IJCAI 2021poster0 citations
RAFT: Recurrent All-Pairs Field Transforms for Optical Flow (Extended Abstract)
Abstract
We introduce Recurrent All-Pairs Field Transforms (RAFT), a new deep network architecture for optical flow. RAFT extracts per-pixel features, builds multi-scale 4D correlation volumes for all pairs of pixels, and iteratively updates a flow field through a recurrent unit that performs lookups on the correlation volumes. RAFT achieves state-of-the-art performance on the KITTI and Sintel datasets. In addition, RAFT has strong cross-dataset generalization as well as high efficiency in inference time, training speed, and parameter count.
Computer Vision: Motion and TrackingComputer Vision: 2D and 3D Computer Vision
BibTeX
@inproceedings{ijcai2021p662,
title = {RAFT: Recurrent All-Pairs Field Transforms for Optical Flow (Extended Abstract)},
author = {Teed, Zachary and Deng, Jia},
booktitle = {Proceedings of the Thirtieth International Joint Conference on
Artificial Intelligence, {IJCAI-21}},
publisher = {International Joint Conferences on Artificial Intelligence Organization},
editor = {Zhi-Hua Zhou},
pages = {4839--4843},
year = {2021},
month = {8},
note = {Sister Conferences Best Papers},
doi = {10.24963/ijcai.2021/662},
url = {https://doi.org/10.24963/ijcai.2021/662},
}