We present SMURF, a method for unsupervised learning of optical flow that improves state of the art on all benchmarks by 36% to 40% (over the prior best method UFlow) and even outperforms several supervised approaches such as PWC-Net and FlowNet2. Our method integrates architecture improvements from supervised optical flow, i.e. the RAFT model, with new ideas for unsupervised learning that include a sequence-aware self-supervision loss, a technique for handling out-of-frame motion, and an approach for learning effectively from multi-frame video data while still only requiring two frames for inference.
@article{arxiv.2105.07014,
title = {SMURF: Self-Teaching Multi-Frame Unsupervised RAFT with Full-Image Warping},
author = {Austin Stone and Daniel Maurer and Alper Ayvaci and Anelia Angelova and Rico Jonschkowski},
journal= {arXiv preprint arXiv:2105.07014},
year = {2021}
}
Comments
Accepted at CVPR 2021, all code available at https://github.com/google-research/google-research/tree/master/smurf