Anonymous-sub commited on
Commit
f8b4b37
1 Parent(s): 5500fcd

Update app.py

Browse files
Files changed (1) hide show
  1. app.py +26 -2
app.py CHANGED
@@ -639,7 +639,7 @@ def process2(*args):
639
 
640
 
641
  DESCRIPTION = '''
642
- ## Rerender A Video
643
  ### This space provides the function of key frame translation. Full code for full video translation will be released upon the publication of the paper.
644
  ### To avoid overload, we set limitations to the **maximum frame number** (8) and the maximum frame resolution (512x768).
645
  ### The running time of a video of size 512x640 is about 1 minute per keyframe under T4 GPU.
@@ -647,7 +647,7 @@ DESCRIPTION = '''
647
  1. **Run 1st Key Frame**: only translate the first frame, so you can adjust the prompts/models/parameters to find your ideal output appearance before run the whole video.
648
  2. **Run Key Frames**: translate all the key frames based on the settings of the first frame
649
  3. **Run All**: **Run 1st Key Frame** and **Run Key Frames**
650
- 4. **Run Propagation**: propogate the key frames to other frames for full video translation. This part will be released upon the publication of the paper.
651
  ### Tips:
652
  1. This method cannot handle large or quick motions where the optical flow is hard to estimate. **Videos with stable motions are preferred**.
653
  2. Pixel-aware fusion may not work for large or quick motions.
@@ -661,6 +661,28 @@ DESCRIPTION = '''
661
  [![Duplicate this Space](https://huggingface.co/datasets/huggingface/badges/raw/main/duplicate-this-space-sm-dark.svg)](https://huggingface.co/spaces/Anonymous-sub/Rerender?duplicate=true) for no queue on your own hardware.
662
  '''
663
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
664
  block = gr.Blocks().queue()
665
  with block:
666
  with gr.Row():
@@ -903,6 +925,8 @@ with block:
903
  outputs=[result_image, result_keyframe],
904
  cache_examples=True)
905
 
 
 
906
  def input_uploaded(path):
907
  frame_count = get_frame_count(path)
908
  if frame_count <= 2:
 
639
 
640
 
641
  DESCRIPTION = '''
642
+ ## [Rerender A Video](https://github.com/williamyang1991/Rerender_A_Video)
643
  ### This space provides the function of key frame translation. Full code for full video translation will be released upon the publication of the paper.
644
  ### To avoid overload, we set limitations to the **maximum frame number** (8) and the maximum frame resolution (512x768).
645
  ### The running time of a video of size 512x640 is about 1 minute per keyframe under T4 GPU.
 
647
  1. **Run 1st Key Frame**: only translate the first frame, so you can adjust the prompts/models/parameters to find your ideal output appearance before run the whole video.
648
  2. **Run Key Frames**: translate all the key frames based on the settings of the first frame
649
  3. **Run All**: **Run 1st Key Frame** and **Run Key Frames**
650
+ 4. **Run Propagation**: propogate the key frames to other frames for full video translation. This function is supported [here](https://github.com/williamyang1991/Rerender_A_Video#webui-recommended)
651
  ### Tips:
652
  1. This method cannot handle large or quick motions where the optical flow is hard to estimate. **Videos with stable motions are preferred**.
653
  2. Pixel-aware fusion may not work for large or quick motions.
 
661
  [![Duplicate this Space](https://huggingface.co/datasets/huggingface/badges/raw/main/duplicate-this-space-sm-dark.svg)](https://huggingface.co/spaces/Anonymous-sub/Rerender?duplicate=true) for no queue on your own hardware.
662
  '''
663
 
664
+
665
+ ARTICLE = r"""
666
+ If Rerender-A-Video is helpful, please help to ⭐ the <a href='https://github.com/williamyang1991/Rerender_A_Video' target='_blank'>Github Repo</a>. Thanks!
667
+ [![GitHub Stars](https://img.shields.io/github/stars/williamyang1991/Rerender_A_Video?style=social)](https://github.com/williamyang1991/Rerender_A_Video)
668
+ ---
669
+ 📝 **Citation**
670
+ If our work is useful for your research, please consider citing:
671
+ ```bibtex
672
+ @inproceedings{yang2023rerender,
673
+  title = {Rerender A Video: Zero-Shot Text-Guided Video-to-Video Translation},
674
+  author = {Yang, Shuai and Zhou, Yifan and Liu, Ziwei and and Loy, Chen Change},
675
+ booktitle = {ACM SIGGRAPH Asia Conference Proceedings},
676
+  year = {2023},
677
+ }
678
+ ```
679
+ 📋 **License**
680
+ This project is licensed under <a rel="license" href="https://github.com/williamyang1991/Rerender_A_Video/blob/main/LICENSE.md">S-Lab License 1.0</a>.
681
+ Redistribution and use for non-commercial purposes should follow this license.
682
+ 📧 **Contact**
683
+ If you have any questions, please feel free to reach me out at <b>williamyang@pku.edu.cn</b>.
684
+ """
685
+
686
  block = gr.Blocks().queue()
687
  with block:
688
  with gr.Row():
 
925
  outputs=[result_image, result_keyframe],
926
  cache_examples=True)
927
 
928
+ gr.Markdown(ARTICLE)
929
+
930
  def input_uploaded(path):
931
  frame_count = get_frame_count(path)
932
  if frame_count <= 2: