Research map / Label-efficient and robust learning / Recognition and long-tail
Identity-Guided Collaborative Learning for Cloth-Changing Person Reidentification
Zan Gao, Shengxun Wei, Weili Guan, Lei Zhu, Meng Wang, Shengyong Chen
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 clothing attention degradation (CAD) stream (ablation) | ViT baseline · PRCC | rank-1: 46.3 -> 54.2 | Improved |
| add clothing attention degradation (CAD) stream (ablation) | ViT baseline · VC-Clothes | mAP: 70.6 -> 71.8 | No clear effect |
| add multi-scale (three) attention maps instead of F1 only (ablation) | CAD stream · PRCC | rank-1: 53.3 -> 54.2 | Improved |
| set clothing elimination weight alpha to 0.1 (ablation) | CAD stream · PRCC | rank-1: 52.0 -> 54.2 | Improved |
| increase alpha to 0.5 (ablation) | CAD stream · Celeb-reID-light | mAP: 21.3 -> 18.0 | Hurt |
| add human semantic attention (foreground semantic image) (ablation) | ViT baseline · PRCC | rank-1: 46.3 -> 63.4 | Improved |
| add human semantic attention (ablation) | ViT baseline · Celeb-reID-light | mAP: 17.1 -> 22.2 | Improved |
| add body jigsaw to semantic stream (ablation) | SAJ stream · Celeb-reID-light | rank-1: 38.6 -> 39.5 | Improved |
| add body jigsaw to semantic stream (ablation) | SAJ stream · PRCC | mAP: 59.6 -> 59.5 | No clear effect |
| add pedestrian identity enhancement (PIE) stream on original image (ablation) | ViT baseline · Celeb-reID-light | rank-1: 30.2 -> 36.9 | Improved |
| use foreground image with shielding of upper clothes in PIE (ablation) | PIE stream · PRCC | rank-1: 47.3 -> 52.6 | Improved |
| add PIE to CAD+SAJ (ablation) | IGCL · PRCC | mAP: 60.1 -> 59.5 | Hurt |
Closest papers by meaning
- DIFFER: Disentangling Identity Features via Semantic Cues for Clothes-Changing Person Re-ID (CVPR 2025 · similarity 0.89)
- Identity-Clothing Similarity Modeling for Unsupervised Clothing Change Person Re-Identification (CVPR 2025 · similarity 0.88)
- Instruct-ReID: A Multi-purpose Person Re-identification Task with Instructions (CVPR 2024 · similarity 0.87)
- From Poses to Identity: Training-Free Person Re-Identification via Feature Centralization (CVPR 2025 · similarity 0.86)
- DKC: Differentiated Knowledge Consolidation for Cloth-Hybrid Lifelong Person Re-identification (CVPR 2025 · similarity 0.85)
- Distilling CLIP with Dual Guidance for Learning Discriminative Human Body Shape Representation (CVPR 2024 · similarity 0.85)
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.
