Research map / Continual learning / Domain and prompt-based
SCORE: SubDistribution-aware Collaborative Knowledge Reinforcing for Cloth-Hybrid Lifelong Person Re-Identification
Kunlun Xu, Liangyu Ma, Jiangmeng Li, Xin Tong, Xiaode Liu, Yufei Guo, Jiahuan Zhou
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 Adaptive SubDistribution Modeling (ASD) to sequential finetuning (ablation) | SFT · CH-LReID Order-1 cloth-consistent | mAP: 40.1 -> 50.1 | Improved |
| add ASD to sequential finetuning (ablation) | SFT · CH-LReID Order-1 cloth-changing | mAP: 14.4 -> 22.3 | Improved |
| add Instance-oriented Structural Knowledge Retention (ISK) (ablation) | SCORE (ASD) · CH-LReID Order-1 cloth-consistent | mAP: 50.1 -> 52.8 | Improved |
| add ISK (ablation) | SCORE (ASD) · CH-LReID Order-1 cloth-changing | mAP: 22.3 -> 23.0 | Improved |
| add Distributional Knowledge Reinforcement (DKR) (ablation) | SCORE (ASD+ISK) · CH-LReID Order-1 cloth-consistent | mAP: 52.8 -> 57.5 | Improved |
| add DKR (ablation) | SCORE (ASD+ISK) · CH-LReID Order-1 cloth-changing | Rank-1: 29.6 -> 31.1 | Improved |
| use fixed k=5 subprototypes instead of dynamic k (ablation) | SCORE · CH-LReID Order-1 overall | mAP: 47.2 -> 47.8 | Improved |
| use SCORE instead of DKC on cloth-changing data (Order-2) | SCORE · CH-LReID Order-2 cloth-changing | mAP: 25.6 -> 25.4 | No clear effect |
| use SCORE instead of DKC on cloth-changing data (Order-1) | SCORE · CH-LReID Order-1 cloth-changing | mAP: 23.5 -> 23.6 | No clear effect |
| use SCORE instead of DKC overall | SCORE · CH-LReID Order-2 overall | mAP: 43.7 -> 46.4 | Improved |
| combine cloth-changing ReID methods (DLCR, PFAC) with DASK instead of SCORE | DASK+PFAC · CH-LReID overall | mAP: 47.8 -> 40.3 | Hurt |
| use SCORE instead of DKC on forgetting measure | SCORE · CH-LReID | FM (mAP): 9.1 -> 3.5 | 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.
- DASK — DASK: Distribution Rehearsing via Adaptive Style Kernel Learning for Exemplar-Free Lifelong Person Re-Identification (AAAI 2025) · better on 6 of 6 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
- DKC: Differentiated Knowledge Consolidation for Cloth-Hybrid Lifelong Person Re-identification (CVPR 2025 · similarity 0.94)
- Distribution-Aware Knowledge Aligning and Prototyping for Non-Exemplar Lifelong Person Re-Identification (TPAMI 2025 · similarity 0.91)
- Long Short-Term Knowledge Decomposition and Consolidation for Lifelong Person Re-Identification (TPAMI 2025 · similarity 0.91)
- Vision-Language Attribute Disentanglement and Reinforcement for Lifelong Person Re-Identification (CVPR 2026 · similarity 0.90)
- LSTKC: Long Short-Term Knowledge Consolidation for Lifelong Person Re-identification (AAAI 2024 · similarity 0.90)
- Distribution-aware Knowledge Prototyping for Non-exemplar Lifelong Person Re-identification (CVPR 2024 · similarity 0.90)
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.
