|
--- |
|
{} |
|
--- |
|
# AM-RADIO: Reduce All Domains Into One |
|
|
|
Mike Ranzinger, Greg Heinrich, Jan Kautz, Pavlo Molchanov |
|
|
|
[NVIDIA Research](https://www.nvidia.com/en-us/research/) |
|
|
|
\[[AM-RADIO Paper](https://arxiv.org/abs/2312.06709)\] |
|
\[[PHI-S Paper](https://arxiv.org/abs/2410.01680)\] |
|
\[[BibTex](#citing-radio)\]\[[GitHub examples](https://github.com/NVlabs/RADIO)\] |
|
\[[Tech report on v2.5](https://github.com/NVlabs/RADIO/blob/main/RADIOv2.5_tech_report.md)\] |
|
|
|
|
|
### HuggingFace Hub |
|
|
|
You can pull the model from a Python script: |
|
|
|
```Python |
|
import torch |
|
from PIL import Image |
|
from transformers import AutoModel, CLIPImageProcessor |
|
|
|
hf_repo = "nvidia/RADIO-L" |
|
|
|
image_processor = CLIPImageProcessor.from_pretrained(hf_repo) |
|
model = AutoModel.from_pretrained(hf_repo, trust_remote_code=True) |
|
model.eval().cuda() |
|
|
|
image = Image.open('./assets/radio.png').convert('RGB') |
|
pixel_values = image_processor(images=image, return_tensors='pt', do_resize=True).pixel_values |
|
pixel_values = pixel_values.cuda() |
|
|
|
summary, features = model(pixel_values) |
|
``` |
|
|
|
### Usage |
|
|
|
RADIO will return a tuple with two tensors. The `summary` is similar to the `cls_token` in ViT and is meant to represent the general concept of the entire image. It has shape $(B,C)$ with $B$ being the batch dimension, and $C$ being some number of channels. The `spatial_features` represent more localized content which should be suitable for dense tasks such as semantic segmentation, or for integration into an LLM. It has shape $(B,T,D)$ with $T$ being the flattened spatial tokens, and $D$ being the channels for spatial features. Note that $C \neq D$ in general. |
|
|
|
Converting to a spatial tensor format can be done using the downsampling size of the model, combined with the input tensor shape. For 'radio_v1', the patch size is 14. |
|
```Python |
|
from einops import rearrange |
|
spatial_features = rearrange(spatial_features, 'b (h w) d -> b d h w', h=x.shape[-2] // patch_size, w=x.shape[-1] // patch_size) |
|
``` |
|
|
|
The resulting tensor will have shape $(B,D,H,W)$, as is typically seen with computer vision models. |
|
|
|
### RADIOv2.5 Notes |
|
|
|
See the [RADIOv2.5 technical report](https://github.com/NVlabs/RADIO/blob/main/RADIOv2.5_tech_report.md). |
|
|
|
## License |
|
|
|
RADIO code and weights are released under the [NSCLv1 License](LICENSE). |
|
|
|
## Citing RADIO |
|
|
|
If you find this repository useful, please consider giving a star and citation: |
|
``` |
|
@InProceedings{Ranzinger_2024_CVPR, |
|
author = {Ranzinger, Mike and Heinrich, Greg and Kautz, Jan and Molchanov, Pavlo}, |
|
title = {AM-RADIO: Agglomerative Vision Foundation Model Reduce All Domains Into One}, |
|
booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)}, |
|
month = {June}, |
|
year = {2024}, |
|
pages = {12490-12500} |
|
} |
|
``` |
|
|
|
``` |
|
@misc{ranzinger2024phisdistributionbalancinglabelfree, |
|
title={PHI-S: Distribution Balancing for Label-Free Multi-Teacher Distillation}, |
|
author={Mike Ranzinger and Jon Barker and Greg Heinrich and Pavlo Molchanov and Bryan Catanzaro and Andrew Tao}, |
|
year={2024}, |
|
eprint={2410.01680}, |
|
archivePrefix={arXiv}, |
|
primaryClass={cs.LG}, |
|
url={https://arxiv.org/abs/2410.01680}, |
|
} |
|
``` |
|
|
|
|
