Research map / Label-efficient and robust learning / Recognition and long-tail
CarGait: Cross-Attention based Re-ranking for Gait recognition
Gavriel Habib, Noa Barzilay, Or Shimshi, Rami Ben-Ari, Nir Darshan
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 cross-attention re-ranker CarGait | GaitBase · GREW | Rank-1: 60.1 -> 67.2 | Improved |
| add CarGait re-ranker | SwinGait-3D · Gait3D | Rank-5: 86.7 -> 88.6 | Improved |
| add CarGait re-ranker | DGV2-P3D · Gait3D | Rank-5: 88.0 -> 87.5 | Hurt |
| add CarGait instead of KR/LBR/GCR re-rankers | SwinGait-3D · Gait3D | Rank-1: 76.3 vs 66.7, 64.0, 74.1 | Improved |
| apply KR or LBR re-ranker instead of CarGait | GaitBase · OU-MVLP | Rank-1: 90.8 -> 72.9 (KR), 74.4 (LBR) | Hurt |
| add extra training of the global model (ablation) | SwinGait-3D · Gait3D | Rank-1: 75.0 -> 75.0 | No clear effect |
| add cross-attention instead of binary classifier re-ranker (ablation) | SwinGait-3D · Gait3D | Rank-1: 71.6 -> 76.3 | Improved |
| add classification loss (ablation) | CarGait · Gait3D | Rank-1: 76.0 -> 76.3 R1; 89.1 -> 88.6 R5 | Mixed |
| add loss damping (beta=0.1) (ablation) | CarGait · Gait3D | Rank-1: 75.5 -> 76.3 | Improved |
| use inference K=10 instead of K=5 (ablation) | CarGait · Gait3D | Rank-5: 86.7 -> 88.6 | Improved |
| use inference K=20 instead of K=10 (ablation) | CarGait · Gait3D | mAP: 67.59 -> 67.68 | No clear effect |
| train with v=30 instead of v=20 or v=40 (ablation) | CarGait · Gait3D | Rank-1: 76.3 (v=20), 76.0 (v=40) vs 76.3 | Mixed |
Closest papers by meaning
- OpenGait: A Comprehensive Benchmark Study for Gait Recognition Toward Better Practicality (TPAMI 2025 · similarity 0.89)
- BiggerGait: Unlocking Gait Recognition with Layer-wise Representations from Large Vision Models (NeurIPS 2025 · similarity 0.88)
- GaitSet: Cross-view Gait Recognition through Utilizing Gait as a Deep Set (TPAMI 2021 · similarity 0.88)
- BigGait: Learning Gait Representation You Want by Large Vision Models (CVPR 2024 · similarity 0.87)
- Cross-Covariate Gait Recognition: A Benchmark (AAAI 2024 · similarity 0.87)
- Learning Visual Prompt for Gait Recognition (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.
