Research map / Label-efficient and robust learning / Recognition and long-tail
Meta Pairwise Relationship Distillation for Unsupervised Person Re-Identification
Haoxuanye Ji, Le Wang, Sanping Zhou, Wei Tang, Nanning Zheng, Gang Hua
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 GCN pairwise pseudo-label estimation instead of similarity threshold (SS) (ablation) | MPRD · Market-1501 | Rank-1: 72.8 -> 83.0 | Improved |
| add GCN pairwise pseudo-label estimation instead of similarity threshold (SS) (ablation) | MPRD · Market-1501 | mAP: 39.9 -> 51.1 | Improved |
| train GCN with unlabeled data in addition to labeled meta data (ablation) | MPRD · Market-1501 | Rank-1: 75.8 -> 83.0 | Improved |
| train GCN with unlabeled data in addition to labeled meta data (ablation) | MPRD · Market-1501 | mAP: 43.1 -> 51.1 | Improved |
| add CamStyle data augmentation (ablation) | MPRD · Market-1501 | Rank-1: 73.0 -> 83.0 | Improved |
| add CamStyle data augmentation (ablation) | MPRD · Market-1501 | mAP: 39.3 -> 51.1 | Improved |
| add 5 labeled identities as meta data jump start for GCN (ablation) | MPRD · Market-1501 | Rank-1: 80.9 -> 83.0 | Improved |
| add 5 labeled identities as meta data jump start for GCN (ablation) | MPRD · DukeMTMC-reID | mAP: 40.1 -> 43.7 | Improved |
| give MLCR the same 5 labeled identities as extra supervision (ablation) | MLCR · Market-1501 | Rank-1: 80.1 -> 80.2 | No clear effect |
| use KNN instead of similarity threshold for pseudo labels (ablation) | pseudo-label generation · Market-1501 | Rank-1: 72.8 vs 72.7 | No clear effect |
| raise likelihood threshold mu beyond about 0.5 (ablation) | MPRD · Market-1501 | Rank-1 and mAP | Hurt |
| MPRD versus MLCR | unsupervised Re-ID · Market-1501 | Rank-1: 80.3 -> 83.0 | Improved |
Closest papers by meaning
- Unsupervised Person Re-Identification by Soft Multilabel Learning (CVPR 2019 · similarity 0.90)
- Intra-Inter Domain Similarity for Unsupervised Person Re-Identification (TPAMI 2022 · similarity 0.90)
- Intra-Inter Camera Similarity for Unsupervised Person Re-Identification (CVPR 2021 · similarity 0.89)
- Mutual Mean-Teaching: Pseudo Label Refinery for Unsupervised Domain Adaptation on Person Re-identification (ICLR 2020 · similarity 0.89)
- Part-Based Pseudo Label Refinement for Unsupervised Person Re-Identification (CVPR 2022 · similarity 0.89)
- Asymmetric Co-Teaching for Unsupervised Cross-Domain Person Re-Identification (AAAI 2020 · 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.
