Research map / Label-efficient and robust learning / Recognition and long-tail
Perceive Where to Focus: Learning Visibility-Aware Part-Level Features for Partial 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 |
|---|---|---|---|
| replace global features with visibility-aware region features | person re-identification · Market-1501, γ=0.5 | R-1: 64.5 -> 70.9 | Helped |
| replace global features with visibility-aware region features | person re-identification · Market-1501, γ=0.5 | mAP: 44.4 -> 48.8 | Helped |
| replace global features with visibility-aware region features | person re-identification · DukeMTMC-reID, γ=0.5 | R-1: 65.0 -> 69.5 | Helped |
| use VPM instead of SFR with bilateral crop training | Partial-REID · Partial-REID | R-1: +10.6% | Helped |
| use VPM instead of SFR with top-crop training | Partial-iLIDS · Partial-iLIDS | R-1: +3.3% | Helped |
| add triplet loss (ablation) | VPM · Partial-iLIDS | R-1: 57.1 -> 67.2 | Helped |
| add triplet loss (ablation) | VPM · Market-1501 | mAP: 77.8 -> 80.8 | Helped |
| use visibility-aware shared regions instead of all region features at testing (ablation) | VPM · Partial-iLIDS | R-1: 63.0 -> 67.2 | Helped |
| use visibility-aware shared regions instead of all region features at testing (ablation) | VPM · Market-1501 | mAP: 79.7 -> 80.8 | Helped |
| apply self-supervision to triplet-loss region selection (ablation) | VPM · Partial-iLIDS | R-1: 61.3 -> 67.2 | Helped |
| apply self-supervision to identity-loss region selection (ablation) | VPM · Partial-iLIDS | R-1: 58.8 -> 67.2 | Helped |
| apply self-supervision to both triplet and identity losses (ablation) | VPM · Partial-iLIDS | R-1: 59.7 -> 67.2 | Helped |
Closest papers by meaning
- Partial Person Re-Identification With Part-Part Correspondence Learning (CVPR 2021 · similarity 0.88)
- Relation-Aware Global Attention for Person Re-Identification (CVPR 2020 · similarity 0.87)
- Deep Spatial Feature Reconstruction for Partial Person Re-Identification: Alignment-Free Approach (CVPR 2018 · similarity 0.87)
- Camera-Aware Proxies for Unsupervised Person Re-Identification (AAAI 2021 · similarity 0.87)
- Similarity Learning With Spatial Constraints for Person Re-Identification (CVPR 2016 · similarity 0.87)
- Implicit Discriminative Knowledge Learning for Visible-Infrared Person Re-Identification (CVPR 2024 · similarity 0.86)
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.