STC-Flow: Spatio-temporal context-aware optical flow estimation

作者：

Highlights：

• A general contextual attention framework of multi-level features & complicated fusion.

• STC-Flow to capture rich spatio-temporal long-range dependencies for flow estimation.

• PSC, TCC & RRCU for the effect of feature extraction, correlation, and reconstruction.

• Promising performance in Sintel & KITTI for two-frame based optical flow estimation.

摘要

•A general contextual attention framework of multi-level features & complicated fusion.•STC-Flow to capture rich spatio-temporal long-range dependencies for flow estimation.•PSC, TCC & RRCU for the effect of feature extraction, correlation, and reconstruction.•Promising performance in Sintel & KITTI for two-frame based optical flow estimation.

论文关键词：Spatio-temporal network,Context modeling,Optical flow estimation,Deep learning

论文评审过程：Received 11 October 2020, Revised 28 June 2021, Accepted 11 August 2021, Available online 22 August 2021, Version of Record 1 September 2021.

论文官网地址：https://doi.org/10.1016/j.image.2021.116441