RAFT: Recurrent All-Pairs Field Transforms for Optical Flow (Extended Abstract)

Teed, Zachary; Deng, Jia

doi:10.24963/ijcai.2021/662

RAFT: Recurrent All-Pairs Field Transforms for Optical Flow (Extended Abstract)

Download BibTeX Cite

Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence. Sister Conferences Best Papers. Pages 4839-4843.

https://doi.org/10.24963/ijcai.2021/662

Keywords

Computer Vision: 2D and 3D Computer Vision
Computer Vision: Motion and Tracking

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.

Zachary Teed, and Jia Deng. 2021. RAFT: Recurrent All-Pairs Field Transforms for Optical Flow (Extended Abstract). In Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, IJCAI-21. IJCAI, 4839–4843. https://doi.org/10.24963/ijcai.2021/662

@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    = {Prof. 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},
}

TY  - CPAPER
TI  - RAFT: Recurrent All-Pairs Field Transforms for Optical Flow (Extended Abstract)
AU  - Teed, Zachary
AU  - Deng, Jia
T2  - Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, IJCAI-21
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2021
SP  - 4839
EP  - 4843
DO  - 10.24963/ijcai.2021/662
UR  - https://doi.org/10.24963/ijcai.2021/662
ER  -