README.md

June 13, 2019 ยท View on GitHub

PWC-Net (PyTorch v1.0.1)

Pytorch implementation of PWC-Net: CNNs for Optical Flow Using Pyramid, Warping, and Cost Volume. We made it as a off-the-shelf package:

  • After installation, just copy the whole folder PWC_src to your codebase to use. See demo.py for details.

Environment

This code has been test with Python3.6 and PyTorch1.0.1, with a Tesla K80 GPU. The system is Ubuntu 14.04, and the CUDA version is 10.0. All the required python packages can be found in requirements.txt.

Installation

# install custom layers
cd PWC_src/correlation_package
python setup.py install

Note: you might need to add gencode here, according to the GPU you use. You can find more information about gencode here and here.

Converted Caffe Pre-trained Models

You can find them in models folder.

Inference mode

Modify the path to your input, then

python demo.py

If installation is sucessful, you should see the following: PWC-Net Sample Prediction

Reference

If you find this implementation useful in your work, please acknowledge it appropriately and cite the paper using:

@inproceedings{sun2018pwc,
  title={PWC-Net: CNNs for optical flow using pyramid, warping, and cost volume},
  author={Sun, Deqing and Yang, Xiaodong and Liu, Ming-Yu and Kautz, Jan},
  booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},
  pages={8934--8943},
  year={2018}
}

Acknowledgments