# GaitTAKE: Gait Recognition by Temporal Attention and Keypoint-guided Embedding

arXiv: https://arxiv.org/abs/2207.03608 · HTML: https://mlsanity.com/paper/2207.03608/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 2 of its 19 references that are in the map.
- First posted on arXiv 7 Jul 2022.

## Builds on (most cited first, 2 of 2)

- [Mask R-CNN](https://mlsanity.com/paper/1703.06870/) (2017)
- [GaitSet: Regarding Gait as a Set for Cross-View Gait Recognition](https://mlsanity.com/paper/1811.06186/) (2018)

## Built on by (most cited first, 6 of 6)

- [QAGait: Revisit Gait Recognition from a Quality Perspective](https://mlsanity.com/paper/2401.13531/) (2024)
- [Condition-Adaptive Graph Convolution Learning for Skeleton-Based Gait Recognition](https://mlsanity.com/paper/2308.06707/) (2023)
- [GaitSnippet: Gait Recognition Beyond Unordered Sets and Ordered Sequences](https://mlsanity.com/paper/2508.07782/) (2025)
- [Evidence-based Match-status-Aware Gait Recognition for Out-of-Gallery Gait Identification](https://mlsanity.com/paper/2211.08007/) (2022)
- [CVVNet: A Cross-Vertical-View Network for Gait Recognition](https://mlsanity.com/paper/2505.01837/) (2025)
- [TinySiamese Network for Biometric Analysis](https://mlsanity.com/paper/2307.00578/) (2023)

## Abstract (arXiv)

Gait recognition, which refers to the recognition or identification of a person based on their body shape and walking styles, derived from video data captured from a distance, is widely used in crime prevention, forensic identification, and social security. However, to the best of our knowledge, most of the existing methods use appearance, posture and temporal feautures without considering a learned temporal attention mechanism for global and local information fusion. In this paper, we propose a novel gait recognition framework, called Temporal Attention and Keypoint-guided Embedding (GaitTAKE), which effectively fuses temporal-attention-based global and local appearance feature and temporal aggregated human pose feature. Experimental results show that our proposed method achieves a new SOTA in gait recognition with rank-1 accuracy of 98.0% (normal), 97.5% (bag) and 92.2% (coat) on the CASIA-B gait dataset; 90.4% accuracy on the OU-MVLP gait dataset.