Files changed (1) hide show
  1. app.py +15 -15
app.py CHANGED
@@ -192,21 +192,21 @@ def get_example():
192
  [
193
  "./examples/musk_resize.jpeg",
194
  "./examples/poses/pose2.jpg",
195
- "a man flying in the sky in Mars",
196
  "Mars",
197
  "(lowres, low quality, worst quality:1.2), (text:1.2), watermark, (frame:1.2), deformed, ugly, deformed eyes, blur, out of focus, blurry, deformed cat, deformed, photo, anthropomorphic cat, monochrome, photo, pet collar, gun, weapon, blue, 3d, drones, drone, buildings in background, green",
198
  ],
199
  [
200
  "./examples/sam_resize.png",
201
  "./examples/poses/pose4.jpg",
202
- "a man doing a silly pose wearing a suite",
203
  "Jungle",
204
  "(lowres, low quality, worst quality:1.2), (text:1.2), watermark, (frame:1.2), deformed, ugly, deformed eyes, blur, out of focus, blurry, deformed cat, deformed, photo, anthropomorphic cat, monochrome, photo, pet collar, gun, weapon, blue, 3d, drones, drone, buildings in background, gree",
205
  ],
206
  [
207
  "./examples/schmidhuber_resize.png",
208
  "./examples/poses/pose3.jpg",
209
- "a man sit on a chair",
210
  "Neon",
211
  "(lowres, low quality, worst quality:1.2), (text:1.2), watermark, (frame:1.2), deformed, ugly, deformed eyes, blur, out of focus, blurry, deformed cat, deformed, photo, anthropomorphic cat, monochrome, photo, pet collar, gun, weapon, blue, 3d, drones, drone, buildings in background, green",
212
  ],
@@ -433,11 +433,11 @@ title = r"""
433
  description = r"""
434
  <b>Official 🤗 Gradio demo</b> for <a href='https://github.com/InstantID/InstantID' target='_blank'><b>InstantID: Zero-shot Identity-Preserving Generation in Seconds</b></a>.<br>
435
 
436
- We are organizing a Spring Festival event with HuggingFace from 2.7 to 2.25, and you can now generate pictures of Spring Festival costumes. Happy Dragon Year 🐲 ! Share the joy with your family.<br>
437
 
438
  How to use:<br>
439
- 1. Upload an image with a face. For images with multiple faces, we will only detect the largest face. Ensure the face is not too small and is clearly visible without significant obstructions or blurring.
440
- 2. (Optional) You can upload another image as a reference for the face pose. If you don't, we will use the first detected face image to extract facial landmarks. If you use a cropped face at step 1, it is recommended to upload it to define a new face pose.
441
  3. (Optional) You can select multiple ControlNet models to control the generation process. The default is to use the IdentityNet only. The ControlNet models include pose skeleton, canny, and depth. You can adjust the strength of each ControlNet model to control the generation process.
442
  4. Enter a text prompt, as done in normal text-to-image models.
443
  5. Click the <b>Submit</b> button to begin customization.
@@ -447,7 +447,7 @@ article = r"""
447
  ---
448
  📝 **Citation**
449
  <br>
450
- If our work is helpful for your research or applications, please cite us via:
451
  ```bibtex
452
  @article{wang2024instantid,
453
  title={InstantID: Zero-shot Identity-Preserving Generation in Seconds},
@@ -458,15 +458,15 @@ If our work is helpful for your research or applications, please cite us via:
458
  ```
459
  📧 **Contact**
460
  <br>
461
- If you have any questions, please feel free to open an issue or directly reach us out at <b>haofanwang.ai@gmail.com</b>.
462
  """
463
 
464
  tips = r"""
465
- ### Usage tips of InstantID
466
- 1. If you're not satisfied with the similarity, try increasing the weight of "IdentityNet Strength" and "Adapter Strength."
467
  2. If you feel that the saturation is too high, first decrease the Adapter strength. If it remains too high, then decrease the IdentityNet strength.
468
- 3. If you find that text control is not as expected, decrease Adapter strength.
469
- 4. If you find that realistic style is not good enough, go for our Github repo and use a more realistic base model.
470
  """
471
 
472
  css = """
@@ -493,7 +493,7 @@ with gr.Blocks(css=css) as demo:
493
  # prompt
494
  prompt = gr.Textbox(
495
  label="Prompt",
496
- info="Give simple prompt is enough to achieve good face fidelity",
497
  placeholder="A photo of a person",
498
  value="",
499
  )
@@ -501,7 +501,7 @@ with gr.Blocks(css=css) as demo:
501
  submit = gr.Button("Submit", variant="primary")
502
  enable_LCM = gr.Checkbox(
503
  label="Enable Fast Inference with LCM", value=enable_lcm_arg,
504
- info="LCM speeds up the inference step, the trade-off is the quality of the generated image. It performs better with portrait face images rather than distant faces",
505
  )
506
  style = gr.Dropdown(
507
  label="Style template",
@@ -527,7 +527,7 @@ with gr.Blocks(css=css) as demo:
527
  with gr.Accordion("Controlnet"):
528
  controlnet_selection = gr.CheckboxGroup(
529
  ["pose", "canny", "depth"], label="Controlnet", value=["pose"],
530
- info="Use pose for skeleton inference, canny for edge detection, and depth for depth map estimation. You can try all three to control the generation process"
531
  )
532
  pose_strength = gr.Slider(
533
  label="Pose strength",
 
192
  [
193
  "./examples/musk_resize.jpeg",
194
  "./examples/poses/pose2.jpg",
195
+ "a man flying through the sky on mars",
196
  "Mars",
197
  "(lowres, low quality, worst quality:1.2), (text:1.2), watermark, (frame:1.2), deformed, ugly, deformed eyes, blur, out of focus, blurry, deformed cat, deformed, photo, anthropomorphic cat, monochrome, photo, pet collar, gun, weapon, blue, 3d, drones, drone, buildings in background, green",
198
  ],
199
  [
200
  "./examples/sam_resize.png",
201
  "./examples/poses/pose4.jpg",
202
+ "a man doing a silly pose wearing a suit",
203
  "Jungle",
204
  "(lowres, low quality, worst quality:1.2), (text:1.2), watermark, (frame:1.2), deformed, ugly, deformed eyes, blur, out of focus, blurry, deformed cat, deformed, photo, anthropomorphic cat, monochrome, photo, pet collar, gun, weapon, blue, 3d, drones, drone, buildings in background, gree",
205
  ],
206
  [
207
  "./examples/schmidhuber_resize.png",
208
  "./examples/poses/pose3.jpg",
209
+ "a man sitting in a chair",
210
  "Neon",
211
  "(lowres, low quality, worst quality:1.2), (text:1.2), watermark, (frame:1.2), deformed, ugly, deformed eyes, blur, out of focus, blurry, deformed cat, deformed, photo, anthropomorphic cat, monochrome, photo, pet collar, gun, weapon, blue, 3d, drones, drone, buildings in background, green",
212
  ],
 
433
  description = r"""
434
  <b>Official 🤗 Gradio demo</b> for <a href='https://github.com/InstantID/InstantID' target='_blank'><b>InstantID: Zero-shot Identity-Preserving Generation in Seconds</b></a>.<br>
435
 
436
+ We are organizing a Spring Festival event with HuggingFace from February 7 to February 25. You can now generate pictures of Spring Festival costumes. Happy Dragon Year 🐲! Share the joy with your family.<br>
437
 
438
  How to use:<br>
439
+ 1. Upload an image with a face. For images with multiple faces, we will only detect the largest face. Ensure the face is clearly visible without significant obstructions or blurring.
440
+ 2. (Optional) You can upload another image as a reference for the face pose. If you don't, we will use the first detected face image to extract facial landmarks. If you used a cropped face in step 1, it is recommended to upload it to define a new face pose.
441
  3. (Optional) You can select multiple ControlNet models to control the generation process. The default is to use the IdentityNet only. The ControlNet models include pose skeleton, canny, and depth. You can adjust the strength of each ControlNet model to control the generation process.
442
  4. Enter a text prompt, as done in normal text-to-image models.
443
  5. Click the <b>Submit</b> button to begin customization.
 
447
  ---
448
  📝 **Citation**
449
  <br>
450
+ If our work is helpful to you for your research or applications, please cite us via:
451
  ```bibtex
452
  @article{wang2024instantid,
453
  title={InstantID: Zero-shot Identity-Preserving Generation in Seconds},
 
458
  ```
459
  📧 **Contact**
460
  <br>
461
+ If you have any questions, please feel free to open an issue or reach out to us directly at <b>haofanwang.ai@gmail.com</b>.
462
  """
463
 
464
  tips = r"""
465
+ ### Usage Tips for InstantID
466
+ 1. If you're not satisfied with the similarity, try increasing the weight of "IdentityNet Strength" and "Adapter Strength."
467
  2. If you feel that the saturation is too high, first decrease the Adapter strength. If it remains too high, then decrease the IdentityNet strength.
468
+ 3. If text control is not as expected, decrease the Adapter strength.
469
+ 4. If the realistic style is not good enough, visit our GitHub repo and use a more realistic base model.
470
  """
471
 
472
  css = """
 
493
  # prompt
494
  prompt = gr.Textbox(
495
  label="Prompt",
496
+ info="Give a simple prompt in order to achieve good face fidelity.",
497
  placeholder="A photo of a person",
498
  value="",
499
  )
 
501
  submit = gr.Button("Submit", variant="primary")
502
  enable_LCM = gr.Checkbox(
503
  label="Enable Fast Inference with LCM", value=enable_lcm_arg,
504
+ info="LCM speeds up the inference step, but the trade-off is the quality of the generated image. It performs better with portrait face images rather than distant faces.",
505
  )
506
  style = gr.Dropdown(
507
  label="Style template",
 
527
  with gr.Accordion("Controlnet"):
528
  controlnet_selection = gr.CheckboxGroup(
529
  ["pose", "canny", "depth"], label="Controlnet", value=["pose"],
530
+ info="Use pose for skeleton inference, canny for edge detection, and depth for depth map estimation. You can try all three to control the generation process."
531
  )
532
  pose_strength = gr.Slider(
533
  label="Pose strength",