Research map / Video / Action recognition
Context-Sensitive Temporal Feature Learning for Gait Recognition
Xiaohu Huang, Duowang Zhu, Hao Wang, Xinggang Wang, Bo Yang, Botao He, Wenyu Liu, Bin Feng
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add multi-scale temporal extraction (MSTE) (ablation) | Baseline · CASIA-B mean | Mean rank-1 accuracy (%): 85.4 -> 89.6 | Improved |
| add adaptive temporal aggregation (ATA) to MSTE (ablation) | Baseline + MSTE · CASIA-B mean | Mean rank-1 accuracy (%): 89.6 -> 90.1 | Improved |
| add salient spatial feature learning (SSFL) to MSTE (ablation) | Baseline + MSTE · CASIA-B mean | Mean rank-1 accuracy (%): 89.6 -> 91.1 | Improved |
| add SSFL on top of MSTE and ATA (ablation) | Baseline + MSTE + ATA · CASIA-B mean | Mean rank-1 accuracy (%): 90.1 -> 91.9 | Improved |
| add ATA to MSTE on the coat-wearing condition (ablation) | Baseline + MSTE · CASIA-B CL probe | Rank-1 accuracy (%): 81.0 -> 79.1 | Hurt |
| add SSFL to MSTE on the coat-wearing condition (ablation) | Baseline + MSTE · CASIA-B CL probe | Rank-1 accuracy (%): 81.0 -> 83.7 | Improved |
| use all three temporal scales instead of short-term only (ablation) | CSTL · CASIA-B mean | Mean rank-1 accuracy (%): 90.2 -> 91.9 | Improved |
| use long-term features alone instead of frame-level alone (ablation) | CSTL · CASIA-B mean | Mean rank-1 accuracy (%): 88.5 -> 87.5 | Hurt |
| use short-term features alone instead of frame-level alone (ablation) | CSTL · CASIA-B mean | Mean rank-1 accuracy (%): 88.5 -> 90.2 | Improved |
| use ATA instead of max pooling (ablation) | CSTL · CASIA-B mean | Mean rank-1 accuracy (%): 91.1 -> 91.9 | Improved |
| use ATA instead of average pooling (ablation) | CSTL · CASIA-B mean | Mean rank-1 accuracy (%): 90.6 -> 91.9 | Improved |
| use max pooling instead of average pooling (ablation) | CSTL · CASIA-B mean | Mean rank-1 accuracy (%): 90.6 -> 91.1 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- GaitPart — GaitPart: Temporal Part-Based Model for Gait Recognition (CVPR 2020) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- GaitPart: Temporal Part-Based Model for Gait Recognition (CVPR 2020 · similarity 0.90)
- Gait Recognition via Effective Global-Local Feature Representation and Local Temporal Aggregation (ICCV 2021 · similarity 0.90)
- Learning Visual Prompt for Gait Recognition (CVPR 2024 · similarity 0.90)
- Hierarchical Spatio-Temporal Representation Learning for Gait Recognition (ICCV 2023 · similarity 0.89)
- GaitSnippet: Gait Recognition Beyond Unordered Sets and Ordered Sequences (ICLR 2026 · similarity 0.89)
- Vocabulary-Guided Gait Recognition (NeurIPS 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
