Research map / Label-efficient and robust learning / Recognition and long-tail
CGCC: Towards Generalizable Clothes-Changing Person Re-Identification
Yizhi Wu, Fangyi Liu, wei yu, Mang Ye
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 BGPA (biometric-guided prototype alignment) (ablation) | EVA02-CLIP baseline · CGCC -> PRCC | Rank-1: 54.5 -> 56.4 | Improved |
| add SAOP (spectrum-aware orthogonal projection) (ablation) | EVA02-CLIP baseline · CGCC -> PRCC | Rank-1: 54.5 -> 59.0 | Improved |
| add both BGPA and SAOP (ablation) | EVA02-CLIP baseline · CGCC -> PRCC | Rank-1: 54.5 -> 62.0 | Improved |
| add both BGPA and SAOP (ablation) | EVA02-CLIP baseline · CGCC -> PRCC | mAP: 50.6 -> 56.9 | Improved |
| add SAOP on top of BGPA (ablation) | TGIR · CGCC -> PRCC | Rank-1: 56.4 -> 62.0 | Improved |
| train on CGCC instead of LaST | MADE · DEEP target | Rank-1: 51.4 -> 59.6 | Improved |
| train on CGCC instead of DEEP | CAL · LTCC target | Rank-1: 13.0 -> 19.4 | Improved |
| train on CGCC instead of DEEP | EVA02-CLIP · LTCC target | Rank-1: 31.6 -> 37.2 | Improved |
| train on CGCC instead of LaST | EVA02-CLIP · PRCC target | Rank-1: 55.5 -> 54.5 | Hurt |
| train on CGCC instead of LaST | TransReID · LTCC target | Rank-1: 12.2 -> 21.9 | Improved |
| use TGIR instead of MADE on CGCC-trained models | MADE · average over LTCC/PRCC/DEEP | Rank-1: 52.7 -> 54.8 | Improved |
| use TGIR instead of Differ on CGCC-trained models | Differ · average over LTCC/PRCC/DEEP | Rank-1: 49.8 -> 54.8 | 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.
- Differ — DIFFER: Disentangling Identity Features via Semantic Cues for Clothes-Changing Person Re-ID (CVPR 2025) · better on 9 of 10 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
- CMCC-ReID: Cross-Modality Clothing-Change Person Re-Identification (ECCV 2026 · similarity 0.92)
- DIFFER: Disentangling Identity Features via Semantic Cues for Clothes-Changing Person Re-ID (CVPR 2025 · similarity 0.91)
- Identity-Clothing Similarity Modeling for Unsupervised Clothing Change Person Re-Identification (CVPR 2025 · similarity 0.90)
- Instruct-ReID: A Multi-purpose Person Re-identification Task with Instructions (CVPR 2024 · similarity 0.88)
- DKC: Differentiated Knowledge Consolidation for Cloth-Hybrid Lifelong Person Re-identification (CVPR 2025 · similarity 0.88)
- From Poses to Identity: Training-Free Person Re-Identification via Feature Centralization (CVPR 2025 · similarity 0.88)
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.
