AutoFlow: Learning a Better Training Set for Optical Flow

Deqing Sun, Daniel Vlasic, Charles Herrmann, Varun Jampani, Michael Krainin, Huiwen Chang, Ramin Zabih, William T. Freeman, Ce Liu

CVPR 2021 pp. 10093-10102

doi:10.1109/CVPR46437.2021.00996 /cvpr/2021/sun2021cvpr-autoflow/

Abstract

Synthetic datasets play a critical role in pre-training CNN models for optical flow, but they are painstaking to generate and hard to adapt to new applications. To automate the process, we present AutoFlow, a simple and effective method to render training data for optical flow that optimizes the performance of a model on a target dataset. AutoFlow takes a layered approach to render synthetic data, where the motion, shape, and appearance of each layer are controlled by learnable hyperparameters. Experimental results show that AutoFlow achieves state-of-the-art accuracy in pre-training both PWC-Net and RAFT. Our code and data are available at autoflow-google.github.io.

PDF CVPR Semantic Scholar

Cite

Text

Sun et al. "AutoFlow: Learning a Better Training Set for Optical Flow." Conference on Computer Vision and Pattern Recognition, 2021. doi:10.1109/CVPR46437.2021.00996

Markdown

[Sun et al. "AutoFlow: Learning a Better Training Set for Optical Flow." Conference on Computer Vision and Pattern Recognition, 2021.](https://mlanthology.org/cvpr/2021/sun2021cvpr-autoflow/) doi:10.1109/CVPR46437.2021.00996

BibTeX

@inproceedings{sun2021cvpr-autoflow,
  title     = {{AutoFlow: Learning a Better Training Set for Optical Flow}},
  author    = {Sun, Deqing and Vlasic, Daniel and Herrmann, Charles and Jampani, Varun and Krainin, Michael and Chang, Huiwen and Zabih, Ramin and Freeman, William T. and Liu, Ce},
  booktitle = {Conference on Computer Vision and Pattern Recognition},
  year      = {2021},
  pages     = {10093-10102},
  doi       = {10.1109/CVPR46437.2021.00996},
  url       = {https://mlanthology.org/cvpr/2021/sun2021cvpr-autoflow/}
}