Research map / Label-efficient and robust learning / Recognition and long-tail
Hierarchical Clustering With Hard-Batch Triplet Loss for Person Re-Identification
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| use HCT instead of BUC | fully unsupervised person re-identification · Market-1501 | mAP: 38.3 -> 56.4 | Helped |
| use HCT instead of BUC | fully unsupervised person re-identification · DukeMTMC-reID | mAP: 27.5 -> 50.7 | Helped |
| set merge step s = 12 instead of s = 13 (ablation) | HCT · Market-1501 | mAP: 56.4 -> 46.2 | Hurt |
| set merge step s = 14 instead of s = 13 (ablation) | HCT · Market-1501 | mAP: 56.4 -> × | Hurt |
| set merge percent mp = 0.04 instead of mp = 0.07 (ablation) | HCT · Market-1501 | mAP: 56.4 -> 55.3 | Hurt |
| set merge percent mp = 0.05 instead of mp = 0.07 (ablation) | HCT · Market-1501 | mAP: 56.4 -> 55.0 | Hurt |
| set merge percent mp = 0.06 instead of mp = 0.07 (ablation) | HCT · Market-1501 | mAP: 56.4 -> 54.3 | Hurt |
| set merge percent mp = 0.08 instead of mp = 0.07 (ablation) | HCT · Market-1501 | mAP: 56.4 -> 53.0 | Hurt |
| set merge percent mp = 0.09 instead of mp = 0.07 (ablation) | HCT · Market-1501 | mAP: 56.4 -> 54.6 | Hurt |
| set merge percent mp = 0.1 instead of mp = 0.07 (ablation) | HCT · Market-1501 | mAP: 56.4 -> 53.7 | Hurt |
Closest papers by meaning
- Group-aware Label Transfer for Domain Adaptive 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 0.88)
- Learning Incremental Triplet Margin for Person Re-Identification (AAAI 2019 · similarity 0.88)
- Unsupervised Person Re-Identification via Softened Similarity Learning (CVPR 2020 · similarity 0.87)
- Divide-and-Regroup Clustering for Domain Adaptive Person Re-identification (AAAI 2022 · similarity 0.87)
- A Bottom-Up Clustering Approach to Unsupervised Person Re-Identification (AAAI 2019 · similarity 0.87)
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.