent.com/112162847/222353078-46fd7af6-95b2-4380-955e-5ce195d67b59.png)\r\n",
"@pawani2v It appears that you've modified the validation script/source code. It would be a good idea to run the null implementation with the original, unmodified scripts/code. The expected order of tests and output should like look the below:\r\n\r\n------------------------------\r\n Running 1:1 validation\r\n------------------------------\r\nChecking for hard-coded config directory [SUCCESS]\r\nCreating Enrollment Templates (Single Process) [SUCCESS]\r\nCreating Enrollment Templates on Multiple Images per Subject (Single Process) [SUCCESS]\r\nCreating Enrollment Templates (Multiple Processes) [SUCCESS]\r\nCreating Verification Templates (Multiple Processes) [SUCCESS]\r\nMatching Templates (Multiple Processes) [SUCCESS]\r\nCreating Verification Templates for Multiple Persons Detected in an Image (Single Process) [SUCCESS]\r\nMatching Multi-person Templates (Single Process) [SUCCESS]\r\nSanity checking validation output [SUCCESS]\r\nCreating submission package [SUCCESS]\r\n",
"I found the problem. I'm using ONNXInference and forgot to set IntraOpNumThreads to Ort::SessionOptions.\r\n\r\n`Ort::SesstionOptions options;`\r\n`options.SetIntraOpNumThreads(1);`\r\n`...`",
"@RomanShnurov did you get it running after this change ?",
"@pawani2v, yep, all tests were passed after this change."
] | 2023-02-24T15:06:00 | 2023-03-05T13:40:13 | null | NONE | null | null | null | I am prepared my shared lib and now i am running `run_validate_11.sh` but the `validate11` does not return even after it has run on all the images.
```
Checking installation of required packages [SUCCESS]
Looking for core implementation library in /home/i2v/Downloads/NIST/frvt/11/lib.[SUCCESS] Found core implementation library /home/i2v/Downloads/NIST/frvt/11/lib/libfrvt_11_i2v_001.so.
Attempting to compile and link /home/i2v/Downloads/NIST/frvt/11/lib/libfrvt_11_i2v_001.so against test harness.
Scanning dependencies of target validate11
[ 33%] Building CXX object src/testdriver/CMakeFiles/validate11.dir/home/i2v/Downloads/NIST/frvt/common/src/util/util.cpp.o
[ 66%] Building CXX object src/testdriver/CMakeFiles/validate11.dir/validate11.cpp.o
[100%] Linking CXX executable ../../../bin/validate11
[100%] Built target validate11
[SUCCESS] Built executable in /home/i2v/Downloads/NIST/frvt/11/bin.
------------------------------
Running 1:1 validation
------------------------------
Checking for hard-coded config directory
Exiting with status 0
```
How do i know `validate11` does not return control to `run_testdriver.sh` script: I put a `std::cout` in `validate11.cpp` just before return and changed the `switch` code to:
```
std::cout << "Exiting with status " << exitStatus << std::endl;
return exitStatus;
```
```
for (auto &inputFile : inputFileVector) {
/* Fork */
switch(fork()) {
case 0: /* Child */
if (action == Action::CreateTemplate)
{ exitStatus = createTemplate(
implPtr,
inputFile,
outputDir + "/" + outputFileStem + ".log." + to_string(i),
templatesDir,
role);
}
else if (action == Action::CreateMultiTemplates)
{ exitStatus = createMultiTemplates(
implPtr,
inputFile,
outputDir + "/" + outputFileStem + ".log." + to_string(i),
templatesDir,
role);
}
else if (action == Action::Match)
{ exitStatus = match(
implPtr,
inputFile,
templatesDir,
outputDir + "/" + outputFileStem + ".log." + to_string(i));
}
break;
case -1: /* Error */
cerr << "Problem forking" << endl;
break;
```
I can see the `Exiting with status 0` printed but can't see the echo that i did in `run_testdriver.sh` script below the `retEnroll=$?`
echo i put in script: `echo "retEnroll: $retEnroll"`
But if i open `htop` and kill the process i prints `[SUCCESS]` and proceeds further, but the same issue occurs in next steps, **what is this issue and how do i resolve it.**
I have run this on my development machine , and also on a virtual machine that i created using the image provided by NIST:
My Machine:
![image](https://user-images.githubusercontent.com/112162847/221212134-6996f1a7-e661-4034-ae84-c853cf872af5.png)
Virtual machine:
![image](https://user-images.githubusercontent.com/112162847/221212283-8b6bf3d9-d7f2-4fbe-8513-b79d1806bccf.png)
The issue occurs on both machines.
| {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/92/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/92/timeline | null | null |
https://api.github.com/repos/usnistgov/frvt/issues/91 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/91/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/91/comments | https://api.github.com/repos/usnistgov/frvt/issues/91/events | https://github.com/usnistgov/frvt/issues/91 | 1,574,584,461 | I_kwDOBIKFiM5d2kCN | 91 | Pad track: evasion | {
"login": "mlourencoeb",
"id": 24304201,
"node_id": "MDQ6VXNlcjI0MzA0MjAx",
"avatar_url": "https://avatars.githubusercontent.com/u/24304201?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/mlourencoeb",
"html_url": "https://github.com/mlourencoeb",
"followers_url": "https://api.github.com/users/mlourencoeb/followers",
"following_url": "https://api.github.com/users/mlourencoeb/following{/other_user}",
"gists_url": "https://api.github.com/users/mlourencoeb/gists{/gist_id}",
"starred_url": "https://api.github.com/users/mlourencoeb/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/mlourencoeb/subscriptions",
"organizations_url": "https://api.github.com/users/mlourencoeb/orgs",
"repos_url": "https://api.github.com/users/mlourencoeb/repos",
"events_url": "https://api.github.com/users/mlourencoeb/events{/privacy}",
"received_events_url": "https://api.github.com/users/mlourencoeb/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"Per the API and evaluation plan, it is not required to implement both PAD functions. For the function that you choose not to implement, please return ReturnCode::NotImplemented.",
"Thanks for the quick reply. In that case, I believe that the algorithm is not even considered for the other evaluation type, right? \r\nBest,\r\n/m",
"That is correct."
] | 2023-02-07T15:56:14 | 2023-02-07T16:11:03 | 2023-02-07T16:02:28 | NONE | null | null | null | Dear @mlngan
I am writing to ask if there is mandatory to implement evasion for the new pad challenge. If case it is optional, should we return not implemented error status to not be considered for that track?
Best,
/m | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/91/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/91/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/90 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/90/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/90/comments | https://api.github.com/repos/usnistgov/frvt/issues/90/events | https://github.com/usnistgov/frvt/issues/90 | 1,554,981,701 | I_kwDOBIKFiM5cryNF | 90 | chmod: changing permissions of 'config': Permission denied chmod: changing permissions of 'otherConfig': Permission denied | {
"login": "AwaisPF",
"id": 97581451,
"node_id": "U_kgDOBdD5iw",
"avatar_url": "https://avatars.githubusercontent.com/u/97581451?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/AwaisPF",
"html_url": "https://github.com/AwaisPF",
"followers_url": "https://api.github.com/users/AwaisPF/followers",
"following_url": "https://api.github.com/users/AwaisPF/following{/other_user}",
"gists_url": "https://api.github.com/users/AwaisPF/gists{/gist_id}",
"starred_url": "https://api.github.com/users/AwaisPF/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/AwaisPF/subscriptions",
"organizations_url": "https://api.github.com/users/AwaisPF/orgs",
"repos_url": "https://api.github.com/users/AwaisPF/repos",
"events_url": "https://api.github.com/users/AwaisPF/events{/privacy}",
"received_events_url": "https://api.github.com/users/AwaisPF/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"Here is more detail I found in Valgrind output file \r\nInvalid read of size 8\r\n==1662818== at 0x4EB1C5E: PyObject_GetAttrString (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)\r\n==1662818== by 0x484D998: FRVT_PAD::FaceImplFRVTPad::detectEvasionPA(FRVT_PAD::Media const&, bool&, double&, std::vector<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > > > >&) (in face_001/2023_01_17/lib/libfrvt_pad_face_001.so)\r\n==1662818== by 0x1140B5: runPad(std::shared_ptr<FRVT_PAD::Interface>&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&, Action) (in validation/pad/bin/validate_pad)\r\n==1662818== by 0x10F4C2: main (in validation/pad/bin/validate_pad)\r\n==1662818== Address 0x8 is not stack'd, malloc'd or (recently) free'd\r\n could some please help understand this error\r\n\r\n\r\nProcess terminating with default action of signal 11 (SIGSEGV)\r\n==1662818== Access not within mapped region at address 0x8\r\n==1662818== at 0x4EB1C5E: PyObject_GetAttrString (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)\r\n\r\n"
] | 2023-01-24T13:24:28 | 2023-02-14T04:15:08 | 2023-02-14T04:15:08 | NONE | null | null | null | After testing it on the local machine with the same environment, I submitted my package to NIST which was specifically for PAD validation but on their side, I received the error using vallgrind
detectImpersonationPA - checking for hard-coded config directory chmod: changing permissions of 'config': Permission denied
chmod: changing permissions of 'otherConfig': Permission denied
==1662817== Memcheck, a memory error detector
==1662817== Copyright (C) 2002-2017, and GNU GPL'd, by Julian Seward et al.
==1662817== Using Valgrind-3.15.0-608cb11914-20190413 and LibVEX; rerun with -h for copyright info
==1662817== Command: bin/validate_pad detectImpersonationPA -c otherconfig -o validation -h detectImpersonationPA -i input/short_pad.txt -t 1
.
.
.
.
==1662817== Invalid read of size 4
==1662817== at 0x4EA4A21: ??? (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==1662817== by 0x4EC4367: ??? (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==1662817== by 0x4EC59A8: PyDict_SetItem (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==1662817== by 0x4CD9CFE: PyType_Ready (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==1662817== by 0x4EAFDE4: _PyTypes_Init (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==1662817== by 0x4DEAAEA: ??? (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==1662817== by 0x4DEB222: Py_InitializeFromConfig (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==1662817== by 0x4DEB2CB: Py_InitializeEx (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==1662817== by 0x484D6E0: FRVT_PAD::FacedirectImplFRVTPad::initialize(std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const&) (in facedirect_001/2023_01_17/lib/libfrvt_pad_facedirect_001.so)
.
.
.
==1662817== by 0x4CC946C: _PyEval_EvalFrameDefault (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==1662817== by 0x4CD306A: ??? (in /usr/lib/x86_64-linux-gnu/libpython3.8.so.1.0)
==1662817==
==1662817== ERROR SUMMARY: 1122 errors from 108 contexts (suppressed: 0 from 0)
chmod: changing permissions of 'otherConfig': Permission denied
[ERROR] Detection of hard-coded config directory in your software. Please fix!
as far as I understand the issue is with permission related I have file on my system using root user but I did not receive this kind of error I think you are using valgrind then which kind flags are using. or if some one know exact issue with this kindly help me
thanks in advance
| {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/90/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/90/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/89 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/89/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/89/comments | https://api.github.com/repos/usnistgov/frvt/issues/89/events | https://github.com/usnistgov/frvt/issues/89 | 1,518,193,089 | I_kwDOBIKFiM5afcnB | 89 | run_validation_11 not found | {
"login": "thuanbvv",
"id": 17841056,
"node_id": "MDQ6VXNlcjE3ODQxMDU2",
"avatar_url": "https://avatars.githubusercontent.com/u/17841056?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/thuanbvv",
"html_url": "https://github.com/thuanbvv",
"followers_url": "https://api.github.com/users/thuanbvv/followers",
"following_url": "https://api.github.com/users/thuanbvv/following{/other_user}",
"gists_url": "https://api.github.com/users/thuanbvv/gists{/gist_id}",
"starred_url": "https://api.github.com/users/thuanbvv/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/thuanbvv/subscriptions",
"organizations_url": "https://api.github.com/users/thuanbvv/orgs",
"repos_url": "https://api.github.com/users/thuanbvv/repos",
"events_url": "https://api.github.com/users/thuanbvv/events{/privacy}",
"received_events_url": "https://api.github.com/users/thuanbvv/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"Humm? I'm not from the NIST but on the master branch you have the file you're looking for : https://github.com/usnistgov/frvt/blob/master/11/run_validate_11.sh\r\nBest regards",
"There was a typo in the README.md file. We've fixed the name of the script."
] | 2023-01-04T02:21:27 | 2023-01-09T12:43:34 | 2023-01-09T12:43:33 | NONE | null | null | null | Hi
I don't see any file with name run_validation_11.sh like README.md in 11.
Please review that.
Thanks, | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/89/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/89/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/88 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/88/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/88/comments | https://api.github.com/repos/usnistgov/frvt/issues/88/events | https://github.com/usnistgov/frvt/issues/88 | 1,469,509,349 | I_kwDOBIKFiM5Xlu7l | 88 | Errors when running install_packages.sh | {
"login": "4sfaloth",
"id": 8202087,
"node_id": "MDQ6VXNlcjgyMDIwODc=",
"avatar_url": "https://avatars.githubusercontent.com/u/8202087?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/4sfaloth",
"html_url": "https://github.com/4sfaloth",
"followers_url": "https://api.github.com/users/4sfaloth/followers",
"following_url": "https://api.github.com/users/4sfaloth/following{/other_user}",
"gists_url": "https://api.github.com/users/4sfaloth/gists{/gist_id}",
"starred_url": "https://api.github.com/users/4sfaloth/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/4sfaloth/subscriptions",
"organizations_url": "https://api.github.com/users/4sfaloth/orgs",
"repos_url": "https://api.github.com/users/4sfaloth/repos",
"events_url": "https://api.github.com/users/4sfaloth/events{/privacy}",
"received_events_url": "https://api.github.com/users/4sfaloth/received_events",
"type": "User",
"site_admin": false
} | [] | open | false | null | [] | null | [
"Thanks for bringing this to our attention. We will plan to include the actual packages themselves along with an updated install script to address this issue (hopefully in the next week or so).",
"[ubuntu iso copy provided at this repo is being used.](https://nigos.nist.gov/evaluations/ubuntu-20.04.3-live-server-amd64.iso)\r\n\r\nI have the same problem specifically the GCC compiler.\r\n\r\nThe documentation states that GCC version is 9.3.0\r\n\r\nwhile it is 9.4.0 after running install_packages.sh \r\n\r\nThere are multiple packages that have the same issue.\r\n\r\nShould I wait for modification? or package my library and submit with the default installed library?\r\n\r\nI appreciate your help @mlngan \r\n\r\n\r\n",
"Please go ahead and run/validate with the default installed libraries. The original approach to fixing install_packages.sh is getting a bit more complicated than we originally thought. We will keep this issue open and try to come up with a reasonable solution."
] | 2022-11-30T11:52:27 | 2023-02-14T18:05:37 | null | NONE | null | null | null | Hi,
we are preparing a NIST submission and are facing some issues with the setup of our build & validation environments. In order to make sure we are including all required files in `/lib` for dynamic linking we
- building our library and exporting the output folders (eg: `lib`, `config` and `doc`) in a `development` image
- copying these output folders only into a `validation` image and running the validation scripts there
This validation image is just a raw `focal` image where we have run
- apt update
- apt -y install sudo lsb-release
And then we run the `install_packages.sh` script. Problem is the script fails to install many of the packages seemingly due to missing versions on the package repo. For instance **_libavcodec-dev_** install fails with
`E: Version '7:4.2.4-1ubuntu0.1' for 'libavcodec-dev' was not found`
If we edit the **_install_packages.sh_** script removing the pinned versions then everything seems to work correctly, though we would no longer be sure to be running in the exact same environment as the NIST benchmark.
Is it possible that the **_install_packages.sh_** is pointing to versions no longer available on the public repo and thus needs to be updated, or am I missing something?
Thanks in advance,
Filipe Morais | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/88/reactions",
"total_count": 1,
"+1": 1,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/88/timeline | null | null |
https://api.github.com/repos/usnistgov/frvt/issues/87 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/87/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/87/comments | https://api.github.com/repos/usnistgov/frvt/issues/87/events | https://github.com/usnistgov/frvt/issues/87 | 1,459,333,589 | I_kwDOBIKFiM5W-6nV | 87 | Thread warning when using OpenVINO backend | {
"login": "ddfabbro",
"id": 17279638,
"node_id": "MDQ6VXNlcjE3Mjc5NjM4",
"avatar_url": "https://avatars.githubusercontent.com/u/17279638?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ddfabbro",
"html_url": "https://github.com/ddfabbro",
"followers_url": "https://api.github.com/users/ddfabbro/followers",
"following_url": "https://api.github.com/users/ddfabbro/following{/other_user}",
"gists_url": "https://api.github.com/users/ddfabbro/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ddfabbro/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ddfabbro/subscriptions",
"organizations_url": "https://api.github.com/users/ddfabbro/orgs",
"repos_url": "https://api.github.com/users/ddfabbro/repos",
"events_url": "https://api.github.com/users/ddfabbro/events{/privacy}",
"received_events_url": "https://api.github.com/users/ddfabbro/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"By default it will attempt to use threads so you need to tell it not to.",
"> By default it will attempt to use threads so you need to tell it not to.\r\n\r\nThanks. I will search through the OpenVINO docs to see how to tell it not use threads.\r\n\r\nWere you able to set the max number of threads for one inference without recompiling OpenVINO?",
"I recompile all the frameworks anyway but they should have a method without doing so.\r\nUsually it is based on the internal threading library they use (TBB, etc) rather than the inferencing framework itself. Being an Intel framework, I'd assume it is TBB so you can just call, off the top of my head:\r\n```C++\r\n#include <tbb/global_control.h>\r\n\r\n...\r\n\r\ntbb::global_control MAX_THREADS(tbb::global_control::max_allowed_parallelism, 1);\r\n```",
"> I recompile all the frameworks anyway but they should have a method without doing so.\r\n\r\nIt seems that they do. Going over the IE c++ api, you can specify how many threads the OpenVINO CPU Plugin (aka MKLDNN) can use for inference, among others: https://docs.openvino.ai/2021.1/openvino_docs_IE_DG_supported_plugins_CPU.html\r\n\r\nAfter reading that page until the end, you can find the suggested combination of parameters for single thread inference:\r\n\r\n> **NOTE:** To disable all internal threading, use the following set of configuration parameters: KEY_CPU_THROUGHPUT_STREAMS=0, KEY_CPU_THREADS_NUM=1, KEY_CPU_BIND_THREAD=NO.\r\n\r\nAccording to my stack tracing on the validation binary, it seems I got OpenVINO multi threads out of the way now, but I seem to have found that Boost and OpenBLAS are also creating threads at some point. Luckily, for the PAD implementation, I can remove the OpenBLAS dependency entirely, so that is solved. But I am still looking on how to force boost to run on a single thread.\r\n",
"Which part of boost is using threads for you? Normally you'd have to explicitly call some boost function that is asynchronous. There's usually a synchronous version.",
"> Which part of boost is using threads for you? Normally you'd have to explicitly call some boost function that is asynchronous. There's usually a synchronous version.\r\n\r\nOh I see now. We are explicitly creating threads with ```boost::asio::thread_pool threadPool(std::thread::hardware_concurrency());``` at model initialization.\r\n\r\n"
] | 2022-11-22T07:56:19 | 2022-11-23T07:31:07 | 2022-11-23T07:31:07 | NONE | null | null | null | My implementation is single threaded, but it will be using OpenVINO as the inference backend, which is threaded. Upon running the validation script, I received the following warning:
```
[WARNING] We've detected that your software may be threading or using other multiprocessing techniques. The number of
processes detected was 22 and it should be 2. Per the API document, implementations must run single-threaded. In the test
environment, there is no advantage to threading, because NIST will distribute workload across multiple blades and multiple
processes. We highly recommend that you fix this issue prior to submission.
```
I saw someone mentioning in the past Issues that all the main frameworks such as tensorflow, pytorch and mxnet have been known to work in NIST.
Do you also know from past submissions if OpenVINO had any issues with NIST? | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/87/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/87/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/86 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/86/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/86/comments | https://api.github.com/repos/usnistgov/frvt/issues/86/events | https://github.com/usnistgov/frvt/issues/86 | 1,459,317,734 | I_kwDOBIKFiM5W-2vm | 86 | NIST FRVT version number symbols | {
"login": "ddfabbro",
"id": 17279638,
"node_id": "MDQ6VXNlcjE3Mjc5NjM4",
"avatar_url": "https://avatars.githubusercontent.com/u/17279638?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ddfabbro",
"html_url": "https://github.com/ddfabbro",
"followers_url": "https://api.github.com/users/ddfabbro/followers",
"following_url": "https://api.github.com/users/ddfabbro/following{/other_user}",
"gists_url": "https://api.github.com/users/ddfabbro/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ddfabbro/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ddfabbro/subscriptions",
"organizations_url": "https://api.github.com/users/ddfabbro/orgs",
"repos_url": "https://api.github.com/users/ddfabbro/repos",
"events_url": "https://api.github.com/users/ddfabbro/events{/privacy}",
"received_events_url": "https://api.github.com/users/ddfabbro/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"There's nothing you have to do explicitly. Those version number symbols are defined in ./pad/src/testdriver/CMakeLists.txt\r\n\r\nset (CMAKE_CXX_FLAGS \"${CMAKE_CXX_FLAGS} -Wall -std=c++17 -DNIST_EXTERN_FRVT_STRUCTS_VERSION -DNIST_EXTERN_API_VERSION\")"
] | 2022-11-22T07:44:54 | 2022-11-28T13:04:19 | 2022-11-28T13:04:18 | NONE | null | null | null | According to `frvt_structs.h` I must define NIST FRVT version number symbols in the core library.
https://github.com/usnistgov/frvt/blob/7c2e52155467f0858cb30c9c688cb47e56142366/common/src/include/frvt_structs.h#L502-L519
What should the values be?
I am basing my implementation on the [nullimplfrvtpad.cpp](https://github.com/usnistgov/frvt/blob/master/pad/src/nullImpl/nullimplfrvtpad.cpp), but I don't see it defined in the source code. | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/86/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/86/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/85 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/85/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/85/comments | https://api.github.com/repos/usnistgov/frvt/issues/85/events | https://github.com/usnistgov/frvt/issues/85 | 1,449,359,484 | I_kwDOBIKFiM5WY3h8 | 85 | Pose Orientation at Test Time FRVT 1-1 | {
"login": "tanpengshi",
"id": 24505990,
"node_id": "MDQ6VXNlcjI0NTA1OTkw",
"avatar_url": "https://avatars.githubusercontent.com/u/24505990?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/tanpengshi",
"html_url": "https://github.com/tanpengshi",
"followers_url": "https://api.github.com/users/tanpengshi/followers",
"following_url": "https://api.github.com/users/tanpengshi/following{/other_user}",
"gists_url": "https://api.github.com/users/tanpengshi/gists{/gist_id}",
"starred_url": "https://api.github.com/users/tanpengshi/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/tanpengshi/subscriptions",
"organizations_url": "https://api.github.com/users/tanpengshi/orgs",
"repos_url": "https://api.github.com/users/tanpengshi/repos",
"events_url": "https://api.github.com/users/tanpengshi/events{/privacy}",
"received_events_url": "https://api.github.com/users/tanpengshi/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"Some of our unconstrained, non-cooperative datasets can contain images with large pose variations."
] | 2022-11-15T08:16:02 | 2022-11-16T12:40:20 | 2022-11-16T12:40:19 | NONE | null | null | null | In validation package, some images are orientated 90 degrees from the front, as such sometimes we are comparing frontal faces against side faces OR side faces against side faces.
May I know would similar facial orientation (frontal against side OR side against side) be used during testing and evaluation by NIST? | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/85/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/85/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/84 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/84/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/84/comments | https://api.github.com/repos/usnistgov/frvt/issues/84/events | https://github.com/usnistgov/frvt/issues/84 | 1,445,977,476 | I_kwDOBIKFiM5WL92E | 84 | Template Creation Failed Situation and Its Evaluation | {
"login": "ErolCitak",
"id": 10463026,
"node_id": "MDQ6VXNlcjEwNDYzMDI2",
"avatar_url": "https://avatars.githubusercontent.com/u/10463026?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ErolCitak",
"html_url": "https://github.com/ErolCitak",
"followers_url": "https://api.github.com/users/ErolCitak/followers",
"following_url": "https://api.github.com/users/ErolCitak/following{/other_user}",
"gists_url": "https://api.github.com/users/ErolCitak/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ErolCitak/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ErolCitak/subscriptions",
"organizations_url": "https://api.github.com/users/ErolCitak/orgs",
"repos_url": "https://api.github.com/users/ErolCitak/repos",
"events_url": "https://api.github.com/users/ErolCitak/events{/privacy}",
"received_events_url": "https://api.github.com/users/ErolCitak/received_events",
"type": "User",
"site_admin": false
} | [] | open | false | null | [] | null | [] | 2022-11-11T20:32:54 | 2022-11-11T20:32:54 | null | NONE | null | null | null | Hi @mlngan ,
I wonder what is the evaluation criteria if I encounter with template creation failed situation. More precisely, suppose I couldn't create a template for a specific image, then filled its vector to handle it in my matcher. And, for example, I return 0 (zero) similarity score for that template in my matcher. Then in the evaluation stage, will Nist evaluate that template for FNMR / FMR calculations? I mean, will Nist take "that failed template-based image" into consideration in the evaluation stage or not? I'm asking this question because if that template will be taken into consideration, I shouldn't give a negative score in my matcher stage. | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/84/reactions",
"total_count": 4,
"+1": 4,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/84/timeline | null | null |
https://api.github.com/repos/usnistgov/frvt/issues/83 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/83/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/83/comments | https://api.github.com/repos/usnistgov/frvt/issues/83/events | https://github.com/usnistgov/frvt/issues/83 | 1,435,803,990 | I_kwDOBIKFiM5VlKFW | 83 | Query on Final Submission - FRVT11 | {
"login": "tanpengshi",
"id": 24505990,
"node_id": "MDQ6VXNlcjI0NTA1OTkw",
"avatar_url": "https://avatars.githubusercontent.com/u/24505990?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/tanpengshi",
"html_url": "https://github.com/tanpengshi",
"followers_url": "https://api.github.com/users/tanpengshi/followers",
"following_url": "https://api.github.com/users/tanpengshi/following{/other_user}",
"gists_url": "https://api.github.com/users/tanpengshi/gists{/gist_id}",
"starred_url": "https://api.github.com/users/tanpengshi/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/tanpengshi/subscriptions",
"organizations_url": "https://api.github.com/users/tanpengshi/orgs",
"repos_url": "https://api.github.com/users/tanpengshi/repos",
"events_url": "https://api.github.com/users/tanpengshi/events{/privacy}",
"received_events_url": "https://api.github.com/users/tanpengshi/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"You just need your python libs. There is no need to include source files (includes).\r\nI recommend spin up another clean VM, extract your package there and attempt to run it. If it works, you're all sweet.",
"Pardon me, but may I also ask what files are required in the ./config folder? I did not require any files in the ./config folder in order to run my validation.\r\n\r\nAnother clarification: When I compile my codes with ./scripts/build_null_impl.sh, should my python3.8_libraries (directory) be already in the /lib folder? Because I only added it in after completion. Initially the python libraries should be called from my local directory. ",
"Well it measures the size of the config directory as your model size so it assumes your models go there. However, you could just as easily put them inside your libs.\r\n\r\nPutting your libs in after will be fine but remember to test it.",
"During test time by FRVT when I have sent my tar.gz file over, would the .so file be run automatically based on the python libraries and models already in the ./lib folder? Would there be any manual adjustments on these files to different directories, or Python installation made by NIST personnel? \r\n\r\nI am asking because I understand that in the test environment Python would not have been installed. I am not sure how would it be set up and tested, and I hope my submission works in the NIST environment.",
"Imagine they run:\r\nLD_LIBRARY_PATH=/path/to/submission/lib testApp\r\n\r\nYou should be able to test this yourself in a VM (not the one you used to build) using the script that is run during the submission file creation.\r\n\r\nIf this isn't coming across right, an easy alternative is: just create the submission on a clean VM \r\n\r\nThey mention what OS image is used in the API document. Assume a clean install of that ISO.",
"You will need to set any Python-related environment variables (e.g., PYTHONPATH, etc.) internally within your code. Consider putting your Python libraries in the configuration directory, which is passed in as a parameter to initialize(). If you want to put your python libraries in ./lib, you may assume that ./lib and ./config are at the same level (i.e., you can access ./lib via configDir/../lib.",
"I am understanding that LD_LIBRARY_PATH is for directories containing .so files, hence I should only be putting my libpython38.so etc files into the ./lib folder.\r\n\r\nAs for installed Python packages (pip install etc), I should be setting the PYTHONPATH variable. My question would be: where can I insert 'export PYTHONPATH'? Is it possible to edit `run_validate_11.sh`? But the submission package does not include `run_validate_11.sh` right?\r\n\r\nOr perhaps can I edit the codes in `nullimplfrvt11.cpp` to include:\r\n```\r\n Py_Initialize();\r\n PyRun_SimpleString(\"import sys\");\r\n PyRun_SimpleString(\"sys.path.append(\"/Path/to/Python/Package/in/config\");\r\n```\r\nThanks for your assistance!",
"You can run this code inside your library implementation.",
"I understand that Ubuntu 20.04.3 LTS comes with the Python 3.8 installed. Can I assume that NIST test environment also uses a clean Ubuntu 20.04.3 LTS with Python 3.8 installed? ",
"Yes, you may assume that Python 3.8 is installed in our test environment.",
"I need to:\r\n```\r\n`export PYTHONPATH=$(pwd)/lib/python3.8_libraries`\r\n`export TORCH_HUB=$(pwd)/lib/python3.8_libraries`\r\n```\r\n\r\nBut when I run:\r\n```\r\nNullImplFRVT11::initialize(const std::string &configDir)\r\n{\r\n Py_Initialize();\r\n PyRun_SimpleString(\"import os\");\r\n PyRun_SimpleString(\"os.system('export PYTHONPATH=$(pwd)/lib/python3.8_libraries')\");\r\n PyRun_SimpleString(\"os.system('export TORCH_HUB=$(pwd)/lib/python3.8_libraries')\");\r\n\r\n return ReturnStatus(ReturnCode::Success);\r\n}\r\n```\r\n\r\nIt does not work. What is the exact codes I can run to export PYTHONPATH and TORCH_HUB from nullimplfrvt11.cpp?\r\n\r\nThank you!\r\n",
"It's probably just due to your $(pwd) being different than you expect.\r\nThe config dir sounds like a more reasonable location, since that is passed in directly to the initialize function.\r\n\r\nWhat happens if you use:\r\n\r\n```C++\r\n#include <dlfcn.h>\r\n\r\nstd::string GetMyPath()\r\n{\r\n Dl_info dl_info;\r\n dladdr((void *)GetMyPath, &dl_info);\r\n return dl_info.dli_fname;\r\n}\r\nReturnStatus NullImplFRVT11::initialize(const std::string &configDir)\r\n{\r\n // Acquire path\r\n auto path = GetMyPath();\r\n auto pythonPath = path.substr(0, path.find_last_of('/') + 1) + \"python3.8_libraries\";\r\n // Set global vars, which will be inherited by forked processes\r\n setenv(\"PYTHONPATH\", pythonPath.c_str(), 1);\r\n setenv(\"TORCH_HUB\", pythonPath.c_str(), 1);\r\n\r\n return ReturnStatus(ReturnCode::Success);\r\n}\r\n```",
"Hi @xsacha, I have tried this and it does not work. \r\n\r\nTypically I just do:\r\n\r\n```\r\n`export PYTHONPATH=$(pwd)/lib/python3.8_libraries`\r\n`export TORCH_HUB=$(pwd)/lib/python3.8_libraries`\r\n```\r\nIn the run_validate_11.sh file.\r\n\r\n\r\nBut @mlngan said this is not allowed for submission\r\n",
"Yeah it wouldn't make sense for them to modify their environment for specific users.\r\n\r\nHow about printing out the location you get from $(pwd)/lib/python3.8_libraries to ensure it is what you expect.\r\n\r\nKeep in mind that might not work in NISTs tests because they could run it from any directory they want -- hence the configDir being provided.\r\n\r\nThe code I posted above works fine for me but I am not using Python so it's possible the Python environment has already been set up somewhere in your code with a different environment.",
"printing $(pwd)/lib/python3.8_libraries does point to frvt/11/lib/python3.8_libraries, also even from the nullimplfrvt11.cpp file.\r\n\r\nThe strange thing is that when I insert \r\n```\r\n`export PYTHONPATH=$(pwd)/lib/python3.8_libraries`\r\n`export TORCH_HUB=$(pwd)/lib/python3.8_libraries`\r\n```\r\nIn the ~/.bashrc file, the code does not work. BUT when I insert the code in the run_validate_11.sh file, it works!",
"For the bashrc file, you have to load a new bash shell for it to get that file to run and it is only set within that shell. Just like running export, it isn't a workable solution here.\r\n\r\nFor the Python code you were running:\r\n```C++\r\n PyRun_SimpleString(\"os.system('export PYTHONPATH=$(pwd)/lib/python3.8_libraries')\");\r\n```\r\nthis runs the code in a separate shell (outside of your application), so it won't affect the environment of your C++ or Python NIST process whatsoever.\r\n\r\nAs for why the code I posted didn't work: unsure. It is working on my machine :) Only thing I can think of is to make sure it is run before you run PyInitialize, in case that is preserving current env and make sure path is right.",
"By the way for your codes posted just now, are u assuming that my python3.8_libraries are in the config directory or the lib directory? More specifically,\r\n\r\n`auto pythonPath = path + \"/python3.8_libraries\";`\r\n\r\nor\r\n\r\n`auto pythonPath = path + \"/lib/python3.8_libraries\";`\r\n\r\nThank you!\r\n",
"It's probably easier if they are in config, but I am assuming they are in lib.\r\nYou can print out path to check which directory it is:\r\n`printf(\"%s\\n\", path.c_str());`",
"printf(\"%s\\n\", path.c_str()) gives me:\r\n\r\n/Documents/frvt/11/lib/libfrvt_11_COMPANY_000.so\r\n\r\nI think it should point to /Documents/frvt/11/lib instead right?",
"Yes, sorry.\r\n\r\nYou can use:\r\nauto pythonPath = path.substr(0, path.find_last_of('/') + 1) + \"python3.8_libraries\";",
"Thank you so much! <3 It worked beautifully!",
"No problems. If you put the libraries in the 'config' path instead, you can ditch that whole GetMyPath() and just put in:\r\nauto pythonPath = configDir + \"/python3.8_libraries\";\r\n\r\nThat's the suggested method from NIST and why they provide that path.",
"I realized I have been using Ubuntu-20.04.3 Desktop installed from the Ubuntu archives instead of https://nigos.nist.gov/evaluations/ubuntu-20.04.3-live-server-amd64.iso when I ran my tests, hence I ran into some issues during submissions with testing over at NIST side.\r\n\r\nDuring NIST import of openCV, i encountered this error:\r\n\r\n```\r\nTraceback (most recent call last):\r\n File autentika_000/lib/python3.8_libraries/extract_template.py\", line 1, in <module> import cv2\r\n File autentika_000/lib/python3.8_libraries/cv2/__init__.py\", line 181, in <module> bootstrap()\r\n File autentika_000/lib/python3.8_libraries/cv2/__init__.py\", line 153, in bootstrap native_module = importlib.import_module(\"cv2\")\r\n File autentika_000/lib/python3.8_libraries/importlib/__init__.py\", line 127, in import_module return _bootstrap._gcd_import(name[level:], package, level)\r\nImportError: libGL.so.1: cannot open shared object file: No such file or directory\r\n```\r\n\r\nThe libGL.so.1 is located in the /usr/lib/x86_64-linux-gnu directory in my Ubuntu Desktop.\r\n\r\nHow can I move it to the /lib/python3.8_libraries directory, and have the import of libGL.so.1 detected? Does this do the trick:\r\n\r\n`setenv(\"PATH\", pythonPath.c_str(), 1);`\r\n\r\n\r\n",
"You want \"LD_LIBRARY_PATH\"\r\n\r\nI guess this is a pre-compiled OpenCV? You would want an OpenCV compiled without OpenGL or image loading normally. You won't be using any of that.",
"Basically my python OpenCV libraries are all located in /lib/python3.8_libraries directory. Apparently i need libGL.so.1 to import python OpenCV",
"setenv( \"LD_LIBRARY_PATH\", pythonPath.c_str(), 1); did not work\r\n\r\nThe system is still looking for the libGL.so.1 file in the /usr/lib/x86_64-linux-gnu directory",
"I have also tried to edit the CMakeLists.txt\r\n\r\n```\r\n# Build the shared libraries\r\nset (CMAKE_SHARED_LINKER_FLAGS \"-Wl,-rpath,$ORIGIN\")\r\nadd_library (frvt_11_autentika_000 SHARED nullimplfrvt11.cpp)\r\ntarget_link_libraries (frvt_11_autentika_000 libpython3.8.so libGL.so.1)\r\n```\r\n\r\nBut i got the error during compilation:\r\n\r\n```\r\n/usr/bin/ld: cannot find -lGL\r\ncollect2: error: ld returned 1 exit status\r\n```\r\n\r\nHow can I linked the shared object files in the /lib directory? ",
"Are you sure it's just libGL it's looking for? I just installed opencv-python package and noticed it has a *massive* amount of dependent libraries I can see in: site-packages/opencv_python.libs/. It has been compiled to include absolutely everything including GUI (which provided Qt on mine).\r\n\r\nIf the dependency comes from there, it should probably be enough to stick all of the missing ones in there along with them.\r\n\r\nOn mine, the dependencies on a single Qt lib provided with opencv-python:\r\n\r\n> ~/.local/lib/python3.10/site-packages/opencv_python.libs/libQt5XcbQpa-c112ba75.so.5.15.0:\r\n> linux-vdso.so.1 (0x00007ffdefd2e000)\r\n> libpng16-57e5e0a0.so.16.37.0 => not found\r\n> libz.so.1 => /lib/x86_64-linux-gnu/libz.so.1 (0x00007f34e7b32000)\r\n> libGL.so.1 => /lib/x86_64-linux-gnu/libGL.so.1 (0x00007f34e7aab000)\r\n> libQt5Gui-48e93776.so.5.15.0 => not found\r\n> libQt5Core-39545cc7.so.5.15.0 => not found\r\n> libpthread.so.0 => /lib/x86_64-linux-gnu/libpthread.so.0 (0x00007f34e7aa4000)\r\n> libX11-xcb-69166bdf.so.1.0.0 => not found\r\n> libxcb-icccm-413c9f41.so.4.0.0 => not found\r\n> libxcb-image-e82a276d.so.0.0.0 => not found\r\n> libxcb-shm-7a199f70.so.0.0.0 => not found\r\n> libxcb-keysyms-21015570.so.1.0.0 => not found\r\n> libxcb-randr-a96a5a87.so.0.1.0 => not found\r\n> libxcb-render-util-43ce00f5.so.0.0.0 => not found\r\n> libxcb-render-637b984a.so.0.0.0 => not found\r\n> libxcb-shape-25c2b258.so.0.0.0 => not found\r\n> libxcb-sync-89374f40.so.1.0.0 => not found\r\n> libxcb-xfixes-9be3ba6f.so.0.0.0 => not found\r\n> libxcb-xinerama-ae147f87.so.0.0.0 => not found\r\n> libxcb-xkb-9ba31ab3.so.1.0.0 => not found\r\n> libxcb.so.1 => /lib/x86_64-linux-gnu/libxcb.so.1 (0x00007f34e7a76000)\r\n> libXext.so.6 => /lib/x86_64-linux-gnu/libXext.so.6 (0x00007f34e7a5f000)\r\n> libX11.so.6 => /lib/x86_64-linux-gnu/libX11.so.6 (0x00007f34e791f000)\r\n> libSM.so.6 => /lib/x86_64-linux-gnu/libSM.so.6 (0x00007f34e7914000)\r\n> libICE.so.6 => /lib/x86_64-linux-gnu/libICE.so.6 (0x00007f34e78f7000)\r\n> libxkbcommon-x11-c65ed502.so.0.0.0 => not found\r\n> libxkbcommon-71ae2972.so.0.0.0 => not found\r\n> ...\r\n\r\nI realise this is very painful going through Python. Tip to anyone else trying this: just use C++ :)",
"Instead of Python OpenCV, do you think I can use skimage or PIL to process images?",
"Sure, I recommend PIL if that works for you. It won't draw in the ~100 .so files that OpenCV does.\r\n\r\nSince you aren't opening images or converting them (provided for you as bytes), you could probably get away with just numpy."
] | 2022-11-04T09:43:10 | 2022-12-05T03:26:49 | 2022-11-15T09:32:11 | NONE | null | null | null | I have successfully generated the tar.gz file for submission, and manual analysis of the validation package shows above 90% F1, so I guess I am all ready for submission for FRVT11 :)
My approach to FRVT11 is to embed my python codes and model in C++, hence in my my CMakeList in ./src/nullImpl directory, I have called:
```
find_package(PythonLibs REQUIRED)
target_link_libraries (frvt_11_COMPANY_000 ${PYTHON_LIBRARIES})
```
plus my include directories and so on. In addition, I have to put my tensorflow model file in ./frvt/11 main directory in order to run run_validate_11.sh
My question is this: Can I put all these python packages, include directories and python model files in the ./frvt/11/lib folder?
For instance in the lib folder, I have:
- libfrvt_11_COMPANY_000.so (file)
- python3.8_include (directory)
- python3.8_libraries (directory)
- tensorflow_model (directory)
Appreciate the assistance, and thanks in advance!
| {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/83/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/83/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/82 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/82/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/82/comments | https://api.github.com/repos/usnistgov/frvt/issues/82/events | https://github.com/usnistgov/frvt/issues/82 | 1,432,875,980 | I_kwDOBIKFiM5VZ_PM | 82 | Distribution of actual matches during validation | {
"login": "tanpengshi",
"id": 24505990,
"node_id": "MDQ6VXNlcjI0NTA1OTkw",
"avatar_url": "https://avatars.githubusercontent.com/u/24505990?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/tanpengshi",
"html_url": "https://github.com/tanpengshi",
"followers_url": "https://api.github.com/users/tanpengshi/followers",
"following_url": "https://api.github.com/users/tanpengshi/following{/other_user}",
"gists_url": "https://api.github.com/users/tanpengshi/gists{/gist_id}",
"starred_url": "https://api.github.com/users/tanpengshi/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/tanpengshi/subscriptions",
"organizations_url": "https://api.github.com/users/tanpengshi/orgs",
"repos_url": "https://api.github.com/users/tanpengshi/repos",
"events_url": "https://api.github.com/users/tanpengshi/events{/privacy}",
"received_events_url": "https://api.github.com/users/tanpengshi/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"This is heavily implementation dependent.\r\nIn the validation package, some of the pairs are incorrect (you can tell because the IDs are different). A match is expected when the ID is the same.\r\n\r\nHere's an example output with 'distance' (opposite of similarity) vs expected match for a solution that gets 100% in validation\r\n[v8.txt](https://github.com/usnistgov/frvt/files/9920710/v8.txt)\r\n"
] | 2022-11-02T10:17:10 | 2022-11-04T02:09:51 | 2022-11-04T02:09:51 | NONE | null | null | null | Hello, I have successfully ran the validation, and it seems that vast majority of matching has very high similarity scores, except for only 1. May I know if that is the norm for the validation package? Are these comparisons suppose to be matches? What should be the distribution of matches in the validation?
Thank you! | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/82/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/82/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/81 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/81/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/81/comments | https://api.github.com/repos/usnistgov/frvt/issues/81/events | https://github.com/usnistgov/frvt/issues/81 | 1,432,534,459 | I_kwDOBIKFiM5VYr27 | 81 | how to pass an array of array from python to C++ (multiple faces in an image) | {
"login": "tanpengshi",
"id": 24505990,
"node_id": "MDQ6VXNlcjI0NTA1OTkw",
"avatar_url": "https://avatars.githubusercontent.com/u/24505990?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/tanpengshi",
"html_url": "https://github.com/tanpengshi",
"followers_url": "https://api.github.com/users/tanpengshi/followers",
"following_url": "https://api.github.com/users/tanpengshi/following{/other_user}",
"gists_url": "https://api.github.com/users/tanpengshi/gists{/gist_id}",
"starred_url": "https://api.github.com/users/tanpengshi/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/tanpengshi/subscriptions",
"organizations_url": "https://api.github.com/users/tanpengshi/orgs",
"repos_url": "https://api.github.com/users/tanpengshi/repos",
"events_url": "https://api.github.com/users/tanpengshi/events{/privacy}",
"received_events_url": "https://api.github.com/users/tanpengshi/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"If you dereference the pointer and there's nothing there, it'll crash. That appears to be the case.\r\nFor example, if c_out[0] is nullptr (0), what do you expect?\r\nI see you print something else and then continue to the crash.\r\n\r\nI'm not sure what the Py* functions are doing but it appears like you have a float* rather than float** and are trying to cast it incorrectly.\r\n\r\nThese questions do not belong here. They aren't related to FRVT."
] | 2022-11-02T05:43:51 | 2022-11-02T06:47:29 | 2022-11-02T06:47:29 | NONE | null | null | null | I am doing template generation for multiple faces in image:
I am returning an array of array (array of templates) from python to c++. Here is a snippet of my codes:
```
npy_intp dims[3] = {height,width,channel};
np_arg = reinterpret_cast<PyArrayObject*>(PyArray_SimpleNewFromData(3,dims,NPY_UINT8,image.data.get()));
pArgs = PyTuple_New(1);
PyTuple_SetItem(pArgs, 0, reinterpret_cast<PyObject*>(np_arg));
pValue = PyObject_CallObject(pFunc, pArgs);
if (pValue==nullptr){
PyErr_Print();
}
np_ret = reinterpret_cast<PyArrayObject*>(pValue);
c_out = reinterpret_cast<float**>(PyArray_DATA(np_ret));
std::cout << c_out[0][0] << std::endl;
```
During runtime, i encounter segmentation error at std::cout << c_out[0][0] << std::endl; (PID 8132 exited due to signal 11)
std::cout << c_out[0] << std::endl; runs fine though.
What did I do wrong? | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/81/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/81/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/80 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/80/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/80/comments | https://api.github.com/repos/usnistgov/frvt/issues/80/events | https://github.com/usnistgov/frvt/issues/80 | 1,431,482,264 | I_kwDOBIKFiM5VUq-Y | 80 | Can you provide minimal docker image? | {
"login": "shubhamwagh",
"id": 22571244,
"node_id": "MDQ6VXNlcjIyNTcxMjQ0",
"avatar_url": "https://avatars.githubusercontent.com/u/22571244?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/shubhamwagh",
"html_url": "https://github.com/shubhamwagh",
"followers_url": "https://api.github.com/users/shubhamwagh/followers",
"following_url": "https://api.github.com/users/shubhamwagh/following{/other_user}",
"gists_url": "https://api.github.com/users/shubhamwagh/gists{/gist_id}",
"starred_url": "https://api.github.com/users/shubhamwagh/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/shubhamwagh/subscriptions",
"organizations_url": "https://api.github.com/users/shubhamwagh/orgs",
"repos_url": "https://api.github.com/users/shubhamwagh/repos",
"events_url": "https://api.github.com/users/shubhamwagh/events{/privacy}",
"received_events_url": "https://api.github.com/users/shubhamwagh/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"The base image seems to be the same you get from the official ubuntu repo for 20.04.3, so I think it should be okay to use a corresponding base dockerfile such as `ubuntu:focal-20220113`. Still would be preferable to have an official response from NIST for sure 🤞",
"Per the our published documentation, the operating system that the submitted implementations shall run on is released as a downloadable file accessible from https://nigos.nist.gov/evaluations/ubuntu-20.04.3-live-server-amd64.iso, which indeed is the 64-bit version of Ubuntu 20.04.3 LTS (Focal Fossa).",
"This is an older issue but I think it is still relevant. Using an iso file is much less convenient than using a docker image in CI/CD systems like jenkins, that are frequently used to automatically build and test releases. \r\nGiven that it the OS requirement is quite specific and not the most recent LTS release of Ubuntu 20, it is quite hard to find something that works. ",
"I use [this](https://hub.docker.com/r/swathip2021/myubuntu) image in the first line of my Dockerfile.\r\n```\r\nFROM swathip2021/myubuntu:20.04 AS ubuntu\r\n```"
] | 2022-11-01T14:01:33 | 2024-05-15T08:00:24 | 2022-11-21T18:26:06 | NONE | null | null | null | Hi, it will be great if you can provide minimal docker image for running validation scripts. Thanks. | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/80/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/80/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/79 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/79/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/79/comments | https://api.github.com/repos/usnistgov/frvt/issues/79/events | https://github.com/usnistgov/frvt/issues/79 | 1,430,853,900 | I_kwDOBIKFiM5VSRkM | 79 | Necessary to output eye coordinates? | {
"login": "tanpengshi",
"id": 24505990,
"node_id": "MDQ6VXNlcjI0NTA1OTkw",
"avatar_url": "https://avatars.githubusercontent.com/u/24505990?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/tanpengshi",
"html_url": "https://github.com/tanpengshi",
"followers_url": "https://api.github.com/users/tanpengshi/followers",
"following_url": "https://api.github.com/users/tanpengshi/following{/other_user}",
"gists_url": "https://api.github.com/users/tanpengshi/gists{/gist_id}",
"starred_url": "https://api.github.com/users/tanpengshi/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/tanpengshi/subscriptions",
"organizations_url": "https://api.github.com/users/tanpengshi/orgs",
"repos_url": "https://api.github.com/users/tanpengshi/repos",
"events_url": "https://api.github.com/users/tanpengshi/events{/privacy}",
"received_events_url": "https://api.github.com/users/tanpengshi/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"Yes, you don't have to populate the eye coordinates.",
"May I also ask: I understand that during test, images contain one and only one face. But in the codes (for multiple faces in an image) in a given in nullImplfrvt11.cpp:\r\n\r\n```\r\nNullImplFRVT11::createTemplate(\r\n const FRVT::Image &image,\r\n FRVT::TemplateRole role,\r\n std::vector<std::vector<uint8_t>> &templs,\r\n std::vector<FRVT::EyePair> &eyeCoordinates)\r\n{\r\n int numFaces = rand() % 4 + 1;\r\n for (int i = 1; i <= numFaces; i++) {\r\n std::vector<uint8_t> templ;\r\n /* Note: example code, potentially not portable across machines. */\r\n std::vector<float> fv = {1.0, 2.0, 8.88, 765.88989};\r\n /* Multiply vector values by scalar */\r\n for_each(fv.begin(), fv.end(), [i](float &f){ f *= i; });\r\n const uint8_t* bytes = reinterpret_cast<const uint8_t*>(fv.data());\r\n int dataSize = sizeof(float) * fv.size();\r\n templ.resize(dataSize);\r\n memcpy(templ.data(), bytes, dataSize);\r\n templs.push_back(templ);\r\n\r\n eyeCoordinates.push_back(EyePair(true, true, i, i, i+1, i+1));\r\n } \r\n \r\n return ReturnStatus(ReturnCode::Success);\r\n}\r\n```\r\nIs it necessary to return multiple templates for a given image of more than one faces? Can I not populate this? Or perhaps return only the template of the largest face?\r\n\r\nAppreciate the clarification :)\r\n\r\n\r\n",
"@tanpengshi You should read the requirements here https://pages.nist.gov/frvt/api/FRVT_ongoing_11_api.pdf. \r\nThere are 2 templates: the first is a template with input Multiface (each image contains only a face and the same face of a person) You can return the largest face if your detector detects the number of faces > 1. The second is a template with input only an image (can contain one or more faces), you must return all faces for that.",
"Can we also assume images at test time are either greyscale (1 channel) or RGB channels?",
"Images used in the evaluation will either be 8-bit grayscale (1 channel) or 24-bit (3 channels) color."
] | 2022-11-01T04:31:20 | 2022-11-02T13:09:34 | 2022-11-01T11:02:16 | NONE | null | null | null | I am working on FRVT 1:1, and my query is : is it necessary to output eye coordinates? I mean the goal is to compare templates between template for enrollment and verification.
Can I leave the eye coordinates as zero?
| {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/79/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/79/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/78 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/78/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/78/comments | https://api.github.com/repos/usnistgov/frvt/issues/78/events | https://github.com/usnistgov/frvt/issues/78 | 1,429,486,673 | I_kwDOBIKFiM5VNDxR | 78 | Query on time limit during testing for 1-1 | {
"login": "tanpengshi",
"id": 24505990,
"node_id": "MDQ6VXNlcjI0NTA1OTkw",
"avatar_url": "https://avatars.githubusercontent.com/u/24505990?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/tanpengshi",
"html_url": "https://github.com/tanpengshi",
"followers_url": "https://api.github.com/users/tanpengshi/followers",
"following_url": "https://api.github.com/users/tanpengshi/following{/other_user}",
"gists_url": "https://api.github.com/users/tanpengshi/gists{/gist_id}",
"starred_url": "https://api.github.com/users/tanpengshi/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/tanpengshi/subscriptions",
"organizations_url": "https://api.github.com/users/tanpengshi/orgs",
"repos_url": "https://api.github.com/users/tanpengshi/repos",
"events_url": "https://api.github.com/users/tanpengshi/events{/privacy}",
"received_events_url": "https://api.github.com/users/tanpengshi/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"It is 1.5 second per feature generation (detection + recognition) on a single core of the CPUs specified in the API document.\r\n\r\nIn your Python example, it is running multi-threaded and looks like you are only doing face recognition on a 112x112 image, which is a far easier task."
] | 2022-10-31T08:53:44 | 2022-11-01T04:29:18 | 2022-11-01T04:29:18 | NONE | null | null | null | I am embedding my Python codes in C++.
My concern is regarding the time taken for each image inference, because I understand the time limit for image enrollment is 1500ms. Using this code above i take about 2.5seconds to 3 seconds (on Virtual Machine CPU) to iterate between each std::cout during runtime, as shown:
![image](https://user-images.githubusercontent.com/24505990/198968574-729f0fd2-7a62-49cc-99b2-10f1eba88921.png)
However, during testing with Python on my CPU work station, each template generation only takes about 190ms, as shown:
![image](https://user-images.githubusercontent.com/24505990/198969045-7465677d-b70b-4ecb-88a6-771f37c03dcc.png)
My question is how are the codes being tested for timings, and would my codes potentially exceed the time limit? Appreciate the help!
| {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/78/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/78/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/77 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/77/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/77/comments | https://api.github.com/repos/usnistgov/frvt/issues/77/events | https://github.com/usnistgov/frvt/issues/77 | 1,417,694,705 | I_kwDOBIKFiM5UgE3x | 77 | [Error] Enrollment validation (multi person in image) failed - when no face detected | {
"login": "dzeus14omega",
"id": 43383923,
"node_id": "MDQ6VXNlcjQzMzgzOTIz",
"avatar_url": "https://avatars.githubusercontent.com/u/43383923?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/dzeus14omega",
"html_url": "https://github.com/dzeus14omega",
"followers_url": "https://api.github.com/users/dzeus14omega/followers",
"following_url": "https://api.github.com/users/dzeus14omega/following{/other_user}",
"gists_url": "https://api.github.com/users/dzeus14omega/gists{/gist_id}",
"starred_url": "https://api.github.com/users/dzeus14omega/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/dzeus14omega/subscriptions",
"organizations_url": "https://api.github.com/users/dzeus14omega/orgs",
"repos_url": "https://api.github.com/users/dzeus14omega/repos",
"events_url": "https://api.github.com/users/dzeus14omega/events{/privacy}",
"received_events_url": "https://api.github.com/users/dzeus14omega/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"Per the [FRVT 1:1 API](https://pages.nist.gov/frvt/api/FRVT_ongoing_11_api.pdf), if the implementation is unable to extract features, the output shall still contain a single template that may be passed to the matchTemplates() function without error. That is, this routine must internally encode \"template creation failed\", and the matcher must transparently handle this. A template may have length zero bytes as long as your matcher handles it."
] | 2022-10-21T04:10:02 | 2022-10-21T13:20:20 | 2022-10-21T13:20:20 | NONE | null | null | null | on function createMultiTemplates() in validate11.cpp, i found that it requires "the output template vector must contain at least one template"; but in some situations, model face-detector cannot detect anything for example, we don't return any templates and that cause this error.
So can anyone help me what i should return at this case or any solution to solve this problem? | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/77/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/77/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/76 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/76/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/76/comments | https://api.github.com/repos/usnistgov/frvt/issues/76/events | https://github.com/usnistgov/frvt/issues/76 | 1,400,072,817 | I_kwDOBIKFiM5Tc2px | 76 | Error Detection of hard-coded config directory when run ./run_validate_11.sh | {
"login": "jatsi",
"id": 33884168,
"node_id": "MDQ6VXNlcjMzODg0MTY4",
"avatar_url": "https://avatars.githubusercontent.com/u/33884168?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/jatsi",
"html_url": "https://github.com/jatsi",
"followers_url": "https://api.github.com/users/jatsi/followers",
"following_url": "https://api.github.com/users/jatsi/following{/other_user}",
"gists_url": "https://api.github.com/users/jatsi/gists{/gist_id}",
"starred_url": "https://api.github.com/users/jatsi/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/jatsi/subscriptions",
"organizations_url": "https://api.github.com/users/jatsi/orgs",
"repos_url": "https://api.github.com/users/jatsi/repos",
"events_url": "https://api.github.com/users/jatsi/events{/privacy}",
"received_events_url": "https://api.github.com/users/jatsi/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"Your program threw a Segmentation Fault error (signal 11). The `run_validate_11.sh` script just checks whether the programs return a successful exit code 0.",
"thanks",
"@jatsi did you solve it?"
] | 2022-10-06T17:42:28 | 2023-09-20T19:15:27 | 2022-10-11T02:15:13 | NONE | null | null | null | Running ./run_validate_11.sh shows me the following error:
**Checking installation of required packages [SUCCESS]
Looking for core implementation library in /home/daksha/Documentos/frvt/11/lib.[SUCCESS] Found core implementation library /home/daksha/Documentos/frvt/11/lib/libfrvt_11_facedaksha_000.so.
Attempting to compile and link /home/daksha/Documentos/frvt/11/lib/libfrvt_11_facedaksha_000.so against test harness.
Scanning dependencies of target validate11
[ 33%] Building CXX object src/testdriver/CMakeFiles/validate11.dir/home/daksha/Documentos/frvt/common/src/util/util.cpp.o
[ 66%] Building CXX object src/testdriver/CMakeFiles/validate11.dir/validate11.cpp.o
[100%] Linking CXX executable ../../../bin/validate11
[100%] Built target validate11
[SUCCESS] Built executable in /home/daksha/Documentos/frvt/11/bin.
------------------------------
Running 1:1 validation
------------------------------**
Checking for hard-coded config directory PID 52211 exited due to signal 11
[ERROR] Detection of hard-coded config directory in your software. Please fix!
I do not hard-coded config in source code, create config folder for my models. | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/76/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/76/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/75 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/75/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/75/comments | https://api.github.com/repos/usnistgov/frvt/issues/75/events | https://github.com/usnistgov/frvt/issues/75 | 1,341,178,773 | I_kwDOBIKFiM5P8MOV | 75 | thread count and mxnet forks | {
"login": "avinashkshirsagar",
"id": 25266500,
"node_id": "MDQ6VXNlcjI1MjY2NTAw",
"avatar_url": "https://avatars.githubusercontent.com/u/25266500?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/avinashkshirsagar",
"html_url": "https://github.com/avinashkshirsagar",
"followers_url": "https://api.github.com/users/avinashkshirsagar/followers",
"following_url": "https://api.github.com/users/avinashkshirsagar/following{/other_user}",
"gists_url": "https://api.github.com/users/avinashkshirsagar/gists{/gist_id}",
"starred_url": "https://api.github.com/users/avinashkshirsagar/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/avinashkshirsagar/subscriptions",
"organizations_url": "https://api.github.com/users/avinashkshirsagar/orgs",
"repos_url": "https://api.github.com/users/avinashkshirsagar/repos",
"events_url": "https://api.github.com/users/avinashkshirsagar/events{/privacy}",
"received_events_url": "https://api.github.com/users/avinashkshirsagar/received_events",
"type": "User",
"site_admin": false
} | [] | open | false | null | [] | null | [] | 2022-08-17T05:15:42 | 2022-08-18T05:59:24 | null | NONE | null | null | null | We are using mxnet for our algorithm, we are able to reduce the number of threads to 5 with only 1 running thread, however to remove display messages we have built the library from source by commenting the logs where we are facing issues with non returning forks. The code executes for the images involved in one phase for example enrolment and then it waits for the process to return.
Can anyone help me to resolve it?
| {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/75/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/75/timeline | null | null |
https://api.github.com/repos/usnistgov/frvt/issues/74 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/74/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/74/comments | https://api.github.com/repos/usnistgov/frvt/issues/74/events | https://github.com/usnistgov/frvt/issues/74 | 1,189,975,620 | I_kwDOBIKFiM5G7ZZE | 74 | Query regarding the Report on demographic groups | {
"login": "aravinthmuthu",
"id": 17089804,
"node_id": "MDQ6VXNlcjE3MDg5ODA0",
"avatar_url": "https://avatars.githubusercontent.com/u/17089804?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/aravinthmuthu",
"html_url": "https://github.com/aravinthmuthu",
"followers_url": "https://api.github.com/users/aravinthmuthu/followers",
"following_url": "https://api.github.com/users/aravinthmuthu/following{/other_user}",
"gists_url": "https://api.github.com/users/aravinthmuthu/gists{/gist_id}",
"starred_url": "https://api.github.com/users/aravinthmuthu/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/aravinthmuthu/subscriptions",
"organizations_url": "https://api.github.com/users/aravinthmuthu/orgs",
"repos_url": "https://api.github.com/users/aravinthmuthu/repos",
"events_url": "https://api.github.com/users/aravinthmuthu/events{/privacy}",
"received_events_url": "https://api.github.com/users/aravinthmuthu/received_events",
"type": "User",
"site_admin": false
} | [] | open | false | null | [] | null | [] | 2022-04-01T15:41:04 | 2022-04-01T15:41:04 | null | NONE | null | null | null | Hello
We have a few queries regarding the upcoming "Report on how error rate variations across demographic groups."
1. Will this report be based only on the algorithms submitted to FRVT 1:1 leaderboard, or will this report also contain an analysis of algorithms submitted to the FRVT 1:N leaderboard?
2. Will the new algorithms submitted to FRVT 1:1 and FRVT 1:N leaderboards next week also be a part of this demographics report?
3. What will be the frequency of this report? Will this be released monthly like the FRVT 1:1 and FRVT 1:N reports?
Thank you! | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/74/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/74/timeline | null | null |
https://api.github.com/repos/usnistgov/frvt/issues/73 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/73/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/73/comments | https://api.github.com/repos/usnistgov/frvt/issues/73/events | https://github.com/usnistgov/frvt/issues/73 | 1,188,016,117 | I_kwDOBIKFiM5Gz6_1 | 73 | Query on the updated protocol that supports the existence of multiple faces detected in an image | {
"login": "manideep2510",
"id": 25865501,
"node_id": "MDQ6VXNlcjI1ODY1NTAx",
"avatar_url": "https://avatars.githubusercontent.com/u/25865501?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/manideep2510",
"html_url": "https://github.com/manideep2510",
"followers_url": "https://api.github.com/users/manideep2510/followers",
"following_url": "https://api.github.com/users/manideep2510/following{/other_user}",
"gists_url": "https://api.github.com/users/manideep2510/gists{/gist_id}",
"starred_url": "https://api.github.com/users/manideep2510/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/manideep2510/subscriptions",
"organizations_url": "https://api.github.com/users/manideep2510/orgs",
"repos_url": "https://api.github.com/users/manideep2510/repos",
"events_url": "https://api.github.com/users/manideep2510/events{/privacy}",
"received_events_url": "https://api.github.com/users/manideep2510/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"In the updated protocol that supports multiple faces detected in an image, we take the maximum similarity score across all template comparisons between each enrollment and verification image pair. The high ranks (poor FNMR) are due to higher FMR under the max(score) policy. The upside to this approach is of reduced FNMR – this can occur because in a genuine comparison of say an image with 1 face present with another image with 2 faces present, detection of the correct mated person was not guaranteed before we changed the API. With the new API, one of the two faces should score highly. We take largest score over those produced in comparing K1 templates from image 1 with K2 templates from image 2, i.e. max of K1 * K2 scores. We do this for mated **and** non-mated comparisons."
] | 2022-03-31T12:37:04 | 2022-03-31T13:04:42 | 2022-03-31T13:04:41 | NONE | null | null | null | Hello
In the recently released [FRVT 1:1 leaderboard](https://pages.nist.gov/frvt/html/frvt11.html) along with benchmark numbers (in blue) with the updated protocol that supports the existence of multiple faces detected in an image, the ranks and the FNMR numbers (in blue) on the `BORDER Photos` and `KIOSK Photos` benchmark sets are extremely large compared to other benchmarks where this updated protocol isn't used even for the algorithms which are very well ranked on the other benchmark sets.
Could you kindly give us more clarity on how these FNMR values reported in blue on the leaderboard should be read?
And are these high FNMR rates in blue with updated protocol due to inaccuracies in the submitted algorithms, or should we take into account anything else from the package and benchmark data point of view?
I really appreciate any help you can provide. Thank you. | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/73/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/73/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/72 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/72/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/72/comments | https://api.github.com/repos/usnistgov/frvt/issues/72/events | https://github.com/usnistgov/frvt/issues/72 | 1,169,509,757 | I_kwDOBIKFiM5FtU19 | 72 | Errors while compiling validation package 1:1 | {
"login": "mahdisomjee123",
"id": 90239736,
"node_id": "MDQ6VXNlcjkwMjM5NzM2",
"avatar_url": "https://avatars.githubusercontent.com/u/90239736?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/mahdisomjee123",
"html_url": "https://github.com/mahdisomjee123",
"followers_url": "https://api.github.com/users/mahdisomjee123/followers",
"following_url": "https://api.github.com/users/mahdisomjee123/following{/other_user}",
"gists_url": "https://api.github.com/users/mahdisomjee123/gists{/gist_id}",
"starred_url": "https://api.github.com/users/mahdisomjee123/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/mahdisomjee123/subscriptions",
"organizations_url": "https://api.github.com/users/mahdisomjee123/orgs",
"repos_url": "https://api.github.com/users/mahdisomjee123/repos",
"events_url": "https://api.github.com/users/mahdisomjee123/events{/privacy}",
"received_events_url": "https://api.github.com/users/mahdisomjee123/received_events",
"type": "User",
"site_admin": false
} | [] | open | false | null | [] | null | [
"As a sanity check, do you get the same error when you build the null implementation by running ./scripts/build_null_impl.sh and then running ./run_validate_11.sh?",
"I am facing same problem. \r\n/usr/bin/ld: CMakeFiles/validate_morph.dir/validate_morph.o: in function `main':\r\nvalidate_morph.cpp:(.text+0xf9f): undefined reference to `FRVT::FRVT_STRUCTS_MAJOR_VERSION'\r\n/usr/bin/ld: validate_morph.cpp:(.text+0xfaf): undefined reference to `FRVT::FRVT_STRUCTS_MINOR_VERSION'\r\n/usr/bin/ld: validate_morph.cpp:(.text+0xfd9): undefined reference to `FRVT::FRVT_STRUCTS_MAJOR_VERSION'\r\n/usr/bin/ld: validate_morph.cpp:(.text+0xfff): undefined reference to `FRVT::FRVT_STRUCTS_MINOR_VERSION'\r\n/usr/bin/ld: validate_morph.cpp:(.text+0x1087): undefined reference to `FRVT_MORPH::API_MAJOR_VERSION'\r\n/usr/bin/ld: validate_morph.cpp:(.text+0x1097): undefined reference to `FRVT_MORPH::API_MINOR_VERSION'\r\n/usr/bin/ld: validate_morph.cpp:(.text+0x10c1): undefined reference to `FRVT_MORPH::API_MAJOR_VERSION'\r\n/usr/bin/ld: validate_morph.cpp:(.text+0x10e7): undefined reference to `FRVT_MORPH::API_MINOR_VERSION'\r\n/usr/bin/ld: validate_morph.cpp:(.text+0x17b2): undefined reference to `FRVT_MORPH::Interface::getImplementation()'\r\ncollect2: error: ld returned 1 exit status\r\nmake[2]: *** [CMakeFiles/validate_morph.dir/build.make:99: ../bin/validate_morph] Error 1\r\nmake[1]: *** [CMakeFiles/Makefile2:76: CMakeFiles/validate_morph.dir/all] Error 2\r\nmake: *** [Makefile:84: all] Error 2\r\nsms@sms:~/Downloads/morph/src/testdriver_wvu/build4$ \r\n\r\n\r\nCan anyone give me the solution of this two error in the new FRVT?\r\n",
"As a sanity check, do you get the same error when you build the null implementation by running ./scripts/build_null_impl.sh and then running ./run_validate_morph.sh?",
"I do not get any error in null implementation (./scripts/build_null_impl.sh).",
"Can you confirm that your implementation library was compiled and built on Ubuntu? Can you also confirm that you **DO NOT** have \"-DNIST_EXTERN_FRVT_STRUCTS_VERSION -DNIST_EXTERN_API_VERSION\" included in the CMakeLists.txt/Makefile that you built your library with?",
"I have implemented into Ubuntu 20.04.3. I do not have \"-DNIST_EXTERN_FRVT_STRUCTS_VERSION -DNIST_EXTERN_API_VERSION\" those flags.",
"After building the null implementation, did you run ./run_validate_morph.sh to confirm there were no errors with linking against the validation testdriver?",
"After Null implementation I have run ./run_validate_morph.sh but there have no issue and no error.\r\nMoreover,\r\n./src/testdriver/ when i build through cmake and make 2 errors occur always.\r\n",
"sms@sms:~/Downloads/frvt-master_23March/morph/src/testdriver/build$ cmake ..\r\nsms@sms:~/Downloads/frvt-master_23March/morph/src/testdriver/build$ make\r\n\r\nI have not change any code of frvt into Ubuntu 20.04.3 . I just want to build test driver and got same error. That is\r\n\r\nsms@sms:~/Downloads/frvt-master_23March/morph/src/testdriver/build$ make\r\nScanning dependencies of target validate_morph\r\n[ 33%] Building CXX object CMakeFiles/validate_morph.dir/home/sms/Downloads/frvt-master_23March/common/src/util/util.o\r\n[ 66%] Building CXX object CMakeFiles/validate_morph.dir/validate_morph.o\r\n[100%] Linking CXX executable ../bin/validate_morph\r\n/usr/bin/ld: CMakeFiles/validate_morph.dir/validate_morph.o: in function `main':\r\nvalidate_morph.cpp:(.text+0xf9f): undefined reference to `FRVT::FRVT_STRUCTS_MAJOR_VERSION'\r\n/usr/bin/ld: validate_morph.cpp:(.text+0xfaf): undefined reference to `FRVT::FRVT_STRUCTS_MINOR_VERSION'\r\n/usr/bin/ld: validate_morph.cpp:(.text+0xfd9): undefined reference to `FRVT::FRVT_STRUCTS_MAJOR_VERSION'\r\n/usr/bin/ld: validate_morph.cpp:(.text+0xfff): undefined reference to `FRVT::FRVT_STRUCTS_MINOR_VERSION'\r\n/usr/bin/ld: validate_morph.cpp:(.text+0x1087): undefined reference to `FRVT_MORPH::API_MAJOR_VERSION'\r\n/usr/bin/ld: validate_morph.cpp:(.text+0x1097): undefined reference to `FRVT_MORPH::API_MINOR_VERSION'\r\n/usr/bin/ld: validate_morph.cpp:(.text+0x10c1): undefined reference to `FRVT_MORPH::API_MAJOR_VERSION'\r\n/usr/bin/ld: validate_morph.cpp:(.text+0x10e7): undefined reference to `FRVT_MORPH::API_MINOR_VERSION'\r\n/usr/bin/ld: validate_morph.cpp:(.text+0x17b2): undefined reference to `FRVT_MORPH::Interface::getImplementation()'\r\ncollect2: error: ld returned 1 exit status\r\nmake[2]: *** [CMakeFiles/validate_morph.dir/build.make:99: ../bin/validate_morph] Error 1\r\nmake[1]: *** [CMakeFiles/Makefile2:76: CMakeFiles/validate_morph.dir/all] Error 2\r\nmake: *** [Makefile:84: all] Error 2\r\n\r\n\r\nCan I try to packaging the FRVT (old version) into CentOS because of Ubuntu 20.04.3 provide that issue.Please, Solve this issue,it is very urgent because we (WVU team) will submit the new FRVT encrypted package to NIST. \r\nThanks\r\n",
"Ok, I think I see what your issue is. You can't just go into the ./src/testdriver directory and build the testdriver directly. There is a variable called FRVT_IMPL_LIB that needs to be set in order to know what implementation library to link to. Please follow the instructions from the ./README.md file. You need to place your implementation library in ./lib and then run run_validate_morph.sh, which will set up all necessary variables and paths.",
"I will follow the steps that you say and ./README.md file depicts. Thanks for your kind reply.",
"I am able to run run_validate_morph.sh (with this: libfrvt_morph_wvu_001.so in library) into CentOS . But new FRVT with Ubuntu causes 2 error for running run_validate_morph.sh .\r\n**Does NIST allow to previous version of FRVT with CentOS. That will be easy for me to compile and packaging the software.**\r\nThe generated shared object (.so file) causes the problem in Ubuntu. But that shared object (libfrvt_morph_wvu_001.so) able to compile in the previous version of FRVT (operating system is CentOS). \r\nThanks. ",
"Can you confirm that your libfrvt_morph_wvu_001.so file was compiled and built on Ubuntu? What are the errors that you're encountering when running on Ubuntu?",
"Yes. \r\nlibfrvt_morph_wvu_001.so causes the error. When I put libfrvt_morph_null_001.so there have no error. Moreover I have compiled libfrvt_morph_wvu_001.so into the ubuntu. (Also I have checked previous version of libfrvt_morph_wvu_001.so too. ) But facing the same problem for libfrvt_morph_wvu_001.so in Ubuntu. \r\nBut when I switch to CentOS with our previous libfrvt_morph_wvu_001.so it compiled easily. **Can we are allowed to package our software through previous FRVT with CentOS for NIST submission?**",
"Can you please provide the exact error messages that you are seeing when you run ./run_validate_morph.sh on Ubuntu?",
"Thanks for your kind reply.\r\n`sms@sms:~/Downloads/frvt-master_24_march/morph$ ./run_validate_morph.sh\r\nChecking installation of required packages [SUCCESS]\r\nLooking for core implementation library in /home/sms/Downloads/frvt-master_24_march/morph/lib.[SUCCESS] Found core implementation library /home/sms/Downloads/frvt-master_24_march/morph/lib/libfrvt_morph_wvu_001.so.\r\nAttempting to compile and link /home/sms/Downloads/frvt-master_24_march/morph/lib/libfrvt_morph_wvu_001.so against test harness.\r\nScanning dependencies of target validate_morph\r\n[ 33%] Building CXX object src/testdriver/CMakeFiles/validate_morph.dir/home/sms/Downloads/frvt-master_24_march/common/src/util/util.cpp.o\r\n[ 66%] Building CXX object src/testdriver/CMakeFiles/validate_morph.dir/validate_morph.cpp.o\r\n[100%] Linking CXX executable ../../../bin/validate_morph\r\n/usr/bin/ld: /home/sms/Downloads/libtorch/lib/libc10.so: .dynsym local symbol at index 138 (>= sh_info of 2)\r\n/usr/bin/ld: /home/sms/Downloads/libtorch/lib/libc10.so: .dynsym local symbol at index 140 (>= sh_info of 2)\r\n/usr/bin/ld: /home/sms/Downloads/libtorch/lib/libtorch_cpu.so: .dynsym local symbol at index 149 (>= sh_info of 2)\r\n/usr/bin/ld: /home/sms/Downloads/libtorch/lib/libtorch_cpu.so: .dynsym local symbol at index 227 (>= sh_info of 2)\r\n/usr/bin/ld: /home/sms/Downloads/libtorch/lib/libtorch_cpu.so: .dynsym local symbol at index 252 (>= sh_info of 2)\r\n/usr/bin/ld: /home/sms/Downloads/libtorch/lib/libtorch_cpu.so: .dynsym local symbol at index 291 (>= sh_info of 2)\r\n/usr/bin/ld: /home/sms/Downloads/libtorch/lib/libtorch_cpu.so: .dynsym local symbol at index 574 (>= sh_info of 2)\r\n/usr/bin/ld: /home/sms/Downloads/libtorch/lib/libtorch_cpu.so: .dynsym local symbol at index 625 (>= sh_info of 2)\r\n/usr/bin/ld: /home/sms/Downloads/libtorch/lib/libtorch_cpu.so: .dynsym local symbol at index 739 (>= sh_info of 2)\r\n/usr/bin/ld: /home/sms/Downloads/libtorch/lib/libtorch_cpu.so: .dynsym local symbol at index 742 (>= sh_info of 2)\r\n/usr/bin/ld: /home/sms/Downloads/libtorch/lib/libtorch_cpu.so: .dynsym local symbol at index 743 (>= sh_info of 2)\r\n/usr/bin/ld: /home/sms/Downloads/libtorch/lib/libtorch_cpu.so: .dynsym local symbol at index 747 (>= sh_info of 2)\r\n/usr/bin/ld: /home/sms/Downloads/libtorch/lib/libtorch_cpu.so: .dynsym local symbol at index 748 (>= sh_info of 2)\r\n/usr/bin/ld: /home/sms/Downloads/libtorch/lib/libtorch_cpu.so: .dynsym local symbol at index 749 (>= sh_info of 2)\r\n/usr/bin/ld: /home/sms/Downloads/libtorch/lib/libtorch_cpu.so: .dynsym local symbol at index 750 (>= sh_info of 2)\r\n/usr/bin/ld: /home/sms/Downloads/libtorch/lib/libtorch_cpu.so: .dynsym local symbol at index 779 (>= sh_info of 2)\r\n/usr/bin/ld: /home/sms/Downloads/libtorch/lib/libtorch_cpu.so: .dynsym local symbol at index 785 (>= sh_info of 2)\r\n/usr/bin/ld: ../../../lib/libfrvt_morph_wvu_001.so: undefined reference to `at::_ops::view::call(at::Tensor const&, c10::ArrayRef<long>)'\r\n/usr/bin/ld: ../../../lib/libfrvt_morph_wvu_001.so: undefined reference to `at::_ops::div_Scalar::call(at::Tensor const&, c10::Scalar const&)'\r\n/usr/bin/ld: ../../../lib/libfrvt_morph_wvu_001.so: undefined reference to `torch::jit::Method::operator()(std::vector<c10::IValue, std::allocator<c10::IValue> >, std::unordered_map<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> >, c10::IValue, std::hash<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::equal_to<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > >, std::allocator<std::pair<std::__cxx11::basic_string<char, std::char_traits<char>, std::allocator<char> > const, c10::IValue> > > const&) const'\r\n/usr/bin/ld: ../../../lib/libfrvt_morph_wvu_001.so: undefined reference to `at::_ops::full::call(c10::ArrayRef<long>, c10::Scalar const&, c10::optional<c10::ScalarType>, c10::optional<c10::Layout>, c10::optional<c10::Device>, c10::optional<bool>)'\r\n/usr/bin/ld: ../../../lib/libfrvt_morph_wvu_001.so: undefined reference to `at::TensorBase::__dispatch_contiguous(c10::MemoryFormat) const'\r\n/usr/bin/ld: ../../../lib/libfrvt_morph_wvu_001.so: undefined reference to `at::_ops::select_int::call(at::Tensor const&, long, long)'\r\n/usr/bin/ld: ../../../lib/libfrvt_morph_wvu_001.so: undefined reference to `at::_ops::to_dtype::call(at::Tensor const&, c10::ScalarType, bool, bool, c10::optional<c10::MemoryFormat>)'\r\n/usr/bin/ld: ../../../lib/libfrvt_morph_wvu_001.so: undefined reference to `at::_ops::permute::call(at::Tensor const&, c10::ArrayRef<long>)'\r\n/usr/bin/ld: ../../../lib/libfrvt_morph_wvu_001.so: undefined reference to `vtable for torch::jit::Method'\r\n/usr/bin/ld: ../../../lib/libfrvt_morph_wvu_001.so: undefined reference to `at::_ops::copy_::call(at::Tensor&, at::Tensor const&, bool)'\r\n/usr/bin/ld: ../../../lib/libfrvt_morph_wvu_001.so: undefined reference to `at::_ops::sub_Scalar::call(at::Tensor const&, c10::Scalar const&, c10::Scalar const&)'\r\n/usr/bin/ld: ../../../lib/libfrvt_morph_wvu_001.so: undefined reference to `at::_ops::softmax_int::call(at::Tensor const&, long, c10::optional<c10::ScalarType>)'\r\n/usr/bin/ld: ../../../lib/libfrvt_morph_wvu_001.so: undefined reference to `at::_ops::clone::call(at::Tensor const&, c10::optional<c10::MemoryFormat>)'\r\ncollect2: error: ld returned 1 exit status\r\nmake[2]: *** [src/testdriver/CMakeFiles/validate_morph.dir/build.make:100: ../bin/validate_morph] Error 1\r\nmake[1]: *** [CMakeFiles/Makefile2:94: src/testdriver/CMakeFiles/validate_morph.dir/all] Error 2\r\nmake: *** [Makefile:84: all] Error 2\r\n[ERROR] There were errors during compilation of your library with the validation test harness. Please investigate and re-compile.\r\nsms@sms:~/Downloads/frvt-master_24_march/morph$ \r\n",
"Your errors appear to be related to the version of libtorch that you're linking to may have been compiled with an older compiler - https://github.com/pytorch/pytorch/issues/13541. Have you tried installing a new version of libtorch that's compatible with g++ 9.x or tried re-compiling torch with -D_GLIBCXX_USE_CXX11_ABI=0?",
"I have changed Libtorch also opencv (new version) but there are 2 errors\r\n```\r\nsms@sms:~/Downloads/frvt-master_24_march/morph$ ./run_validate_morph.sh\r\nChecking installation of required packages [SUCCESS]\r\nLooking for core implementation library in /home/sms/Downloads/frvt-master_24_march/morph/lib.[SUCCESS] Found core implementation library /home/sms/Downloads/frvt-master_24_march/morph/lib/libfrvt_morph_wvu_001.so.\r\nAttempting to compile and link /home/sms/Downloads/frvt-master_24_march/morph/lib/libfrvt_morph_wvu_001.so against test harness.\r\nScanning dependencies of target validate_morph\r\n[ 33%] Building CXX object src/testdriver/CMakeFiles/validate_morph.dir/home/sms/Downloads/frvt-master_24_march/common/src/util/util.cpp.o\r\n[ 66%] Building CXX object src/testdriver/CMakeFiles/validate_morph.dir/validate_morph.cpp.o\r\n[100%] Linking CXX executable ../../../bin/validate_morph\r\n/usr/bin/ld: ../../../lib/libfrvt_morph_wvu_001.so: undefined reference to `cv::dnn::dnn4_v20200310::Net::setInput(cv::_InputArray const&, std::string const&, double, cv::Scalar_<double> const&)'\r\n/usr/bin/ld: ../../../lib/libfrvt_morph_wvu_001.so: undefined reference to `cv::dnn::dnn4_v20200310::Net::forward(cv::_OutputArray const&, std::vector<std::string, std::allocator<std::string> > const&)'\r\n/usr/bin/ld: ../../../lib/libfrvt_morph_wvu_001.so: undefined reference to `cv::dnn::dnn4_v20200310::readNetFromCaffe(std::string const&, std::string const&)'\r\ncollect2: error: ld returned 1 exit status\r\nmake[2]: *** [src/testdriver/CMakeFiles/validate_morph.dir/build.make:100: ../bin/validate_morph] Error 1\r\nmake[1]: *** [CMakeFiles/Makefile2:94: src/testdriver/CMakeFiles/validate_morph.dir/all] Error 2\r\nmake: *** [Makefile:84: all] Error 2\r\n[ERROR] There were errors during compilation of your library with the validation test harness. Please investigate and re-compile.\r\nsms@sms:~/Downloads/frvt-master_24_march/morph$ \r\n\r\n```\r\n",
"Could you please solve the above error?",
"@ShoaibMerajSami this looks like an error on your side. It is missing symbols from OpenCV DNN lib",
"Thanks @xsacha . I have tried to solve this issue, so, I have install opencv 4.5.3 from source then I have added the opencv path into the cmake. But there remain 2 error. How can I solve this error. Could you give me yours valuable suggestion for solve OpenCV DNN lib issue?\r\nThe error is:\r\n```\r\nsms@sms:~/Downloads/frvt-master_24_march/morph$ ./run_validate_morph.sh\r\nChecking installation of required packages [SUCCESS]\r\nLooking for core implementation library in /home/sms/Downloads/frvt-master_24_march/morph/lib.[SUCCESS] Found core implementation library /home/sms/Downloads/frvt-master_24_march/morph/lib/libfrvt_morph_wvu_001.so.\r\nAttempting to compile and link /home/sms/Downloads/frvt-master_24_march/morph/lib/libfrvt_morph_wvu_001.so against test harness.\r\nScanning dependencies of target validate_morph\r\n[ 33%] Building CXX object src/testdriver/CMakeFiles/validate_morph.dir/home/sms/Downloads/frvt-master_24_march/common/src/util/util.cpp.o\r\n[ 66%] Building CXX object src/testdriver/CMakeFiles/validate_morph.dir/validate_morph.cpp.o\r\n[100%] Linking CXX executable ../../../bin/validate_morph\r\n/usr/bin/ld: ../../../lib/libfrvt_morph_wvu_001.so: undefined reference to `cv::dnn::dnn4_v20210608::Net::forward(cv::_OutputArray const&, std::vector<std::string, std::allocator<std::string> > const&)'\r\n/usr/bin/ld: ../../../lib/libfrvt_morph_wvu_001.so: undefined reference to `cv::dnn::dnn4_v20210608::Net::setInput(cv::_InputArray const&, std::string const&, double, cv::Scalar_<double> const&)'\r\n/usr/bin/ld: ../../../lib/libfrvt_morph_wvu_001.so: undefined reference to `cv::dnn::dnn4_v20210608::readNetFromCaffe(std::string const&, std::string const&)'\r\ncollect2: error: ld returned 1 exit status\r\nmake[2]: *** [src/testdriver/CMakeFiles/validate_morph.dir/build.make:100: ../bin/validate_morph] Error 1\r\nmake[1]: *** [CMakeFiles/Makefile2:94: src/testdriver/CMakeFiles/validate_morph.dir/all] Error 2\r\nmake: *** [Makefile:84: all] Error 2\r\n[ERROR] There were errors during compilation of your library with the validation test harness. Please investigate and re-compile.\r\n```",
"Is there an OpenCV library with the word 'dnn' in it that you have forgotten to link?\r\nIf building from source it may be easier for you to statically link it in this case."
] | 2022-03-15T10:54:09 | 2022-03-28T20:14:58 | null | NONE | null | null | null | /usr/bin/ld: validate11.cpp:(.text.startup+0x35): undefined reference to `FRVT::FRVT_STRUCTS_MINOR_VERSION'
/usr/bin/ld: validate11.cpp:(.text.startup+0x5b): undefined reference to `FRVT::FRVT_STRUCTS_MAJOR_VERSION'
/usr/bin/ld: validate11.cpp:(.text.startup+0x85): undefined reference to `FRVT::FRVT_STRUCTS_MINOR_VERSION'
/usr/bin/ld: validate11.cpp:(.text.startup+0x12d): undefined reference to `FRVT_11::API_MAJOR_VERSION'
/usr/bin/ld: validate11.cpp:(.text.startup+0x13b): undefined reference to `FRVT_11::API_MINOR_VERSION'
/usr/bin/ld: validate11.cpp:(.text.startup+0x375): undefined reference to `FRVT_11::API_MAJOR_VERSION'
/usr/bin/ld: validate11.cpp:(.text.startup+0x39f): undefined reference to `FRVT_11::API_MINOR_VERSION'
/usr/bin/ld: validate11.cpp:(.text.startup+0x538): undefined reference to `FRVT_11::Interface::getImplementation()'
collect2: error: ld returned 1 exit status
make[2]: *** [src/testdriver/CMakeFiles/validate11.dir/build.make:100:../bin/v
alidate11] Error 1
make[1]: *** [CMakeFiles/Makefilez:94:src/testdriver/CMakeFiles/validate11.dir
/all] Error 2
make: *** [Makefile:84: all] Error 2
I am facing the above issue when I placed my .so file in libs folder
Please suggest as to what changes needs to be done in my .so file or any other file to solve the issue | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/72/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/72/timeline | null | null |
https://api.github.com/repos/usnistgov/frvt/issues/71 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/71/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/71/comments | https://api.github.com/repos/usnistgov/frvt/issues/71/events | https://github.com/usnistgov/frvt/issues/71 | 1,164,674,822 | I_kwDOBIKFiM5Fa4cG | 71 | If my code needs to read the model externally, where should I put the face recognition related models ? | {
"login": "powermano",
"id": 24365567,
"node_id": "MDQ6VXNlcjI0MzY1NTY3",
"avatar_url": "https://avatars.githubusercontent.com/u/24365567?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/powermano",
"html_url": "https://github.com/powermano",
"followers_url": "https://api.github.com/users/powermano/followers",
"following_url": "https://api.github.com/users/powermano/following{/other_user}",
"gists_url": "https://api.github.com/users/powermano/gists{/gist_id}",
"starred_url": "https://api.github.com/users/powermano/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/powermano/subscriptions",
"organizations_url": "https://api.github.com/users/powermano/orgs",
"repos_url": "https://api.github.com/users/powermano/repos",
"events_url": "https://api.github.com/users/powermano/events{/privacy}",
"received_events_url": "https://api.github.com/users/powermano/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"The models should be located in the 'config' directory.\r\n\r\nI don't think there's any guarantee over the other directories which is why the configDir is supplied in the function.",
"> The models should be located in the 'config' directory.\r\n> \r\n> I don't think there's any guarantee over the other directories which is why the configDir is supplied in the function.\r\n\r\nthanks.\r\n\r\n"
] | 2022-03-10T02:52:33 | 2022-03-10T02:58:28 | 2022-03-10T02:58:28 | NONE | null | null | null | Take 1:N as an example. In the NullImplFRVT1N::initializeTemplateCreation func, I should load the face recognition related models. If puting the face recognition related models in the same level directory as run_validate_1N.sh, may i use './face_model' in the initializeTemplateCreation function?
```
NullImplFRVT1N::initializeTemplateCreation(
const std::string &configDir,
FRVT::TemplateRole role)
{
face_model model = load('./face_model');
return ReturnStatus(ReturnCode::Success);
}
```
the following is the tree struct of 1N directory.
|-- CMakeLists.txt
|-- README.md
|-- bin
| `-- validate1N
|-- config
|-- doc
| `-- version.txt
|-- input
| |-- enroll_1N.txt
| |-- enroll_1N_multiface.txt
| |-- enroll_1N_short.txt
| |-- insert.txt
| |-- search_1N.txt
| `-- search_1N_multiperson.txt
|-- lib
| `-- libfrvt_1N_xxx_000.so
|-- libfrvt_1N_xxx_000.tar.gz
|-- run_validate_1N.sh
|-- scripts
| |-- build_null_impl.sh
| |-- compile_and_link.sh
| `-- run_testdriver.sh
`-- validation
| {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/71/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/71/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/70 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/70/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/70/comments | https://api.github.com/repos/usnistgov/frvt/issues/70/events | https://github.com/usnistgov/frvt/issues/70 | 1,162,549,843 | I_kwDOBIKFiM5FSxpT | 70 | match against failed templates | {
"login": "mtamburrano",
"id": 1943264,
"node_id": "MDQ6VXNlcjE5NDMyNjQ=",
"avatar_url": "https://avatars.githubusercontent.com/u/1943264?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/mtamburrano",
"html_url": "https://github.com/mtamburrano",
"followers_url": "https://api.github.com/users/mtamburrano/followers",
"following_url": "https://api.github.com/users/mtamburrano/following{/other_user}",
"gists_url": "https://api.github.com/users/mtamburrano/gists{/gist_id}",
"starred_url": "https://api.github.com/users/mtamburrano/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/mtamburrano/subscriptions",
"organizations_url": "https://api.github.com/users/mtamburrano/orgs",
"repos_url": "https://api.github.com/users/mtamburrano/repos",
"events_url": "https://api.github.com/users/mtamburrano/events{/privacy}",
"received_events_url": "https://api.github.com/users/mtamburrano/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"You can set the similarity score to whatever value you think is a good idea, for example, returning a very low score to represent a rejection or setting the similarity score to -1.0."
] | 2022-03-08T11:36:10 | 2022-03-08T13:22:53 | 2022-03-08T13:22:52 | NONE | null | null | null | Hi,
for frvt_11, I read in th documentation of the template creation:
```
In all cases, even when unable to extract features, the output shall be a template that may be
passed to the matchTemplates() function without error. That is, this routine must internally encode "template
creation failed" and the matcher must transparently handle this.
```
Fine, but what about similarity?
Evene if I return some error during template creation step, that template is still being passed to `matchTemplates` and returning a similarity makes no sense.
The failed templates shouldn't be supposed to not being passed to match functions? If not, what similarity should I return in those cases?
Thanks | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/70/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/70/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/69 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/69/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/69/comments | https://api.github.com/repos/usnistgov/frvt/issues/69/events | https://github.com/usnistgov/frvt/issues/69 | 1,149,775,199 | I_kwDOBIKFiM5EiC1f | 69 | Tensorflow is not fork safe | {
"login": "FranciscoSierra27",
"id": 73256804,
"node_id": "MDQ6VXNlcjczMjU2ODA0",
"avatar_url": "https://avatars.githubusercontent.com/u/73256804?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/FranciscoSierra27",
"html_url": "https://github.com/FranciscoSierra27",
"followers_url": "https://api.github.com/users/FranciscoSierra27/followers",
"following_url": "https://api.github.com/users/FranciscoSierra27/following{/other_user}",
"gists_url": "https://api.github.com/users/FranciscoSierra27/gists{/gist_id}",
"starred_url": "https://api.github.com/users/FranciscoSierra27/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/FranciscoSierra27/subscriptions",
"organizations_url": "https://api.github.com/users/FranciscoSierra27/orgs",
"repos_url": "https://api.github.com/users/FranciscoSierra27/repos",
"events_url": "https://api.github.com/users/FranciscoSierra27/events{/privacy}",
"received_events_url": "https://api.github.com/users/FranciscoSierra27/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"All the main frameworks such as tensorflow, pytorch and mxnet have been known to work in NIST.\r\nYou may just need to compile it differently or set it up (usually a setting in each framework) so it does not attempt to use multi-threading.",
"Okay, I'll see if it is a multi thread problem and if I can recompile it without multi-threading.\r\nAlso, I'm closing this issue for now.",
"Hi NIST team.\r\nI've searching a lot and sadly, I've had the same problem. \r\nI even migrated my TF Rust code to CPP with no luck.\r\n\r\nResearching a little more I foud an open issue regarding NIST implementatinos in the official TF github and they say as follows:\r\n[https://github.com/tensorflow/tensorflow/issues/51832#issuecomment-926205545\r\n](https://github.com/tensorflow/tensorflow/issues/51832#issuecomment-926205545)\r\nSo basically NIST API is not compatible with all model inferences with TF and the TF developers have not intenssions to make TF fork safe.\r\n\r\n¿Do you have any plans to make more compatible your API with TF?\r\nFor me it's contradictory that you use forks when TF, one of the more spreaded IA framekors, is not fork safe.\r\n\r\nI've seen that TF allows \"sessions\" as servers, and \"session-clients\". But that will require an extra process to use it. I assume that that implementation is not allowed by NIST.",
"We've received a number of submissions using TF. It's possible to move your model initialization to the child process (after the fork) and invoke it once during the first API call. As for measuring timing, we consider and report the median time across thousands of API calls (e.g. createTemplate()), so even if your very first API function invocation takes longer, it should not impact the timing benchmark.",
"Thanks. That was my first approach, but TF in Rust takes a long time to initialize. In c++ the initialization is faster.\r\nNow I'm initializing at \"matchTemplate\" once per fork.\r\n\r\nJust as conclussion for this issue. If you initialize your TF model at \"initialize()\"w and it hangs at \"matchTemlate\", you should try to initialize it once per fork at \"matchTemplate\" or \"createTemplate\". Regarding the time, it may not affect considerably the benchmark due the mean calculatiuon of thousands of API calls.\r\n\r\n¿Am I right?\r\n\r\n"
] | 2022-02-24T21:37:47 | 2022-03-02T19:13:22 | 2022-03-02T18:52:48 | NONE | null | null | null | Hi NIST team.
I've created a dynamic library from Rust that my implementation of "frvt.h" in c++ uses, but I have a problem.
My model inference is made with TensorFlow, when I run the "run_validate.sh". It hangs indefinitely.
I've read that TF is not fork safe so that might be the problem.
¿Do you suggest to try another Machine Learning framekork or is there a workaround for this issue?
Edit: added a missing word "framework".
| {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/69/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/69/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/68 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/68/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/68/comments | https://api.github.com/repos/usnistgov/frvt/issues/68/events | https://github.com/usnistgov/frvt/issues/68 | 1,136,858,124 | I_kwDOBIKFiM5DwxQM | 68 | Does the implementation of FRVT 1:N search also need to be limited to a single process? | {
"login": "powermano",
"id": 24365567,
"node_id": "MDQ6VXNlcjI0MzY1NTY3",
"avatar_url": "https://avatars.githubusercontent.com/u/24365567?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/powermano",
"html_url": "https://github.com/powermano",
"followers_url": "https://api.github.com/users/powermano/followers",
"following_url": "https://api.github.com/users/powermano/following{/other_user}",
"gists_url": "https://api.github.com/users/powermano/gists{/gist_id}",
"starred_url": "https://api.github.com/users/powermano/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/powermano/subscriptions",
"organizations_url": "https://api.github.com/users/powermano/orgs",
"repos_url": "https://api.github.com/users/powermano/repos",
"events_url": "https://api.github.com/users/powermano/events{/privacy}",
"received_events_url": "https://api.github.com/users/powermano/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"Yes, the implementation of 1:N search needs to be limited to a single process."
] | 2022-02-14T06:57:33 | 2022-02-14T13:25:26 | 2022-02-14T13:25:26 | NONE | null | null | null | As shown in the following picture, you need to test the speed of 1:N search with one single-core, but in the run_testdriver.sh script, only the number of processes during createTemplate is counted, but not during 1;N search. Does the implementation of FRVT 1:N search also need to be limited to a single process?
![image](https://user-images.githubusercontent.com/24365567/153814230-b302548e-8708-41fa-9f0e-649698e14b63.png)
| {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/68/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/68/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/67 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/67/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/67/comments | https://api.github.com/repos/usnistgov/frvt/issues/67/events | https://github.com/usnistgov/frvt/issues/67 | 1,129,555,592 | I_kwDOBIKFiM5DU6aI | 67 | input/match_multiperson.txt does not exists? | {
"login": "powermano",
"id": 24365567,
"node_id": "MDQ6VXNlcjI0MzY1NTY3",
"avatar_url": "https://avatars.githubusercontent.com/u/24365567?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/powermano",
"html_url": "https://github.com/powermano",
"followers_url": "https://api.github.com/users/powermano/followers",
"following_url": "https://api.github.com/users/powermano/following{/other_user}",
"gists_url": "https://api.github.com/users/powermano/gists{/gist_id}",
"starred_url": "https://api.github.com/users/powermano/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/powermano/subscriptions",
"organizations_url": "https://api.github.com/users/powermano/orgs",
"repos_url": "https://api.github.com/users/powermano/repos",
"events_url": "https://api.github.com/users/powermano/events{/privacy}",
"received_events_url": "https://api.github.com/users/powermano/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"The match_multiperson.txt generated from the following code;\r\n```\r\ndo\r\n grep \"${line}_\" $outputDir/$outputStem.log | awk -v enroll=$line '{ print enroll\".template \" $1\".template\" }' >> $inputFileMulti\r\ndone\r\n```"
] | 2022-02-10T07:13:11 | 2022-02-10T07:18:24 | 2022-02-10T07:16:10 | NONE | null | null | null | ```
echo -n "Matching Multi-person Templates (Single Process) "
inputFileMulti=input/match_multiperson.txt
rm -f $inputFileMulti
sed '1d' $inputFile | awk '{ print $1 }' | while read line
```
https://github.com/usnistgov/frvt/blob/ubuntu/11/scripts/run_testdriver.sh#:~:text=inputFileMulti%3Dinput/match_multiperson.txt
| {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/67/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/67/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/66 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/66/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/66/comments | https://api.github.com/repos/usnistgov/frvt/issues/66/events | https://github.com/usnistgov/frvt/issues/66 | 1,127,036,550 | I_kwDOBIKFiM5DLTaG | 66 | ubuntu 20.04.3 LTS evaluation gcc version | {
"login": "slpal",
"id": 21005856,
"node_id": "MDQ6VXNlcjIxMDA1ODU2",
"avatar_url": "https://avatars.githubusercontent.com/u/21005856?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/slpal",
"html_url": "https://github.com/slpal",
"followers_url": "https://api.github.com/users/slpal/followers",
"following_url": "https://api.github.com/users/slpal/following{/other_user}",
"gists_url": "https://api.github.com/users/slpal/gists{/gist_id}",
"starred_url": "https://api.github.com/users/slpal/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/slpal/subscriptions",
"organizations_url": "https://api.github.com/users/slpal/orgs",
"repos_url": "https://api.github.com/users/slpal/repos",
"events_url": "https://api.github.com/users/slpal/events{/privacy}",
"received_events_url": "https://api.github.com/users/slpal/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"I used gcc 11 with CentOS (via epel) on previous submissions.\r\nUbuntu 20.04 comes with gcc-10 in the standard repo by the way.\r\n\r\nThe version of gcc is unlikely to be the matter but rather the version of glibc, which any compiler on Ubuntu 20.04 should use.",
"Our Ubuntu image is running g++ version 9.3.0."
] | 2022-02-08T10:03:29 | 2022-02-08T13:40:06 | 2022-02-08T13:40:06 | NONE | null | null | null | Should gcc version be 9.3 (ubuntu 20.04.3 LTS default) or 8.3.1 as usual, when switching the FRVT evaluation operating system to Ubuntu 20.04.3 LTS?
Thx | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/66/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/66/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/65 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/65/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/65/comments | https://api.github.com/repos/usnistgov/frvt/issues/65/events | https://github.com/usnistgov/frvt/issues/65 | 1,122,298,340 | I_kwDOBIKFiM5C5Onk | 65 | match_multiperson.txt file is missing. | {
"login": "NessAupera",
"id": 38118941,
"node_id": "MDQ6VXNlcjM4MTE4OTQx",
"avatar_url": "https://avatars.githubusercontent.com/u/38118941?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/NessAupera",
"html_url": "https://github.com/NessAupera",
"followers_url": "https://api.github.com/users/NessAupera/followers",
"following_url": "https://api.github.com/users/NessAupera/following{/other_user}",
"gists_url": "https://api.github.com/users/NessAupera/gists{/gist_id}",
"starred_url": "https://api.github.com/users/NessAupera/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/NessAupera/subscriptions",
"organizations_url": "https://api.github.com/users/NessAupera/orgs",
"repos_url": "https://api.github.com/users/NessAupera/repos",
"events_url": "https://api.github.com/users/NessAupera/events{/privacy}",
"received_events_url": "https://api.github.com/users/NessAupera/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"If your software successfully runs through validation software, match_multiperson.txt gets generated at the end in scripts/run_testdriver.sh."
] | 2022-02-02T19:22:03 | 2022-02-02T19:27:40 | 2022-02-02T19:27:39 | NONE | null | null | null | The last test in ./run_validate_11.sh regarding "Matching Multi-person Templates (Single Process) " needs match_multiperson.txt which is not available in the input folder.
Please advise on how to proceed. | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/65/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/65/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/64 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/64/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/64/comments | https://api.github.com/repos/usnistgov/frvt/issues/64/events | https://github.com/usnistgov/frvt/issues/64 | 999,678,854 | I_kwDOBIKFiM47leOG | 64 | Thread count | {
"login": "avinashkshirsagar",
"id": 25266500,
"node_id": "MDQ6VXNlcjI1MjY2NTAw",
"avatar_url": "https://avatars.githubusercontent.com/u/25266500?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/avinashkshirsagar",
"html_url": "https://github.com/avinashkshirsagar",
"followers_url": "https://api.github.com/users/avinashkshirsagar/followers",
"following_url": "https://api.github.com/users/avinashkshirsagar/following{/other_user}",
"gists_url": "https://api.github.com/users/avinashkshirsagar/gists{/gist_id}",
"starred_url": "https://api.github.com/users/avinashkshirsagar/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/avinashkshirsagar/subscriptions",
"organizations_url": "https://api.github.com/users/avinashkshirsagar/orgs",
"repos_url": "https://api.github.com/users/avinashkshirsagar/repos",
"events_url": "https://api.github.com/users/avinashkshirsagar/events{/privacy}",
"received_events_url": "https://api.github.com/users/avinashkshirsagar/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"We are only interested in the active threads, so given you only have a single active thread, it should be fine. Send us your submission, and we will validate the active thread count on our end. Thanks."
] | 2021-09-17T20:00:24 | 2021-09-17T20:07:26 | 2021-09-17T20:07:26 | NONE | null | null | null | I have limited thread of Tensorflow and Pytorch to 1, but still the warning message mentions about detecting 6 threads. Out of these 6 threads only one thread is running and others are in sleep mode. The following is the output of htop command for the above mentioned scenario. In one of the comment it is mentioned that you check multithreading case by case, Is this case acceptable for submission?
![Screenshot (537) - Copy](https://user-images.githubusercontent.com/25266500/133846525-ad5b2849-f2e9-4389-b67e-1a6f4b5ad2ea.png)
| {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/64/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/64/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/63 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/63/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/63/comments | https://api.github.com/repos/usnistgov/frvt/issues/63/events | https://github.com/usnistgov/frvt/issues/63 | 995,326,349 | MDU6SXNzdWU5OTUzMjYzNDk= | 63 | Resolution constraints for the input image? | {
"login": "sparupat",
"id": 2997890,
"node_id": "MDQ6VXNlcjI5OTc4OTA=",
"avatar_url": "https://avatars.githubusercontent.com/u/2997890?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sparupat",
"html_url": "https://github.com/sparupat",
"followers_url": "https://api.github.com/users/sparupat/followers",
"following_url": "https://api.github.com/users/sparupat/following{/other_user}",
"gists_url": "https://api.github.com/users/sparupat/gists{/gist_id}",
"starred_url": "https://api.github.com/users/sparupat/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/sparupat/subscriptions",
"organizations_url": "https://api.github.com/users/sparupat/orgs",
"repos_url": "https://api.github.com/users/sparupat/repos",
"events_url": "https://api.github.com/users/sparupat/events{/privacy}",
"received_events_url": "https://api.github.com/users/sparupat/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"There is no minimum/maximum resolution per se. In our report and on our leaderboard under the resources tab, you'll see that we run on images as large as 3000x4500, sometimes larger. If your software inherently does not support certain image sizes, it should fail gracefully, meaning return a successful return code and encode the failure within your template. It should not crash."
] | 2021-09-13T20:59:29 | 2021-09-13T21:42:45 | 2021-09-13T21:42:45 | NONE | null | null | null | What is the minimum and maximum resolution of input image to createTemplate method or is it 640x480?
createTemplate(
const Multiface &faces,
TemplateRole role, Input
std::vector<uint8_t> &templ,
std::vector<EyePair> &eyeCoordinates); | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/63/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/63/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/62 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/62/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/62/comments | https://api.github.com/repos/usnistgov/frvt/issues/62/events | https://github.com/usnistgov/frvt/issues/62 | 995,209,551 | MDU6SXNzdWU5OTUyMDk1NTE= | 62 | Repo takes a very long time to clone | {
"login": "cyrusbehr",
"id": 17056751,
"node_id": "MDQ6VXNlcjE3MDU2NzUx",
"avatar_url": "https://avatars.githubusercontent.com/u/17056751?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/cyrusbehr",
"html_url": "https://github.com/cyrusbehr",
"followers_url": "https://api.github.com/users/cyrusbehr/followers",
"following_url": "https://api.github.com/users/cyrusbehr/following{/other_user}",
"gists_url": "https://api.github.com/users/cyrusbehr/gists{/gist_id}",
"starred_url": "https://api.github.com/users/cyrusbehr/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/cyrusbehr/subscriptions",
"organizations_url": "https://api.github.com/users/cyrusbehr/orgs",
"repos_url": "https://api.github.com/users/cyrusbehr/repos",
"events_url": "https://api.github.com/users/cyrusbehr/events{/privacy}",
"received_events_url": "https://api.github.com/users/cyrusbehr/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"I was able to reduce the build time significantly by caching the github repo. "
] | 2021-09-13T18:44:53 | 2022-01-07T19:47:36 | 2022-01-07T19:47:36 | NONE | null | null | null | Due to all the image files which are stored in the github repo, the repo takes a **very** long time to clone.
We build our NIST FRVT implementation as part of a CI pipeline, and it is currently taking over 10 minutes to clone the repository.
Is it possible to store the images outside of the github repo on a faster data storage platform (ex. google cloud storage), and perhaps provide a shell script to download and extract the images to the correct location? | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/62/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/62/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/61 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/61/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/61/comments | https://api.github.com/repos/usnistgov/frvt/issues/61/events | https://github.com/usnistgov/frvt/issues/61 | 978,938,582 | MDU6SXNzdWU5Nzg5Mzg1ODI= | 61 | galleryInsertID in 1N | {
"login": "fisakhan",
"id": 48552619,
"node_id": "MDQ6VXNlcjQ4NTUyNjE5",
"avatar_url": "https://avatars.githubusercontent.com/u/48552619?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/fisakhan",
"html_url": "https://github.com/fisakhan",
"followers_url": "https://api.github.com/users/fisakhan/followers",
"following_url": "https://api.github.com/users/fisakhan/following{/other_user}",
"gists_url": "https://api.github.com/users/fisakhan/gists{/gist_id}",
"starred_url": "https://api.github.com/users/fisakhan/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/fisakhan/subscriptions",
"organizations_url": "https://api.github.com/users/fisakhan/orgs",
"repos_url": "https://api.github.com/users/fisakhan/repos",
"events_url": "https://api.github.com/users/fisakhan/events{/privacy}",
"received_events_url": "https://api.github.com/users/fisakhan/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"The null implementation currently stores templates in a std::map, as an example. If your algorithm stores templates with a different data structure (e.g., to enable faster search), then what's in the null implementation is obviously not going work."
] | 2021-08-25T09:44:13 | 2021-08-25T17:46:26 | 2021-08-25T17:46:26 | NONE | null | null | null | Should I change the implementation of galleryInsertID(const std::vector<uint8_t> &templ, const std::string &id) function frvt/1N ? What do I loose if keep the implementation of this function exactly as that of frvt/1N/src/nullImpl/nullimplfrvt1N.cpp ? | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/61/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/61/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/60 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/60/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/60/comments | https://api.github.com/repos/usnistgov/frvt/issues/60/events | https://github.com/usnistgov/frvt/issues/60 | 971,470,147 | MDU6SXNzdWU5NzE0NzAxNDc= | 60 | Libarchive and cmake version mismatch | {
"login": "sounak07",
"id": 22268899,
"node_id": "MDQ6VXNlcjIyMjY4ODk5",
"avatar_url": "https://avatars.githubusercontent.com/u/22268899?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sounak07",
"html_url": "https://github.com/sounak07",
"followers_url": "https://api.github.com/users/sounak07/followers",
"following_url": "https://api.github.com/users/sounak07/following{/other_user}",
"gists_url": "https://api.github.com/users/sounak07/gists{/gist_id}",
"starred_url": "https://api.github.com/users/sounak07/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/sounak07/subscriptions",
"organizations_url": "https://api.github.com/users/sounak07/orgs",
"repos_url": "https://api.github.com/users/sounak07/repos",
"events_url": "https://api.github.com/users/sounak07/events{/privacy}",
"received_events_url": "https://api.github.com/users/sounak07/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"Interesting. We've never encountered this issue with cmake on our end. Our systems are running CentOS Linux release 8.2.2004, cmake version 3.11.4, and libarchive.so.13.3.2.",
"We are seeing this error with CMake version 3.18. So I think cmake 3.11.4 wont have issue, we can submit the package.",
"For cmake 3.11 libarchive is not required , but for 3.18 which your script installs libarchive might be required, so do u think the repo script should be updated?",
"I just did a fresh install of the CentOS 8.2 image on a VM and the latest cmake package being pulled from AppStream is indeed version 3.18.2-11. I was able to replicate your issue with libarchive, so I've added the libarchive package to the installation list in the validation scripts (which should pull the correct/compatible version).",
"I've just fresh install the CentOS 8.2 on VM, and even the new commit has added libarchive but the issue is still.\r\nIt seems that libarchive 3.3.2 is already installed in my fresh installed CentOS. I solved it by update libarchive to 3.3.3 using:\r\n`sudo yum update libarchive`",
"Thanks for making us aware of this. We have added the libarchive update to the validation scripts."
] | 2021-08-16T08:00:23 | 2021-09-10T15:19:53 | 2021-08-16T18:51:16 | NONE | null | null | null | While running the scripts provided by NIST to compile and build the final package we are getting an error
`cmake: symbol lookup error: cmake: undefined symbol: archive_write_add_filter_zstd`
In order to resolve the issue we had to upgrade the libarchive to 3.3.3. Also, the issue was not seen with CMake version 3.17.
Question: Is this issue been already resolved from NIST's end, since the repo is quite old and maybe has not been updated.
Should we go ahead with a version that works or it has to be the version in NIST's scripts install?
[Similar issue ](https://bugzilla.redhat.com/show_bug.cgi?id=1827927) | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/60/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/60/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/59 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/59/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/59/comments | https://api.github.com/repos/usnistgov/frvt/issues/59/events | https://github.com/usnistgov/frvt/issues/59 | 970,840,974 | MDU6SXNzdWU5NzA4NDA5NzQ= | 59 | Testing the final submission package | {
"login": "manideep2510",
"id": 25865501,
"node_id": "MDQ6VXNlcjI1ODY1NTAx",
"avatar_url": "https://avatars.githubusercontent.com/u/25865501?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/manideep2510",
"html_url": "https://github.com/manideep2510",
"followers_url": "https://api.github.com/users/manideep2510/followers",
"following_url": "https://api.github.com/users/manideep2510/following{/other_user}",
"gists_url": "https://api.github.com/users/manideep2510/gists{/gist_id}",
"starred_url": "https://api.github.com/users/manideep2510/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/manideep2510/subscriptions",
"organizations_url": "https://api.github.com/users/manideep2510/orgs",
"repos_url": "https://api.github.com/users/manideep2510/repos",
"events_url": "https://api.github.com/users/manideep2510/events{/privacy}",
"received_events_url": "https://api.github.com/users/manideep2510/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"If the validation script successfully ran and was able to generate the submission package, then that means it passed all preliminary validation checks, and you can just send us the encrypted tar.gz package to NIST.\r\n\r\nIf there are errors during validation on our end, we will notify you with the issues."
] | 2021-08-14T06:38:05 | 2021-08-16T12:19:50 | 2021-08-16T12:19:50 | NONE | null | null | null | Hello
For example in 1:N package, upon successfully running `run_validate_1N.sh` script (in a fresh CentOS 8.2 system) without any error, it will create a submission package compressed in tar.gz format. How can we use this generated package to further test whether this submission package will run from NIST's side or not?
If we are able to prepare the submission package without any error, but during validation from NIST team, some errors pop up, how can we replicate these errors from our side? Before submitting the encrypted package, is there any script/command that we can use to test the validity of the submission package (the tar.gz file generated) like the NIST team would do?
Thanks in advance. | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/59/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/59/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/58 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/58/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/58/comments | https://api.github.com/repos/usnistgov/frvt/issues/58/events | https://github.com/usnistgov/frvt/issues/58 | 956,958,718 | MDU6SXNzdWU5NTY5NTg3MTg= | 58 | What happens if the submission fails timing tests? | {
"login": "manideep2510",
"id": 25865501,
"node_id": "MDQ6VXNlcjI1ODY1NTAx",
"avatar_url": "https://avatars.githubusercontent.com/u/25865501?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/manideep2510",
"html_url": "https://github.com/manideep2510",
"followers_url": "https://api.github.com/users/manideep2510/followers",
"following_url": "https://api.github.com/users/manideep2510/following{/other_user}",
"gists_url": "https://api.github.com/users/manideep2510/gists{/gist_id}",
"starred_url": "https://api.github.com/users/manideep2510/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/manideep2510/subscriptions",
"organizations_url": "https://api.github.com/users/manideep2510/orgs",
"repos_url": "https://api.github.com/users/manideep2510/repos",
"events_url": "https://api.github.com/users/manideep2510/events{/privacy}",
"received_events_url": "https://api.github.com/users/manideep2510/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"If your submission fails timing, you will have one opportunity to resubmit immediately to resolve the timing issue. We process submissions on a first-come-first-serve basis, so once we receive a working submission from you, we will run it and include it in the next report cycle."
] | 2021-07-30T18:01:02 | 2021-07-30T18:06:08 | 2021-07-30T18:05:49 | NONE | null | null | null | Hello,
If the package we submitted runs successfully without package/code/threads errors, but it does not pass the timing tests (Eg: Under 1500 ms for template creation in 1:1 package). Would we get a chance to fix the package and resend it immediately or would we have to wait for 4 calendar months?
If we can fix and resend the package immediately, will the submission be evaluated in the current iteration leaderboard update or the next update iteration?
Thanks in advance. | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/58/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/58/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/57 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/57/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/57/comments | https://api.github.com/repos/usnistgov/frvt/issues/57/events | https://github.com/usnistgov/frvt/issues/57 | 942,086,938 | MDU6SXNzdWU5NDIwODY5Mzg= | 57 | 1:N submission - How to return templates when faces are not detected? | {
"login": "aravinthmuthu",
"id": 17089804,
"node_id": "MDQ6VXNlcjE3MDg5ODA0",
"avatar_url": "https://avatars.githubusercontent.com/u/17089804?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/aravinthmuthu",
"html_url": "https://github.com/aravinthmuthu",
"followers_url": "https://api.github.com/users/aravinthmuthu/followers",
"following_url": "https://api.github.com/users/aravinthmuthu/following{/other_user}",
"gists_url": "https://api.github.com/users/aravinthmuthu/gists{/gist_id}",
"starred_url": "https://api.github.com/users/aravinthmuthu/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/aravinthmuthu/subscriptions",
"organizations_url": "https://api.github.com/users/aravinthmuthu/orgs",
"repos_url": "https://api.github.com/users/aravinthmuthu/repos",
"events_url": "https://api.github.com/users/aravinthmuthu/events{/privacy}",
"received_events_url": "https://api.github.com/users/aravinthmuthu/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"Per the API document (https://pages.nist.gov/frvt/api/FRVT_ongoing_1N_api.pdf):\r\n\r\nFor enrollment templates (TemplateRole=Enrollment_1N): When the implementation fails to produce a template (i.e. returns a non-successful return code), it shall still return a blank template (which can be zero bytes in length). The template will be included in the enrollment database/manifest like all other enrollment templates, but is not expected to contain any feature information.\r\n\r\nFor identification/probe templates (TemplateRole=Search_1N): If the function returns a non-successful return status, the output template will not be used in subsequent search operations.\r\n"
] | 2021-07-12T13:51:40 | 2021-07-12T14:04:29 | 2021-07-12T14:04:29 | NONE | null | null | null | In cases where faces are not detected by the detector,
1) should we return an error code or a random template (feature vector, eye coordinates)?
2) How will these templates be handled during enrolment, search phase? | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/57/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/57/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/56 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/56/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/56/comments | https://api.github.com/repos/usnistgov/frvt/issues/56/events | https://github.com/usnistgov/frvt/issues/56 | 939,046,248 | MDU6SXNzdWU5MzkwNDYyNDg= | 56 | Timing analysis | {
"login": "sam7177",
"id": 84881532,
"node_id": "MDQ6VXNlcjg0ODgxNTMy",
"avatar_url": "https://avatars.githubusercontent.com/u/84881532?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/sam7177",
"html_url": "https://github.com/sam7177",
"followers_url": "https://api.github.com/users/sam7177/followers",
"following_url": "https://api.github.com/users/sam7177/following{/other_user}",
"gists_url": "https://api.github.com/users/sam7177/gists{/gist_id}",
"starred_url": "https://api.github.com/users/sam7177/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/sam7177/subscriptions",
"organizations_url": "https://api.github.com/users/sam7177/orgs",
"repos_url": "https://api.github.com/users/sam7177/repos",
"events_url": "https://api.github.com/users/sam7177/events{/privacy}",
"received_events_url": "https://api.github.com/users/sam7177/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"We are basically doing the following (pseudocode):\r\n\r\nstart = std::chrono::system_clock::now()\r\ncreateTemplate()\r\nstop = std::chrono::system_clock::now()\r\ntime = stop - start\r\n",
"@mlngan Thanks for the reply.\r\nBut does this code use any forking?\r\nIn your comment here https://github.com/usnistgov/frvt/issues/12#issuecomment-501275609 you mentioned \"we lock all threads to using a single CPU\". Does this mean that the validation package is tested on a multi-cpu system, but only confined to single cpu via code? \r\n\r\n",
"I'm sure they just set the process affinity using taskset or similar.\r\ntaskset -c$assignedcore ./testVendor\r\n\r\nIt is tested on one of the machines they list in the General Evaluation document. \r\n Dual Intel® Xeon® E5-2630 v4 CPU @ 2.20GHz (10 cores each)\r\n Dual Intel® Xeon® E5-2680 v4 CPU @ 2.4GHz (14 cores each)\r\n Dual Intel® Xeon® Gold 6140 CPU @ 2.30GHz (18 cores each)"
] | 2021-07-07T16:07:22 | 2021-07-07T22:16:04 | 2021-07-07T18:48:28 | NONE | null | null | null | Hello,
Do you use a separate script to perform timing analysis?
Would it be possible for you to share the code you use to test template creation time?
Thanks. | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/56/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/56/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/55 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/55/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/55/comments | https://api.github.com/repos/usnistgov/frvt/issues/55/events | https://github.com/usnistgov/frvt/issues/55 | 938,985,741 | MDU6SXNzdWU5Mzg5ODU3NDE= | 55 | Threshold to make the decision in FRVT 1:N package | {
"login": "manideep2510",
"id": 25865501,
"node_id": "MDQ6VXNlcjI1ODY1NTAx",
"avatar_url": "https://avatars.githubusercontent.com/u/25865501?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/manideep2510",
"html_url": "https://github.com/manideep2510",
"followers_url": "https://api.github.com/users/manideep2510/followers",
"following_url": "https://api.github.com/users/manideep2510/following{/other_user}",
"gists_url": "https://api.github.com/users/manideep2510/gists{/gist_id}",
"starred_url": "https://api.github.com/users/manideep2510/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/manideep2510/subscriptions",
"organizations_url": "https://api.github.com/users/manideep2510/orgs",
"repos_url": "https://api.github.com/users/manideep2510/repos",
"events_url": "https://api.github.com/users/manideep2510/events{/privacy}",
"received_events_url": "https://api.github.com/users/manideep2510/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"The **decision** would a best guess. You might apply a threshold, or you might do something else (for example, what if more than one person comes above threshold in a consolidated database?). We'd use the **decision** to compute one point on a Detection Error Tradeoff (DET) curve. It would probably lie on the DET we report (which sweeps across a number of different thresholds).\r\n\r\nWe're not going to use the **decision** for algorithm comparison."
] | 2021-07-07T15:00:43 | 2021-07-07T15:23:18 | 2021-07-07T15:23:18 | NONE | null | null | null | In Table 10 on page 13 of [FRVT 1:N API document](https://pages.nist.gov/frvt/api/FRVT_ongoing_1N_api.pdf), we are asked to return a `decision` (true/false) whether there is a mate within the enrollment database for a query along with the candidate IDs and similarity score. What threshold should be used to make this decision in the [`identifyTemplate`](https://github.com/usnistgov/frvt/blob/master/1N/src/nullImpl/nullimplfrvt1N.cpp#L106) function?
On the [1:N leader board page](https://pages.nist.gov/frvt/html/frvt1N.html), it is told that `"threshold is set to limit to the False Positive Identification Rate (FPIR) to 0.003"` and also that `"The threshold is set for each algorithm and each column separately"`. So, we are assuming NIST would be setting the thresholds to get the final results.
So, we are confused on which threshold to use to make the decision in [`identifyTemplate`](https://github.com/usnistgov/frvt/blob/master/1N/src/nullImpl/nullimplfrvt1N.cpp#L106) function. Any help with this will be much appreciated.
Thanks in advance. | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/55/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/55/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/54 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/54/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/54/comments | https://api.github.com/repos/usnistgov/frvt/issues/54/events | https://github.com/usnistgov/frvt/issues/54 | 930,973,985 | MDU6SXNzdWU5MzA5NzM5ODU= | 54 | Ricard Nellis | {
"login": "KingRj100",
"id": 80036876,
"node_id": "MDQ6VXNlcjgwMDM2ODc2",
"avatar_url": "https://avatars.githubusercontent.com/u/80036876?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/KingRj100",
"html_url": "https://github.com/KingRj100",
"followers_url": "https://api.github.com/users/KingRj100/followers",
"following_url": "https://api.github.com/users/KingRj100/following{/other_user}",
"gists_url": "https://api.github.com/users/KingRj100/gists{/gist_id}",
"starred_url": "https://api.github.com/users/KingRj100/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/KingRj100/subscriptions",
"organizations_url": "https://api.github.com/users/KingRj100/orgs",
"repos_url": "https://api.github.com/users/KingRj100/repos",
"events_url": "https://api.github.com/users/KingRj100/events{/privacy}",
"received_events_url": "https://api.github.com/users/KingRj100/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [] | 2021-06-27T16:50:17 | 2021-06-27T16:50:32 | 2021-06-27T16:50:32 | NONE | null | null | null | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/54/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/54/timeline | null | completed |
|
https://api.github.com/repos/usnistgov/frvt/issues/53 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/53/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/53/comments | https://api.github.com/repos/usnistgov/frvt/issues/53/events | https://github.com/usnistgov/frvt/issues/53 | 913,304,897 | MDU6SXNzdWU5MTMzMDQ4OTc= | 53 | Script for FRVT 1:1 testing of multiface input | {
"login": "GivralNguyen",
"id": 64531434,
"node_id": "MDQ6VXNlcjY0NTMxNDM0",
"avatar_url": "https://avatars.githubusercontent.com/u/64531434?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/GivralNguyen",
"html_url": "https://github.com/GivralNguyen",
"followers_url": "https://api.github.com/users/GivralNguyen/followers",
"following_url": "https://api.github.com/users/GivralNguyen/following{/other_user}",
"gists_url": "https://api.github.com/users/GivralNguyen/gists{/gist_id}",
"starred_url": "https://api.github.com/users/GivralNguyen/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/GivralNguyen/subscriptions",
"organizations_url": "https://api.github.com/users/GivralNguyen/orgs",
"repos_url": "https://api.github.com/users/GivralNguyen/repos",
"events_url": "https://api.github.com/users/GivralNguyen/events{/privacy}",
"received_events_url": "https://api.github.com/users/GivralNguyen/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"We've added a test case in the validation package for template generation using multiple image inputs."
] | 2021-06-07T08:51:51 | 2021-06-07T17:24:15 | 2021-06-07T17:23:46 | NONE | null | null | null | Hi,
According to your documentations, a person can be represented by multiple images (and these images will be represented as a single std::vector<Image> . However your in your test scripts for FRVT 1:1, you only run test for one single enroll image - one verification image. Is there any way so we can test multiple enroll image - one/multiple verification image ?
| {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/53/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/53/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/52 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/52/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/52/comments | https://api.github.com/repos/usnistgov/frvt/issues/52/events | https://github.com/usnistgov/frvt/issues/52 | 894,937,182 | MDU6SXNzdWU4OTQ5MzcxODI= | 52 | Range of similarity | {
"login": "cannguyen275",
"id": 35292523,
"node_id": "MDQ6VXNlcjM1MjkyNTIz",
"avatar_url": "https://avatars.githubusercontent.com/u/35292523?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/cannguyen275",
"html_url": "https://github.com/cannguyen275",
"followers_url": "https://api.github.com/users/cannguyen275/followers",
"following_url": "https://api.github.com/users/cannguyen275/following{/other_user}",
"gists_url": "https://api.github.com/users/cannguyen275/gists{/gist_id}",
"starred_url": "https://api.github.com/users/cannguyen275/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/cannguyen275/subscriptions",
"organizations_url": "https://api.github.com/users/cannguyen275/orgs",
"repos_url": "https://api.github.com/users/cannguyen275/repos",
"events_url": "https://api.github.com/users/cannguyen275/events{/privacy}",
"received_events_url": "https://api.github.com/users/cannguyen275/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"No. Any non-negative range is permitted eg. [0,1] or [300,400].\n\n\nFrom: Kei ***@***.***>\nDate: Tuesday, May 18, 2021 at 22:50\nTo: usnistgov/frvt ***@***.***>\nCc: Subscribed ***@***.***>\nSubject: [usnistgov/frvt] Range of similarity (#52)\n\nHello,\n\nAccording to the document, the similarity value is a similarity score resulting from comparison of the templates, on the range [0,\n1000000].\nWe use cosine similarity to calculate the similarity score, the value is on the range [0,1].\nDo we need to normalize that value from [0,1] to [0, 1000000]?\n\nThanks,\n\n—\nYou are receiving this because you are subscribed to this thread.\nReply to this email directly, view it on GitHub<https://github.com/usnistgov/frvt/issues/52>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ADVZFNILBRDYXBK6X3U6BUTTOMRPTANCNFSM45DWHBEA>.\n"
] | 2021-05-19T02:50:18 | 2021-05-19T03:30:14 | 2021-05-19T03:30:14 | NONE | null | null | null | Hello,
According to the document, the similarity value is a similarity score resulting from comparison of the templates, **on the range [0,
1000000].**
We use cosine similarity to calculate the similarity score, the value is on the range [0,1].
Do we need to normalize that value from [0,1] to [0, 1000000]?
Thanks, | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/52/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/52/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/51 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/51/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/51/comments | https://api.github.com/repos/usnistgov/frvt/issues/51/events | https://github.com/usnistgov/frvt/issues/51 | 806,525,907 | MDU6SXNzdWU4MDY1MjU5MDc= | 51 | How to convert the read images to openCV ? | {
"login": "wfcb-85",
"id": 22932011,
"node_id": "MDQ6VXNlcjIyOTMyMDEx",
"avatar_url": "https://avatars.githubusercontent.com/u/22932011?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/wfcb-85",
"html_url": "https://github.com/wfcb-85",
"followers_url": "https://api.github.com/users/wfcb-85/followers",
"following_url": "https://api.github.com/users/wfcb-85/following{/other_user}",
"gists_url": "https://api.github.com/users/wfcb-85/gists{/gist_id}",
"starred_url": "https://api.github.com/users/wfcb-85/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/wfcb-85/subscriptions",
"organizations_url": "https://api.github.com/users/wfcb-85/orgs",
"repos_url": "https://api.github.com/users/wfcb-85/repos",
"events_url": "https://api.github.com/users/wfcb-85/events{/privacy}",
"received_events_url": "https://api.github.com/users/wfcb-85/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"By doing channel = depth/8, it works.",
"Hi @wfcb-85, when I compile and run with :\r\n\r\n cv::Mat(height, width, CV_8UC3, img_data, channel*width);\r\n\r\nit gives the error:\r\n\r\nerror: (-215:Assertion failed) _step >= minstep in function 'Mat'\r\n\r\nI am new to C++, can you or someone help me?",
"Why is channel divided by 3?",
"I have managed to solve the above issue with the following codes: \r\n\r\n```\r\nfor (auto const& face : faces){\r\n int depth = face.depth;\r\n int width = face.width;\r\n int height = face.height;\r\n int channel = depth/8;\r\n uchar* img_data = face.data.get();\r\n auto mat = cv::Mat(height, width, CV_8UC3, img_data);\r\n cout << typeid(mat).name() << endl;\r\n\r\n }\r\n```\r\n\r\nBecause I would need to feed a vector into my embedded python program (to run template inference), how can I convert the Mat object to 3D Array or Vector of uint8 elements?\r\n",
"Surely it's easier and faster not to use Python?\r\n\r\nA Mat is already a 3d array of uint8.\r\n`uint8_t* = mat.data();`\r\n\r\nIf you want to split a Mat object in to separate channels you simply use:\r\n```\r\nstd::vector<cv::Mat> rgb(3);\r\ncv::split(mat, rgb);\r\n```\r\n\r\nAnyway if you need OpenCV support, I'd recommend reading the docs first or asking on their GitHub. This project has nothing to do with OpenCV.",
"My approach is to embed my Python codes in C++, so that the program can execute in Python. Otherwise I wouldn't know how to execute the Python models, which are written in Tensorflow and Pytorch.",
"I'd imagine that scenario isn't very common. Tensorflow and Pytorch are actually written in C++ first and have Python as a wrapper. You can also port the model to ONNX and run it in any framework for inference.\r\n\r\nFor instance, if you typically use PyTorch, you can grab the libtorch include/libs, link to it and then run your model in that. It's far easier and will prevent any troubles with the NIST environment!\r\n\r\nIt also gives you a lot more control to limit the threading in the way they expect.\r\n\r\n\r\nHere is an actual complete example of code required to load, run, execute and retrieve the result of a PyTorch model (eg. Face recognition) in C++\r\n```\r\n#include <torch/script.h>\r\n// Only do this once:\r\nauto model = torch::jit::load(\"filename.pt\");\r\nconstexpr size_t featureSize = 512; // I assume\r\n\r\n// For each image:\r\nfor (auto const& f : faces)\r\n{\r\n auto tensorImage = torch::from_blob(f.data.get(), {(int64_t)f.height, (int64_t)f.width, (int64_t)f.depth / 8});\r\n tensorImage = tensorImage.permute({2, 0, 1}).unsqueeze(0); // WHC -> NCWH\r\n auto o = model({tensorImage});\r\n std::vector<float> feature{o.data_ptr<float>(), o.data_ptr<float>() + featureSize};\r\n}\r\n```",
"@xsacha Thanks for the tip! Though I am still not sure purely C++ is the way to go. I have model for face detection in pytorch, which means I have to apply functions/attributes to C++ to obtain x-max, x-min, y-max, y-min coordinates. The template inference model is in tensorflow, which i believe can be converted to ONNX.\r\n\r\nFor now, I am just trying out embedding Python in C++ and hopefully it pass the speed test.",
"Another C++ question here: Because this is probably linked to data type in face.data.get()\r\n\r\n```\r\n Py_Initialize();\r\n PyObject *pName, *pModule, *pFunc, *pArgs, *pValue;\r\n pName = PyUnicode_FromString((char*)\"extract_template\");\r\n pModule = PyImport_Import(pName);\r\n if (pModule==nullptr){\r\n PyErr_Print();\r\n }\r\n pFunc = PyObject_GetAttrString(pModule, (char*)\"extract\");\r\n\r\n\r\n for (auto const& face : faces){\r\n int depth = face.depth;\r\n int width = face.width;\r\n int height = face.height;\r\n int channel = depth/8;\r\n int sizes[3] = {height,width,channel};\r\n npy_intp dims[3] = {height,width,channel};\r\n uchar* img_data = face.data.get();\r\n auto mat = cv::Mat(3, sizes, CV_8UC3, img_data);\r\n pArgs = PyTuple_Pack(1,PyArray_SimpleNewFromData(3,dims,NPY_UINT8,reinterpret_cast<void*>(mat)));\r\n pValue = PyObject_CallObject(pFunc, pArgs); \r\n```\r\n\r\nI get the following error:\r\n`error: invalid cast from type ‘cv::Mat’ to type ‘void*’`\r\nHow can I take in data from face.data.get() and pass into Python as numpy array?\r\n\r\n\r\n",
"Just do: mat.data if you want the pointer.\r\nAlthough I'm really confused why you guys are creating a cv::Mat from a pointer, only to retrieve the pointer again.\r\n\r\nHuge hint: you do not need OpenCV at all.\r\n\r\n```\r\nfor (const auto&f : faces)\r\n{\r\n pArgs = PyArray_SimpleNewFromData(3,{f.height, f.width, f.depth / 8}, NPY_UINT8, (void*)f.data.get());\r\n pValue = PyObject_CallObject(pFunc, PyTuple_Pack(1, pArgs));\r\n}\r\n```"
] | 2021-02-11T16:02:16 | 2022-10-28T09:58:03 | 2021-02-11T19:02:28 | NONE | null | null | null | Hello,
I'm trying to set the received images in the same format they are if I read them with cv::imread ( is with that format that I train my models). Unfortunately, I'm unable to get correct images, when I save them, they make no sense. The code I'm trying with inside the 'createTemplate' function is the following:
int depth = faces[0].depth;
int width = faces[0].width;
int height = faces[0].height;
int channel = depth/3;
uchar* img_data = faces[0].data.get();
cv::frame = cv::Mat(height, width, CV_8UC3, img_data, channel*width); | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/51/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/51/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/50 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/50/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/50/comments | https://api.github.com/repos/usnistgov/frvt/issues/50/events | https://github.com/usnistgov/frvt/issues/50 | 805,651,587 | MDU6SXNzdWU4MDU2NTE1ODc= | 50 | Testing Limits | {
"login": "MdScntst",
"id": 36072671,
"node_id": "MDQ6VXNlcjM2MDcyNjcx",
"avatar_url": "https://avatars.githubusercontent.com/u/36072671?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/MdScntst",
"html_url": "https://github.com/MdScntst",
"followers_url": "https://api.github.com/users/MdScntst/followers",
"following_url": "https://api.github.com/users/MdScntst/following{/other_user}",
"gists_url": "https://api.github.com/users/MdScntst/gists{/gist_id}",
"starred_url": "https://api.github.com/users/MdScntst/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/MdScntst/subscriptions",
"organizations_url": "https://api.github.com/users/MdScntst/orgs",
"repos_url": "https://api.github.com/users/MdScntst/repos",
"events_url": "https://api.github.com/users/MdScntst/events{/privacy}",
"received_events_url": "https://api.github.com/users/MdScntst/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"The timing limits for 1:1 are listed in Table 1 of the API document: https://pages.nist.gov/frvt/api/FRVT_ongoing_11_api.pdf.\r\n\r\nPlease see the FRVT Evaluation Specifications for hardware specs: https://pages.nist.gov/frvt/api/FRVT_common.pdf."
] | 2021-02-10T16:08:04 | 2021-02-10T16:12:02 | 2021-02-10T16:12:02 | NONE | null | null | null | I've got that submission has to use 1 CPU thread. What are the other limits: inference time, RAM usage, etc?... | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/50/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/50/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/49 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/49/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/49/comments | https://api.github.com/repos/usnistgov/frvt/issues/49/events | https://github.com/usnistgov/frvt/issues/49 | 793,791,835 | MDU6SXNzdWU3OTM3OTE4MzU= | 49 | free(): invalid pointer after trying to load torch model | {
"login": "wfcb-85",
"id": 22932011,
"node_id": "MDQ6VXNlcjIyOTMyMDEx",
"avatar_url": "https://avatars.githubusercontent.com/u/22932011?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/wfcb-85",
"html_url": "https://github.com/wfcb-85",
"followers_url": "https://api.github.com/users/wfcb-85/followers",
"following_url": "https://api.github.com/users/wfcb-85/following{/other_user}",
"gists_url": "https://api.github.com/users/wfcb-85/gists{/gist_id}",
"starred_url": "https://api.github.com/users/wfcb-85/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/wfcb-85/subscriptions",
"organizations_url": "https://api.github.com/users/wfcb-85/orgs",
"repos_url": "https://api.github.com/users/wfcb-85/repos",
"events_url": "https://api.github.com/users/wfcb-85/events{/privacy}",
"received_events_url": "https://api.github.com/users/wfcb-85/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"There's a test in the validation script that ensures you fail when the config directory is missing.\r\n\r\nIt appears that you are failing to load your torch model, as you should, which will return an invalid pointer.\r\n\r\nHowever, you have failed to verify that 1) the model file existed, 2) permissions are correct and most importantly. 3) the model loaded. So when it comes time to clean up, you have attempted to free an invalid pointer.",
"No, the config folder exists.\r\nAs mentioned in my original message, after I change my code and try no more to load the torch model (I set the null implementation as it originally was), the validation script keeps throwing the same error. With the same original code, it was compiling before.\r\nI think there is a problem with the logs the validation script uses.",
"\"Checking for hard-coded config directory\"\r\nWhen this is printed, the script has renamed config to otherConfig. If the config folder still exists here then something is changed or broken in the configuration or source.\r\n\r\n free(): invalid pointer\r\nWhen this is printed, your code has called free to a bogus pointer, observable in GDB. If you still can't see why this happens, look at the backtrace from GDB.",
"Is not my code that has called free to a pointer.\r\nI set the code back to a version that used to pass all the tests.",
"You may go back to an older version where this test wasn't performed but I believe it won't be accepted.\r\nCan you provide the backtrace?",
"This is what I get when I build the original null Implementation:\r\n\r\nChecking for hard-coded config directory free(): invalid pointer\r\nPID 29153 exited due to signal 6\r\nfree(): invalid pointer\r\nscripts/run_testdriver.sh: line 68: 29512 Aborted",
"You'll need to run it with gdb, such as:\r\ngdb --args myapp\r\nYou can add it to the script.\r\nThen type 'b' and enter once it crashes.\r\nThis should show you which line it crashed on.\r\n\r\nIf you clean the config / otherConfig, I assume your crash at that point goes away. So you should do this in your code instead of the null implementation.",
"Ok, I had to disable the fork() part of the main() of validate11.cpp in order to debug it entirely.\r\nThe error is coming from line 86:\r\nauto ret = implPtr->createTemplate(faces, role, templ, eyes);",
"The error persists even when I do nothing on the 'createTemplate' function and just return the Success status.",
"Hello,\r\nThe problem is solved when I include the same libraries (Torch and Opencv) at the validate11 side, that I used for my implementation. I wonder, will that be a problem when NIST will evaluate my library?",
"You mean you had used different versions of the libraries when running validate11? The correct libraries should be included in your implementation's lib/ path. It all gets zipped (lib, config, doc) for your submission.",
"I linked those libraries statically and the problem is solved. Thank you."
] | 2021-01-25T23:08:00 | 2021-02-01T22:25:44 | 2021-02-01T22:25:44 | NONE | null | null | null | Hello,
After I was trying to load a torch model, and it failed, I get every time I execute run-validate_11.sh an invalid pointer error. The error persists even when I set the code as it originally was and build it again. For info, I'm building the null implementation.
You can find below the message of the error:
-------------------------
Running 1:1 validation
-------------------------
Checking for hard-coded config directory free(): invalid pointer
PID 142393 exited due to signal 6
free(): invalid pointer
scripts/run_testdriver.sh line 68: 142392 Aborted | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/49/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/49/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/48 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/48/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/48/comments | https://api.github.com/repos/usnistgov/frvt/issues/48/events | https://github.com/usnistgov/frvt/issues/48 | 792,052,980 | MDU6SXNzdWU3OTIwNTI5ODA= | 48 | Can I load the models in the constructor? | {
"login": "wfcb-85",
"id": 22932011,
"node_id": "MDQ6VXNlcjIyOTMyMDEx",
"avatar_url": "https://avatars.githubusercontent.com/u/22932011?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/wfcb-85",
"html_url": "https://github.com/wfcb-85",
"followers_url": "https://api.github.com/users/wfcb-85/followers",
"following_url": "https://api.github.com/users/wfcb-85/following{/other_user}",
"gists_url": "https://api.github.com/users/wfcb-85/gists{/gist_id}",
"starred_url": "https://api.github.com/users/wfcb-85/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/wfcb-85/subscriptions",
"organizations_url": "https://api.github.com/users/wfcb-85/orgs",
"repos_url": "https://api.github.com/users/wfcb-85/repos",
"events_url": "https://api.github.com/users/wfcb-85/events{/privacy}",
"received_events_url": "https://api.github.com/users/wfcb-85/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"The initialize() function is called in the testdriver provided in the validation package - https://github.com/usnistgov/frvt/blob/master/11/src/testdriver/validate11.cpp\r\n\r\nDid you follow the instructions contained in the README.md file on putting your library into ./lib? Or did you optionally build the null sdk? Did you successfully compile and link your library against the provided testdriver - e.g. did you see any errors when you ran run_validate_11.sh? Additionally, the path to the configuration directory that you will read your models from is passed in as a parameter to initialize(). Please do not hard-code your configuration directory, because it will not be in the same location on our systems.",
"It works when I build the null sdk."
] | 2021-01-22T14:35:06 | 2021-01-25T23:03:17 | 2021-01-25T23:03:17 | NONE | null | null | null | Hello,
When I execute run_validate.sh, it does not seem to be calling the initialize function. I don't see the messages I print nor a change of the values of the variables. Would it be a problem if I loaded the model files within the constructor? | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/48/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/48/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/47 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/47/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/47/comments | https://api.github.com/repos/usnistgov/frvt/issues/47/events | https://github.com/usnistgov/frvt/issues/47 | 749,389,575 | MDU6SXNzdWU3NDkzODk1NzU= | 47 | Method to convert python model to c++ | {
"login": "quocnhat",
"id": 19263564,
"node_id": "MDQ6VXNlcjE5MjYzNTY0",
"avatar_url": "https://avatars.githubusercontent.com/u/19263564?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/quocnhat",
"html_url": "https://github.com/quocnhat",
"followers_url": "https://api.github.com/users/quocnhat/followers",
"following_url": "https://api.github.com/users/quocnhat/following{/other_user}",
"gists_url": "https://api.github.com/users/quocnhat/gists{/gist_id}",
"starred_url": "https://api.github.com/users/quocnhat/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/quocnhat/subscriptions",
"organizations_url": "https://api.github.com/users/quocnhat/orgs",
"repos_url": "https://api.github.com/users/quocnhat/repos",
"events_url": "https://api.github.com/users/quocnhat/events{/privacy}",
"received_events_url": "https://api.github.com/users/quocnhat/received_events",
"type": "User",
"site_admin": false
} | [] | open | false | null | [] | null | [
"This heavily depends on your deep learning framework. Please provide more details.",
"Thanks for your reply, Sorry for my mistake. My model using pytorch framework. By using NCNN to port pytorch model tobe able to read in C++, I can not utilize Resnet 50 because of the big inference time. But other teams can. May I ask for the solution? ",
"I would simply conver to onnx format and run it directly with c++ onnx interface or (easier) with opencv deep learning inference.\r\n\r\nAny of solutions above will get you there.",
"Thanks very much, But ONNX (NCNN->ONNX) is exactly what I did. It does not help for large models like ResNet 50 or 100 due to big inference time. I will take the time for opencv deep learning interface solution. Thanks",
"What is your execution time? Now you can go up to 1.5secs, and I am sure that the opencv solution will work even in single thread mode.",
"Sorry, but We can not estimate execution time exactly up on Nist's server. The last submission (model verification Resnet 50) failed because of the overloading execution time (they did not feedback the execution time for the failed submission). The Resnet 18 is accepted with 0.675(s). (Assume that detection pharse need 0.5s and it remains 1s for verification)",
"0.5s for detection seems high to me. Which network are you using?",
"The detection net is the modified version of the mobilenetv2(backbone) + retinaFace(head) and can be changed somehow. But let's assume the verification Net remain 1s for inference. By using onnx, Resnet 50( or 101) seems overloading execution time. Opencv deep learning still on my schedule. ",
"You should be able to test performance by using a Xeon processor (AVX-512 compatible) or Core i7 (AVX2 compatible) and building your project without threading or targeting a single core.\r\n\r\nFrom my own testing on the sample images provided in this Github:\r\nMobileNetV2 detector + 101 layer resnet should take approximately 5 minutes to detect and generate models for the 653 enrolment images using off-the-shelf frameworks such as ONNXRuntime or Torch. Or roughly half a second per image.\r\nI was also able to run a custom DetNAS + modified 150 layer resnet in 6 minutes.\r\n\r\nSpeed should not be your limitation. The detector is most certainly the fast bit.\r\n\r\nIf you are taking longer than this, I would profile the app using CPU profiler (Linux: perf record, Windows: Visual Studio) and have a look at where your CPU time is being spent. My guess is it is not spent generating models, but perhaps setting up or manipulating the images. Especially as you seem unfamiliar with C++, you may not recognise the redundant operations and how they may affect performance. Do you have anyone at your company that specialises in C++?",
"> You should be able to test performance by using a Xeon processor (AVX-512 compatible) or Core i7 (AVX2 compatible) and building your project without threading or targeting a single core.\r\n> \r\n> From my own testing on the sample images provided in this Github:\r\n> MobileNetV2 detector + 101 layer resnet should take approximately 5 minutes to detect and generate models for the 653 enrolment images using off-the-shelf frameworks such as ONNXRuntime or Torch. Or roughly half a second per image.\r\n> I was also able to run a custom DetNAS + modified 150 layer resnet in 6 minutes.\r\n> \r\n> Speed should not be your limitation. The detector is most certainly the fast bit.\r\n> \r\n> If you are taking longer than this, I would profile the app using CPU profiler (Linux: perf record, Windows: Visual Studio) and have a look at where your CPU time is being spent. My guess is it is not spent generating models, but perhaps setting up or manipulating the images. Especially as you seem unfamiliar with C++, you may not recognise the redundant operations and how they may affect performance. Do you have anyone at your company that specialises in C++?\r\n\r\nThank you for your information. Would you mind providing more details about your configs, e.g. input image size? \r\nAt present, it takes me about 600ms to run Retinaface for only face detection at size of 480x480. ",
"@cao-nv Make sure your detector isn't unnecessarily deep. Most the images in this test are frontal. It sounds like you're using a resnet 50 retinaface judging by the timings. If that's on mobilenet, you are doing too much work somewhere and I'd refer back to my suggestion to profile the app.\r\nI target 640x640 resolution btw.",
"> @cao-nv Make sure your detector isn't unnecessarily deep. Most the images in this test are frontal. It sounds like you're using a resnet 50 retinaface judging by the timings. If that's on mobilenet, you are doing too much work somewhere and I'd refer back to my suggestion to profile the app.\r\n> I target 640x640 resolution btw.\r\n\r\nYes, I used Resnet50. Actually, this is my first try for FRVT. I'll consider backbone. Thank you a lot. ",
"@cao-nv Looks like your model successfully ran in just under 0.5 seconds and without much memory usage. Good work :)\r\nSame with yours @quocnhat.\r\nTime quota is 1.5 seconds so as you can see, there's plenty to play with there.\r\n\r\nIssue can probably be closed now.",
"Actually, I couldn't run RetinaFace with ResNet-50 as its backbone with he input size of 640x640 in less than a seconds. I chose to reduce the input size to 480x480, then it took about 1 second just for detecting faces. ",
"Might just be your machine then because the NIST results show your company doing under 0.5 seconds.",
"> Might just be your machine then because the NIST results show your company doing under 0.5 seconds.\r\n\r\nThat result was our first submission. We are preparing for the next time. \r\nYou are right about the machine, I compiled and tested everything on a laptop with Intel core-i5 10th gen which supports AVX2 but not AVX512 as Xeon E5 CPUs. I don't know if the core-i5 is much slower than Xeon E5. ",
"I have the similar question about convert python inference to c++ inference. Is there any way to not modify the python inference code? I want to use C++ to call python inference. Could anyone give me some advises?",
"> Hi Nist's participants, I have successfully submitted my solutions to Nist for several months. NCNN is the way to convert the python model to c++, but it quite slow for the inference. I find that some teams submitted a larger model ( Resnet 50 or Resnet 100,...), May I ask for the best method to port the python model to inference in c++?\r\n\r\nHi there, How do you convert const FRVT::Image &image to the ncnn:Mat format?",
"Hi @RamatovInomjon \r\n\r\nFrom https://github.com/Tencent/ncnn/blob/master/src/mat.h there is a method you can use to create from_pixels:\r\n\r\n static Mat from_pixels(const unsigned char* pixels, int type, int w, int h, Allocator* allocator = 0);\r\n\r\nThis should be enough.\r\n\r\nBest,\r\n/M",
"> \r\n@mlourencoeb \r\nthanks for replying! \r\nI'm using this, ncnn::Mat enroll_ncnn_img = ncnn::Mat::from_pixels(image.data.get(), ncnn::Mat::PIXEL_RGB, image.width, image.height);\r\n\r\nwhen i save this image \r\n![wface-11](https://user-images.githubusercontent.com/43458270/228139379-42ff9edf-9121-4b29-8c7b-1c97802bb835.jpg)\r\n\r\nactually image is \r\n\r\n![rotated90](https://user-images.githubusercontent.com/43458270/228139867-96ef76aa-9a4a-44b2-a4a1-3256b3220482.jpg)\r\n",
"> I have the similar question about convert python inference to c++ inference. Is there any way to not modify the python inference code? I want to use C++ to call python inference. Could anyone give me some advises?\r\n\r\nyou may use pybind11 or simply `Python.h` to call python runtime."
] | 2020-11-24T06:36:55 | 2023-12-21T15:42:48 | null | NONE | null | null | null | Hi Nist's participants, I have successfully submitted my solutions to Nist for several months. NCNN is the way to convert the python model to c++, but it quite slow for the inference. I find that some teams submitted a larger model ( Resnet 50 or Resnet 100,...), May I ask for the best method to port the python model to inference in c++? | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/47/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/47/timeline | null | null |
https://api.github.com/repos/usnistgov/frvt/issues/46 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/46/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/46/comments | https://api.github.com/repos/usnistgov/frvt/issues/46/events | https://github.com/usnistgov/frvt/issues/46 | 727,855,405 | MDU6SXNzdWU3Mjc4NTU0MDU= | 46 | How to confirm which pictures are the same person | {
"login": "aliqing0122",
"id": 13270978,
"node_id": "MDQ6VXNlcjEzMjcwOTc4",
"avatar_url": "https://avatars.githubusercontent.com/u/13270978?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/aliqing0122",
"html_url": "https://github.com/aliqing0122",
"followers_url": "https://api.github.com/users/aliqing0122/followers",
"following_url": "https://api.github.com/users/aliqing0122/following{/other_user}",
"gists_url": "https://api.github.com/users/aliqing0122/gists{/gist_id}",
"starred_url": "https://api.github.com/users/aliqing0122/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/aliqing0122/subscriptions",
"organizations_url": "https://api.github.com/users/aliqing0122/orgs",
"repos_url": "https://api.github.com/users/aliqing0122/repos",
"events_url": "https://api.github.com/users/aliqing0122/events{/privacy}",
"received_events_url": "https://api.github.com/users/aliqing0122/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"The purpose of the validation package is to ensure your submission adheres to the API and tests a bunch of edge cases to make sure your software doesn't crash. The images published aren't necessarily used to test the accuracy of your software, so we do not publish any sort of ground truth for the identities."
] | 2020-10-23T03:06:56 | 2020-10-23T13:56:53 | 2020-10-23T13:56:53 | NONE | null | null | null | Hi @mlngan how do i categorize the pictures provided on github(https://github.com/usnistgov/frvt/tree/master/common/images)?
How do i know which pictures are the same person? | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/46/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/46/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/45 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/45/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/45/comments | https://api.github.com/repos/usnistgov/frvt/issues/45/events | https://github.com/usnistgov/frvt/issues/45 | 703,957,753 | MDU6SXNzdWU3MDM5NTc3NTM= | 45 | Waiting for 1:1 results.. | {
"login": "deimsdeutsch",
"id": 39468654,
"node_id": "MDQ6VXNlcjM5NDY4NjU0",
"avatar_url": "https://avatars.githubusercontent.com/u/39468654?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/deimsdeutsch",
"html_url": "https://github.com/deimsdeutsch",
"followers_url": "https://api.github.com/users/deimsdeutsch/followers",
"following_url": "https://api.github.com/users/deimsdeutsch/following{/other_user}",
"gists_url": "https://api.github.com/users/deimsdeutsch/gists{/gist_id}",
"starred_url": "https://api.github.com/users/deimsdeutsch/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/deimsdeutsch/subscriptions",
"organizations_url": "https://api.github.com/users/deimsdeutsch/orgs",
"repos_url": "https://api.github.com/users/deimsdeutsch/repos",
"events_url": "https://api.github.com/users/deimsdeutsch/events{/privacy}",
"received_events_url": "https://api.github.com/users/deimsdeutsch/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"Email us: frvt@nist.gov"
] | 2020-09-17T22:48:13 | 2020-09-17T23:02:44 | 2020-09-17T23:02:44 | NONE | null | null | null | Its been more than 10 days and we haven't received results for 1:1 can you please give some clarity as you have moved to a different OS and we would have to migrate for submitting 1:N submission. Very confused. Please update. | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/45/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/45/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/44 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/44/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/44/comments | https://api.github.com/repos/usnistgov/frvt/issues/44/events | https://github.com/usnistgov/frvt/issues/44 | 700,654,622 | MDU6SXNzdWU3MDA2NTQ2MjI= | 44 | Incorrect path | {
"login": "kotbegemot",
"id": 958279,
"node_id": "MDQ6VXNlcjk1ODI3OQ==",
"avatar_url": "https://avatars.githubusercontent.com/u/958279?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/kotbegemot",
"html_url": "https://github.com/kotbegemot",
"followers_url": "https://api.github.com/users/kotbegemot/followers",
"following_url": "https://api.github.com/users/kotbegemot/following{/other_user}",
"gists_url": "https://api.github.com/users/kotbegemot/gists{/gist_id}",
"starred_url": "https://api.github.com/users/kotbegemot/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/kotbegemot/subscriptions",
"organizations_url": "https://api.github.com/users/kotbegemot/orgs",
"repos_url": "https://api.github.com/users/kotbegemot/repos",
"events_url": "https://api.github.com/users/kotbegemot/events{/privacy}",
"received_events_url": "https://api.github.com/users/kotbegemot/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"Unsure why tempConfigDir would be an empty path as it works on my end. But I've reverted back to not appending $(pwd) in front of the config dirs, so it should work now. New OS requirement is now CentOS 8.2. b18c70f"
] | 2020-09-13T21:47:30 | 2020-09-14T10:14:17 | 2020-09-14T10:14:01 | NONE | null | null | null | When the [script]( https://github.com/usnistgov/frvt/blob/master/11/scripts/run_testdriver.sh
) is running, an empty path is passed.
The system centos 7.6 did not repeat itself.
```bash
echo -n "Checking for hard-coded config directory "
tempConfigDir=$(pwd)/otherConfig
chmod 775 $configDir; mv $configDir $tempConfigDir; chmod 550 $tempConfigDir
bin/validate11 createTemplate -x enroll -c $tempConfigDir -o $outputDir -h $outputStem -i $inputFile -t $numForks -j $templatesDir
retEnroll=$?
```
@mlngan | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/44/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/44/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/43 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/43/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/43/comments | https://api.github.com/repos/usnistgov/frvt/issues/43/events | https://github.com/usnistgov/frvt/issues/43 | 695,622,332 | MDU6SXNzdWU2OTU2MjIzMzI= | 43 | update scripts for a centos 8.2 | {
"login": "kotbegemot",
"id": 958279,
"node_id": "MDQ6VXNlcjk1ODI3OQ==",
"avatar_url": "https://avatars.githubusercontent.com/u/958279?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/kotbegemot",
"html_url": "https://github.com/kotbegemot",
"followers_url": "https://api.github.com/users/kotbegemot/followers",
"following_url": "https://api.github.com/users/kotbegemot/following{/other_user}",
"gists_url": "https://api.github.com/users/kotbegemot/gists{/gist_id}",
"starred_url": "https://api.github.com/users/kotbegemot/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/kotbegemot/subscriptions",
"organizations_url": "https://api.github.com/users/kotbegemot/orgs",
"repos_url": "https://api.github.com/users/kotbegemot/repos",
"events_url": "https://api.github.com/users/kotbegemot/events{/privacy}",
"received_events_url": "https://api.github.com/users/kotbegemot/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"We'll be pushing out updates for CentOS 8.2 this week."
] | 2020-09-08T07:20:45 | 2020-09-08T11:39:11 | 2020-09-08T11:39:11 | NONE | null | null | null | When will the scripts be updated for Centos 8.2? @mlngan
| {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/43/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/43/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/42 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/42/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/42/comments | https://api.github.com/repos/usnistgov/frvt/issues/42/events | https://github.com/usnistgov/frvt/issues/42 | 680,777,070 | MDU6SXNzdWU2ODA3NzcwNzA= | 42 | Code in Python | {
"login": "fisakhan",
"id": 48552619,
"node_id": "MDQ6VXNlcjQ4NTUyNjE5",
"avatar_url": "https://avatars.githubusercontent.com/u/48552619?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/fisakhan",
"html_url": "https://github.com/fisakhan",
"followers_url": "https://api.github.com/users/fisakhan/followers",
"following_url": "https://api.github.com/users/fisakhan/following{/other_user}",
"gists_url": "https://api.github.com/users/fisakhan/gists{/gist_id}",
"starred_url": "https://api.github.com/users/fisakhan/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/fisakhan/subscriptions",
"organizations_url": "https://api.github.com/users/fisakhan/orgs",
"repos_url": "https://api.github.com/users/fisakhan/repos",
"events_url": "https://api.github.com/users/fisakhan/events{/privacy}",
"received_events_url": "https://api.github.com/users/fisakhan/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"The NIST FRVT API would require that you either translate or wrap your Python code with C++. We have received submissions where developers have successfully done this."
] | 2020-08-18T07:49:05 | 2020-08-18T11:56:58 | 2020-08-18T11:56:58 | NONE | null | null | null | In order to submit to NIST FRVT, do we need to transfer our code from Python and TensorFlow to C++? Is it possible to call and execute the Python code from within the NIST FRVT C++ API? | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/42/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/42/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/41 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/41/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/41/comments | https://api.github.com/repos/usnistgov/frvt/issues/41/events | https://github.com/usnistgov/frvt/issues/41 | 676,176,505 | MDU6SXNzdWU2NzYxNzY1MDU= | 41 | Twice call to distructor ~NullImplFRVT11() | {
"login": "fisakhan",
"id": 48552619,
"node_id": "MDQ6VXNlcjQ4NTUyNjE5",
"avatar_url": "https://avatars.githubusercontent.com/u/48552619?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/fisakhan",
"html_url": "https://github.com/fisakhan",
"followers_url": "https://api.github.com/users/fisakhan/followers",
"following_url": "https://api.github.com/users/fisakhan/following{/other_user}",
"gists_url": "https://api.github.com/users/fisakhan/gists{/gist_id}",
"starred_url": "https://api.github.com/users/fisakhan/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/fisakhan/subscriptions",
"organizations_url": "https://api.github.com/users/fisakhan/orgs",
"repos_url": "https://api.github.com/users/fisakhan/repos",
"events_url": "https://api.github.com/users/fisakhan/events{/privacy}",
"received_events_url": "https://api.github.com/users/fisakhan/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"The testdriver forks child processes to do work, and the destructor gets called once for each child process + once for the parent process. We use this approach in our actual test, so you will need to address this."
] | 2020-08-10T14:23:19 | 2020-08-18T12:15:32 | 2020-08-18T12:15:32 | NONE | null | null | null | The NIST API calls ~NullImplFRVT11() two times after each call to "bin/validate11 createTemplate -x enroll -c $configDir -o $outputDir -h $outputStem -i $inputFile -t $numForks -j $templatesDir retEnroll=$?" Both the destructors use different process ids. This does not allow my program to release memory. I expect only one call to ~NullImplFRVT11(). Whats the problem and how to restrict processes to only one? | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/41/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/41/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/40 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/40/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/40/comments | https://api.github.com/repos/usnistgov/frvt/issues/40/events | https://github.com/usnistgov/frvt/issues/40 | 672,348,924 | MDU6SXNzdWU2NzIzNDg5MjQ= | 40 | Two submissions before the four months | {
"login": "mlourencoeb",
"id": 24304201,
"node_id": "MDQ6VXNlcjI0MzA0MjAx",
"avatar_url": "https://avatars.githubusercontent.com/u/24304201?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/mlourencoeb",
"html_url": "https://github.com/mlourencoeb",
"followers_url": "https://api.github.com/users/mlourencoeb/followers",
"following_url": "https://api.github.com/users/mlourencoeb/following{/other_user}",
"gists_url": "https://api.github.com/users/mlourencoeb/gists{/gist_id}",
"starred_url": "https://api.github.com/users/mlourencoeb/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/mlourencoeb/subscriptions",
"organizations_url": "https://api.github.com/users/mlourencoeb/orgs",
"repos_url": "https://api.github.com/users/mlourencoeb/repos",
"events_url": "https://api.github.com/users/mlourencoeb/events{/privacy}",
"received_events_url": "https://api.github.com/users/mlourencoeb/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"For the FRVT 1:1, 1:N, and Quality tracks, participants may send **one** submission as often as every four calendar months from the last submission for evaluation. We are no longer allowing two submissions for first-time participants.\r\n",
"Thanks for the quick updade. Appreciate it!",
"@mlngan What if my submission has an error/problem in code, for example, generating too many threads. Do you still start counting 4 month for my submission? Or you start counting 4 month when my submission is running successfully on NIST's end?",
"You will be allowed to submit your next submission 4 months from the date you send us a successfully running algorithm."
] | 2020-08-03T21:19:14 | 2021-04-06T14:08:31 | 2020-08-03T21:23:39 | NONE | null | null | null | Hello,
This is not really an issue, it is more a doubt. I believe last year it was possible to submit two versions of algorithms, and then every submission would need to wait for four months until a new one is done. I did it in the past company, so I'm sure about this.
Is this rule currently valid?
Thanks for the clarification.
Best,
Miguel | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/40/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/40/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/39 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/39/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/39/comments | https://api.github.com/repos/usnistgov/frvt/issues/39/events | https://github.com/usnistgov/frvt/issues/39 | 667,655,012 | MDU6SXNzdWU2Njc2NTUwMTI= | 39 | Error authenticate the Nist key | {
"login": "quocnhat",
"id": 19263564,
"node_id": "MDQ6VXNlcjE5MjYzNTY0",
"avatar_url": "https://avatars.githubusercontent.com/u/19263564?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/quocnhat",
"html_url": "https://github.com/quocnhat",
"followers_url": "https://api.github.com/users/quocnhat/followers",
"following_url": "https://api.github.com/users/quocnhat/following{/other_user}",
"gists_url": "https://api.github.com/users/quocnhat/gists{/gist_id}",
"starred_url": "https://api.github.com/users/quocnhat/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/quocnhat/subscriptions",
"organizations_url": "https://api.github.com/users/quocnhat/orgs",
"repos_url": "https://api.github.com/users/quocnhat/repos",
"events_url": "https://api.github.com/users/quocnhat/events{/privacy}",
"received_events_url": "https://api.github.com/users/quocnhat/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"Your public encryption key is the result of step 2 of the process, which should be named forfrvt11.gpg. You need to email that file to frvt@nist.gov.",
"I need to re-submit my solution to Nist. But unfortunately, my default key (private key) disappears. Can I generate another key and resubmit, including forfrvt11.gpg? It is definitely different from the one printed in the agreement ( already sent by post).\r\nThank you",
"If you generate a new key, you will have to send us an updated participation agreement with the new key fingerprint and email us the new gpg file."
] | 2020-07-29T08:30:59 | 2020-08-05T12:55:39 | 2020-07-29T12:23:46 | NONE | null | null | null | I am confusing at email my public key to Nist, Don't really understand here. What is it and how do I know my public key is sent to Nist successfully?
![image](https://user-images.githubusercontent.com/19263564/88777491-79bbf880-d1b1-11ea-8c39-927a88291729.png)
| {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/39/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/39/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/141 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/141/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/141/comments | https://api.github.com/repos/usnistgov/frvt/issues/141/events | https://github.com/usnistgov/frvt/issues/141 | 2,432,289,049 | I_kwDOBIKFiM6Q-ckZ | 141 | Possibility to relax Ubuntu 20.04 minor version requirement | {
"login": "alexkrz",
"id": 38694722,
"node_id": "MDQ6VXNlcjM4Njk0NzIy",
"avatar_url": "https://avatars.githubusercontent.com/u/38694722?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/alexkrz",
"html_url": "https://github.com/alexkrz",
"followers_url": "https://api.github.com/users/alexkrz/followers",
"following_url": "https://api.github.com/users/alexkrz/following{/other_user}",
"gists_url": "https://api.github.com/users/alexkrz/gists{/gist_id}",
"starred_url": "https://api.github.com/users/alexkrz/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/alexkrz/subscriptions",
"organizations_url": "https://api.github.com/users/alexkrz/orgs",
"repos_url": "https://api.github.com/users/alexkrz/repos",
"events_url": "https://api.github.com/users/alexkrz/events{/privacy}",
"received_events_url": "https://api.github.com/users/alexkrz/received_events",
"type": "User",
"site_admin": false
} | [] | open | false | null | [] | null | [] | 2024-07-26T13:50:07 | 2024-07-26T13:50:07 | null | NONE | null | null | null | Hi,
I am wondering whether it is necessary to specifically check for Ubuntu version 20.04.3 LTS.
The currently supported version is Ubuntu 20.04.6 and contains lots of important security updates.
Since the minor version changes should not have a relevant impact on the included packages, I would advocate to allow all Ubuntu 20.04.x versions.
I will create a PR and hope that the PR can be accepted.
Kind regards | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/141/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/141/timeline | null | null |
https://api.github.com/repos/usnistgov/frvt/issues/139 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/139/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/139/comments | https://api.github.com/repos/usnistgov/frvt/issues/139/events | https://github.com/usnistgov/frvt/issues/139 | 2,408,085,118 | I_kwDOBIKFiM6PiHZ- | 139 | incorrect path to some images for age-estimation | {
"login": "xerus",
"id": 2479721,
"node_id": "MDQ6VXNlcjI0Nzk3MjE=",
"avatar_url": "https://avatars.githubusercontent.com/u/2479721?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/xerus",
"html_url": "https://github.com/xerus",
"followers_url": "https://api.github.com/users/xerus/followers",
"following_url": "https://api.github.com/users/xerus/following{/other_user}",
"gists_url": "https://api.github.com/users/xerus/gists{/gist_id}",
"starred_url": "https://api.github.com/users/xerus/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/xerus/subscriptions",
"organizations_url": "https://api.github.com/users/xerus/orgs",
"repos_url": "https://api.github.com/users/xerus/repos",
"events_url": "https://api.github.com/users/xerus/events{/privacy}",
"received_events_url": "https://api.github.com/users/xerus/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [] | 2024-07-15T07:46:38 | 2024-07-15T20:03:52 | 2024-07-15T20:03:52 | CONTRIBUTOR | null | null | null | Hi, the commit 18038d652b950f919f53957b9ddda8d34fd7d04b add folder "videos" to all image paths in the singleMedia.txt file, however some images are in the `common/images/face/` folder
eg:
`../common/images/face/videos/small.ppm` should be `../common/images/face/small.ppm` | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/139/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/139/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/138 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/138/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/138/comments | https://api.github.com/repos/usnistgov/frvt/issues/138/events | https://github.com/usnistgov/frvt/issues/138 | 2,386,294,354 | I_kwDOBIKFiM6OO_ZS | 138 | Unexpected path of `video1` for `run_validate_ae.sh` | {
"login": "paplhjak",
"id": 52961799,
"node_id": "MDQ6VXNlcjUyOTYxNzk5",
"avatar_url": "https://avatars.githubusercontent.com/u/52961799?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/paplhjak",
"html_url": "https://github.com/paplhjak",
"followers_url": "https://api.github.com/users/paplhjak/followers",
"following_url": "https://api.github.com/users/paplhjak/following{/other_user}",
"gists_url": "https://api.github.com/users/paplhjak/gists{/gist_id}",
"starred_url": "https://api.github.com/users/paplhjak/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/paplhjak/subscriptions",
"organizations_url": "https://api.github.com/users/paplhjak/orgs",
"repos_url": "https://api.github.com/users/paplhjak/repos",
"events_url": "https://api.github.com/users/paplhjak/events{/privacy}",
"received_events_url": "https://api.github.com/users/paplhjak/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"We've updated the file with the correct path. Thank you."
] | 2024-07-02T13:40:03 | 2024-07-02T18:27:45 | 2024-07-02T18:27:45 | NONE | null | null | null | When running `./run_validate_ae.sh`, the execution fails with:
![image](https://github.com/usnistgov/frvt/assets/52961799/ba30b30c-6a18-4efd-ad18-88b48591c287)
This can be fixed by moving the `video1` folder down by one directory, however, this should not be expected of the user. | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/138/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/138/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/137 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/137/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/137/comments | https://api.github.com/repos/usnistgov/frvt/issues/137/events | https://github.com/usnistgov/frvt/issues/137 | 2,381,506,997 | I_kwDOBIKFiM6N8um1 | 137 | Disparity of median enrollment time when running on single core. | {
"login": "rishidd11",
"id": 85778019,
"node_id": "MDQ6VXNlcjg1Nzc4MDE5",
"avatar_url": "https://avatars.githubusercontent.com/u/85778019?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/rishidd11",
"html_url": "https://github.com/rishidd11",
"followers_url": "https://api.github.com/users/rishidd11/followers",
"following_url": "https://api.github.com/users/rishidd11/following{/other_user}",
"gists_url": "https://api.github.com/users/rishidd11/gists{/gist_id}",
"starred_url": "https://api.github.com/users/rishidd11/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/rishidd11/subscriptions",
"organizations_url": "https://api.github.com/users/rishidd11/orgs",
"repos_url": "https://api.github.com/users/rishidd11/repos",
"events_url": "https://api.github.com/users/rishidd11/events{/privacy}",
"received_events_url": "https://api.github.com/users/rishidd11/received_events",
"type": "User",
"site_admin": false
} | [] | closed | false | null | [] | null | [
"During our timing test, we only run with a single process (and limit the number of CPUs to one). We run and time the template generation call over many iterations, then take the median of those timing values."
] | 2024-06-29T05:28:04 | 2024-07-01T11:22:00 | 2024-07-01T11:21:59 | NONE | null | null | null | The NIST document specifies that the median enrollment time must be under 1.5 seconds on a single core, tested on an Intel(R) Xeon(R) Gold 6140 CPU @ 2.30GHz. In the run_testdriver.sh script, the num_forks argument is used to run validate11.cpp.
When I run the script on a single core using the command taskset --cpu-list 1 run_validate11.sh, the enrollment time for multiple processes (i.e., num_forks = 4) increases fourfold compared to a single process. However, there is no increase in time for multiple processes when I run the script without the taskset command.
How does NIST internally check the median enrollment time? Do they use the taskset command to limit the number of CPUs to one? | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/137/reactions",
"total_count": 0,
"+1": 0,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/137/timeline | null | completed |
https://api.github.com/repos/usnistgov/frvt/issues/136 | https://api.github.com/repos/usnistgov/frvt | https://api.github.com/repos/usnistgov/frvt/issues/136/labels{/name} | https://api.github.com/repos/usnistgov/frvt/issues/136/comments | https://api.github.com/repos/usnistgov/frvt/issues/136/events | https://github.com/usnistgov/frvt/issues/136 | 2,352,111,098 | I_kwDOBIKFiM6MMl36 | 136 | Suggest to split and zip "common/images/face" to release files | {
"login": "ImNotMT",
"id": 13098079,
"node_id": "MDQ6VXNlcjEzMDk4MDc5",
"avatar_url": "https://avatars.githubusercontent.com/u/13098079?v=4",
"gravatar_id": "",
"url": "https://api.github.com/users/ImNotMT",
"html_url": "https://github.com/ImNotMT",
"followers_url": "https://api.github.com/users/ImNotMT/followers",
"following_url": "https://api.github.com/users/ImNotMT/following{/other_user}",
"gists_url": "https://api.github.com/users/ImNotMT/gists{/gist_id}",
"starred_url": "https://api.github.com/users/ImNotMT/starred{/owner}{/repo}",
"subscriptions_url": "https://api.github.com/users/ImNotMT/subscriptions",
"organizations_url": "https://api.github.com/users/ImNotMT/orgs",
"repos_url": "https://api.github.com/users/ImNotMT/repos",
"events_url": "https://api.github.com/users/ImNotMT/events{/privacy}",
"received_events_url": "https://api.github.com/users/ImNotMT/received_events",
"type": "User",
"site_admin": false
} | [] | open | false | null | [] | null | [
"`git clone --depth 1` may lower the size of files",
"thank you for the approach. but for my trying, still have 7.7G. the images and video size is 5.7G, cannot avoid that huge .git folder. ",
"I think basically caused by 4.7G videos uploaded recently by this commit 7ca40ad12d548232e88aecb167934bac1e659f29."
] | 2024-06-13T21:39:27 | 2024-06-14T09:25:08 | null | NONE | null | null | null | now git clone the frvt repo would create 10G+ .git files (files in frvt repo not counting yet), while the scripts and source code files only a few KBs.
it is not a good idea to git clone the whole frvt repo while you are mixing the data files (no version control/history required) and source files in git. | {
"url": "https://api.github.com/repos/usnistgov/frvt/issues/136/reactions",
"total_count": 3,
"+1": 3,
"-1": 0,
"laugh": 0,
"hooray": 0,
"confused": 0,
"heart": 0,
"rocket": 0,
"eyes": 0
} | https://api.github.com/repos/usnistgov/frvt/issues/136/timeline | null | null |