Peter Lin
ee86e2636f
Initialize
|
vor 4 Jahren | |
|---|---|---|
| dataset | vor 4 Jahren | |
| doc | vor 4 Jahren | |
| images | vor 4 Jahren | |
| model | vor 4 Jahren | |
| README.md | vor 4 Jahren | |
| export_onnx.py | vor 4 Jahren | |
| export_torchscript.py | vor 4 Jahren | |
| inference_images.py | vor 4 Jahren | |
| inference_speed_test.py | vor 4 Jahren | |
| inference_utils.py | vor 4 Jahren | |
| inference_video.py | vor 4 Jahren | |
| inference_webcam.py | vor 4 Jahren | |
| requirements.txt | vor 4 Jahren |
README.md
Real-Time High-Resolution Background Matting
Official repository for the paper Real-Time High-Resolution Background Matting. Our model produces state-of-the-art matting results at 4K 30fps and HD 60fps on an Nvidia RTX 2080 TI GPU.
Overview
Download
Model / Weights
Video / Image Examples
- HD videos (by Sengupta et al.) (Our model is more robust on HD footage)
- 4K videos and images
Datasets
- VideoMatte240K (Coming soon)
- PhotoMatte85 (Coming soon)
Demo
Scripts
We provide several scripts in this repo for you to experiment with our model. More detail instructions are included in the files.
inference_images.py: Perform matting on a directory of images.inference_video.py: Perform matting on a video.inference_webcam.py: An interactive matting demo using your webcam.
Notebooks
Additionally, you can try our notebooks in Google Colab for performing matting on images and videos.
Usage / Documentation
You can run our model using PyTorch, TorchScript, TensorFlow, and ONNX. For detail about using our model, please check out the Usage / Documentation page.
Training
Training code will be released upon acceptance of the paper.
Project members
- Shanchuan Lin*, University of Washington
- Andrey Ryabtsev*, University of Washington
- Soumyadip Sengupta, University of Washington
- Brian Curless, University of Washington
- Steve Seitz, University of Washington
- Ira Kemelmacher-Shlizerman, University of Washington
* Equal contribution.