Research map / Graph learning / Knowledge and temporal graphs
High-Order Information Matters: Learning Relation and Topology for Occluded 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 HOReID with adaptive relation learning and graph-based topology alignment | ResNet50 person ReID · Occluded-Duke | Rank-1: 51.4 -> 55.1 | Helped |
| Use HOReID with adaptive relation learning and graph-based topology alignment | ResNet50 person ReID · Occluded-Duke | mAP: 37.3 -> 43.8 | Helped |
| Use HOReID | ResNet50 person ReID · Occluded-ReID | Rank-1: 78.3 -> 80.3 | Helped |
| Use HOReID | ResNet50 person ReID · Occluded-ReID | mAP: 77.9 -> 70.2 | Hurt |
| Use HOReID | ResNet50 person ReID · Partial-REID | Rank-1: 82.7 -> 85.3 | Helped |
| Use HOReID | ResNet50 person ReID · Partial-iLIDS | Rank-1: 69.1 -> 72.6 | Helped |
| Use HOReID | ResNet50 person ReID · Market-1501 | Rank-1: 95.4 -> 94.2 | Hurt |
| Use HOReID | ResNet50 person ReID · DukeMTMC-reID | Rank-1: 88.6 -> 86.9 | Hurt |
| Add the semantic module to the global-feature-only IDE baseline (ablation) | HOReID · Occluded-Duke | Rank-1: 49.9 -> 52.4 | Helped |
| Add the high-order relation module (ablation) | HOReID with semantic module · Occluded-Duke | Rank-1: 52.4 -> 53.9 | Helped |
| Add the high-order topology module (ablation) | HOReID with semantic and relation modules · Occluded-Duke | Rank-1: 53.9 -> 55.1 | Helped |
| Add key-point confidence normalization (ablation) | HOReID · Occluded-Duke | Rank-1: 52.5 -> 55.1 | Helped |
Closest papers by meaning
- Dynamic Feature Pruning and Consolidation for Occluded Person Re-identification (AAAI 2024 · similarity 0.90)
- Pose-Guided Feature Disentangling for Occluded Person Re-identification Based on Transformer (AAAI 2022 · similarity 0.89)
- Feature Erasing and Diffusion Network for Occluded Person Re-Identification (CVPR 2022 · similarity 0.89)
- Matching on Sets: Conquer Occluded Person Re-identification Without Alignment (AAAI 2021 · similarity 0.89)
- Text-Based Occluded Person Re-identification via Multi-Granularity Contrastive Consistency Learning (AAAI 2024 · similarity 0.88)
- Cheb-GR: Rethinking K-nearest Neighbor Search in Re-ranking for Person Re-identification (CVPR 2025 · 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.