Research map / Label-efficient and robust learning / Recognition and long-tail
Occluded Person Re-Identification With Single-Scale Global Representations
Cheng Yan, Guansong Pang, Jile Jiao, Xiao Bai, Xuetao Feng, Chunhua Shen
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 compound batch erasing (ablation) | FastReID baseline · Occluded-Duke | Rank-1: 62.8 -> 64.1 | Improved |
| add compound batch erasing (ablation) | FastReID baseline · Occluded-Duke | mAP: 51.9 -> 53.3 | Improved |
| add compound batch erasing (ablation) | FastReID baseline · OPReID | Rank-1: 61.1 -> 62.5 | Improved |
| add BED loss to CBE model (ablation) | Baseline + CBE · Occluded-Duke | Rank-1: 64.1 -> 66.8 | Improved |
| add BED loss to CBE model (ablation) | Baseline + CBE · OPReID | mAP: 63.5 -> 65.7 | Improved |
| add disentangled non-local blocks (ablation) | Baseline + CBE + BED · Occluded-Duke | Rank-1: 66.8 -> 67.0 | Improved |
| add reconstructive pooling (ablation) | Baseline + CBE + BED + DNL · Occluded-Duke | Rank-1: 67.0 -> 68.1 | Improved |
| add reconstructive pooling (ablation) | Baseline + CBE + BED + DNL · OPReID | mAP: 65.9 -> 66.4 | Improved |
| add momentum-effect elimination (ablation) | Baseline + CBE + BED + DNL + RP · Occluded-Duke | Rank-1: 68.1 -> 69.0 | Improved |
| add momentum-effect elimination (ablation) | Baseline + CBE + BED + DNL + RP · Market1501 | mAP: 89.0 -> 89.3 | Improved |
| add momentum-effect elimination (ablation) | Baseline + CBE + BED + DNL + RP · DukeMTMC-ReID | mAP: 80.9 -> 81.3 | Improved |
| use very small BED sensitivity alpha (ablation) | Ours · Occluded-Duke and OPReID | Rank-1 and mAP | Hurt |
Closest papers by meaning
- Feature Erasing and Diffusion Network for Occluded Person Re-Identification (CVPR 2022 · similarity 0.93)
- Adversarially Occluded Samples for Person Re-Identification (CVPR 2018 · similarity 0.93)
- Dynamic Feature Pruning and Consolidation for Occluded Person Re-identification (AAAI 2024 · similarity 0.91)
- From Poses to Identity: Training-Free Person Re-Identification via Feature Centralization (CVPR 2025 · similarity 0.89)
- Text-Based Occluded Person Re-identification via Multi-Granularity Contrastive Consistency Learning (AAAI 2024 · similarity 0.88)
- Exploring the Camera Bias of Person Re-identification (ICLR 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.
