Research map / Label-efficient and robust learning / Recognition and long-tail
CMCC-ReID: Cross-Modality Clothing-Change Person Re-Identification
Xu Haoxuan, Hanzi Wang, Guanglin Niu
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 DBDL module to CAL baseline (ablation) | CAL baseline · SYSU-CMCC V2I | Rank-1: 40.1 -> 43.2 | Improved |
| add orthogonality loss L_orth to DBDL (ablation) | DBDL · SYSU-CMCC V2I | Rank-1: 43.2 -> 45.6 | Improved |
| add intra-modality prototype loss with progressive schedule (ablation) | DBDL + L_orth · SYSU-CMCC V2I | Rank-1: 45.6 -> 46.9 | Improved |
| add inter-modality prototype loss (ablation) | PIA with L_intra · SYSU-CMCC V2I | Rank-1: 46.9 -> 57.0 | Improved |
| add inter-modality prototype loss (ablation) | PIA with L_intra · SYSU-CMCC I2V | Rank-1: 42.7 -> 50.4 | Improved |
| use progressive two-stage training instead of joint training with BPL from start (ablation) | PIA · SYSU-CMCC V2I | Rank-1: 42.8 -> 57.0 | Improved |
| BPL without progressive learning vs DBDL+L_orth only (ablation) | PIA · SYSU-CMCC V2I | Rank-1: 45.6 -> 42.8 | Hurt |
| vary Stage II start epoch (later start, best at 55) (ablation) | PIA · SYSU-CMCC | Rank-1/mAP | Mixed |
| momentum coefficient alpha too small or too large vs 0.9 (ablation) | BPL prototypes · SYSU-CMCC | Rank-1/mAP | Hurt |
| vary lambda1 and lambda2 (ablation) | PIA · SYSU-CMCC | Rank-1/mAP | No clear effect |
Closest papers by meaning
- CGCC: Towards Generalizable Clothes-Changing Person Re-Identification (ECCV 2026 · similarity 0.92)
- Domain Shifting: A Generalized Solution for Heterogeneous Cross-Modality Person Re-Identification (ECCV 2024 · similarity 0.91)
- Instruct-ReID: A Multi-purpose Person Re-identification Task with Instructions (CVPR 2024 · similarity 0.90)
- Identity-Clothing Similarity Modeling for Unsupervised Clothing Change Person Re-Identification (CVPR 2025 · similarity 0.90)
- DIFFER: Disentangling Identity Features via Semantic Cues for Clothes-Changing Person Re-ID (CVPR 2025 · similarity 0.89)
- Implicit Discriminative Knowledge Learning for Visible-Infrared Person Re-Identification (CVPR 2024 · similarity 0.89)
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.
