This repository contains the source code for our optical flow estimation method:
CRAFT: Cross-Attentional Flow Transformers for Robust Optical Flow
Xiuchao Sui, Shaohua Li, Xue Geng, Yan Wu, Xinxing Xu, Yong Liu, Rick Goh, Hongyuan Zhu
You will have to choose cudatoolkit version to match your compute environment. The code is tested on PyTorch 1.8.0 but other versions may also work.
pip install -r requirements.txtsh train-craft-f2full.shsh evaluate-craft-f2full.shWTFPL. See LICENSE file.
@InProceedings{craft,
author="Sui, Xiuchao and Li, Shaohua and Geng, Xue and Wu, Yan and Xu, Xinxing and Liu, Yong and Goh, Rick Siow Mong and Zhu, Hongyuan",
title="CRAFT: Cross-Attentional Flow Transformers for Robust Optical Flow",
booktitle="CVPR",
year="2022"}
The overall code framework is adapted from GMA. We thank the authors for their contributions.