English

GTA: Global Temporal Attention for Video Action Understanding

Computer Vision and Pattern Recognition 2022-03-30 v3

Abstract

Self-attention learns pairwise interactions to model long-range dependencies, yielding great improvements for video action recognition. In this paper, we seek a deeper understanding of self-attention for temporal modeling in videos. We first demonstrate that the entangled modeling of spatio-temporal information by flattening all pixels is sub-optimal, failing to capture temporal relationships among frames explicitly. To this end, we introduce Global Temporal Attention (GTA), which performs global temporal attention on top of spatial attention in a decoupled manner. We apply GTA on both pixels and semantically similar regions to capture temporal relationships at different levels of spatial granularity. Unlike conventional self-attention that computes an instance-specific attention matrix, GTA directly learns a global attention matrix that is intended to encode temporal structures that generalize across different samples. We further augment GTA with a cross-channel multi-head fashion to exploit channel interactions for better temporal modeling. Extensive experiments on 2D and 3D networks demonstrate that our approach consistently enhances temporal modeling and provides state-of-the-art performance on three video action recognition datasets.

Keywords

Cite

@article{arxiv.2012.08510,
  title  = {GTA: Global Temporal Attention for Video Action Understanding},
  author = {Bo He and Xitong Yang and Zuxuan Wu and Hao Chen and Ser-Nam Lim and Abhinav Shrivastava},
  journal= {arXiv preprint arXiv:2012.08510},
  year   = {2022}
}

Comments

Accepted to BMVC 2021

R2 v1 2026-06-23T20:59:42.445Z