ETA: Learning Optical Flow with Efficient Temporal Attention
Bo Wang, Zhenping Sun, Yang Yu, Li Liu, Jian Li, Dewen Hu
Abstract
Considering the potential of using multi-frame information to solve the occlusion problem, we introduce a novel idea of multi-frame information integration, which uses the attention mechanism to fuse the temporal information from the previous frame. The idea can effectively improve the estimation accuracy in occluded regions and optimize the inference speed under multi-frame settings. Meanwhile, we suggest the concept of attention confidence to provide an explicit value criterion for the model to utilize useful attention information more efficiently. Furthermore, we propose an Efficient Temporal Attention network (ETA), which achieves promising results on Sintel and KITTI benchmarks, especially with a 9.4% error reduction compared to the baseline method GMA on Sintel (test) Clean.
BibTeX
@inproceedings{iros2025_etalearningoptic,
title = {ETA: Learning Optical Flow with Efficient Temporal Attention},
author = {Bo Wang and Zhenping Sun and Yang Yu and Li Liu and Jian Li and Dewen Hu},
booktitle = {IROS 2025},
year = {2025}
}