Research map / Video / Action recognition
Hierarchical Spatio-Temporal Representation Learning for Gait Recognition
Lei Wang, Bo Liu, Fangfang Liang, Bincheng Wang
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 ARME (region-independent 3D convolutions) over the non-grouping baseline (ablation) | HSTL · CASIA-B | mean rank-1 accuracy: 91.9 -> 93.0 | Improved |
| add ASTP on top of ARME (ablation) | HSTL · CASIA-B | mean rank-1 accuracy: 93.0 -> 93.4 | Improved |
| add multi-scale FTA (scales 3 and 5) with ARME and ASTP (ablation) | HSTL · CASIA-B | mean rank-1 accuracy: 93.4 -> 94.3 | Improved |
| use single-scale FTA (3 or 5) instead of multi-scale fusion (ablation) | HSTL · CASIA-B | mean rank-1 accuracy: 94.3 -> 93.5 (scale 3), 93.4 (scale 5) | Hurt |
| use hierarchical non-uniform grouping 1-2-4* rather than uniform 1-2-4 (ablation) | HSTL · CASIA-B | mean rank-1 accuracy: +1.0% | Improved |
| hierarchical grouping (1-2-4) instead of non-hierarchical grouping (e.g. 1-1-1) (ablation) | HSTL · CASIA-B | rank-1 accuracy | Improved |
| HSTL versus MetaGait on clothing condition | HSTL · CASIA-B CL | mean rank-1 accuracy: 86.9 -> 88.9 | Improved |
| HSTL versus MetaGait on normal walking | HSTL · CASIA-B NM | mean rank-1 accuracy: 98.1 -> 98.1 | No clear effect |
| HSTL versus MTSGait in the wild | HSTL · GREW | Rank-1 accuracy: 55.32 -> 62.72 | Improved |
| HSTL versus MTSGait in uncontrolled indoor scenes | HSTL · Gait3D | Rank-1 accuracy: 48.70 -> 61.30 | 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.
- MetaGait — MetaGait: Learning to Learn an Omni Sample Adaptive Representation for Gait Recognition (ECCV 2022) · better on 3 of 4 shared results
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
- Gait Recognition via Effective Global-Local Feature Representation and Local Temporal Aggregation (ICCV 2021 · similarity 0.91)
- GaitPart: Temporal Part-Based Model for Gait Recognition (CVPR 2020 · similarity 0.90)
- Context-Sensitive Temporal Feature Learning for Gait Recognition (ICCV 2021 · similarity 0.89)
- Learning Gait Representation From Massive Unlabelled Walking Videos: A Benchmark (TPAMI 2023 · similarity 0.89)
- Learning Visual Prompt for Gait Recognition (CVPR 2024 · similarity 0.88)
- BigGait: Learning Gait Representation You Want by Large Vision Models (CVPR 2024 · similarity 0.87)
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.
