Research map / Label-efficient and robust learning / Recognition and long-tail
Joint Discriminative and Generative Learning 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 |
|---|---|---|---|
| jointly train image generation and re-id learning with generated images fed online | DG-Net · Market-1501 | mAP: 86.0 vs 82.3 | Helped |
| jointly train image generation and re-id learning with generated images fed online | DG-Net · DukeMTMC-reID | mAP: 74.8 vs 71.8 | Helped |
| jointly train image generation and re-id learning with generated images fed online | DG-Net · MSMT17 | mAP: 52.3 vs 40.4 | Helped |
| add primary feature learning with dynamic soft labels (ablation) | ResNet50 re-id baseline · Market-1501 | mAP: 74.5 -> 84.4 | Helped |
| add primary feature learning with dynamic soft labels (ablation) | ResNet50 re-id baseline · DukeMTMC-reID | mAP: 65.3 -> 72.7 | Helped |
| add primary feature learning with dynamic soft labels (ablation) | ResNet50 re-id baseline · MSMT17 | mAP: 36.2 -> 49.7 | Helped |
| use fine-grained feature mining alone (ablation) | ResNet50 re-id baseline · Market-1501 | mAP: 74.5 -> 75.3 | Helped |
| use fine-grained feature mining alone (ablation) | ResNet50 re-id baseline · DukeMTMC-reID | mAP: 65.3 -> 61.2 | Hurt |
| use fine-grained feature mining alone (ablation) | ResNet50 re-id baseline · MSMT17 | mAP: 36.2 -> 43.5 | Helped |
| combine primary and fine-grained features (ablation) | DG-Net · Market-1501 | mAP: 84.4 -> 86.0 | Helped |
| combine primary and fine-grained features (ablation) | DG-Net · DukeMTMC-reID | mAP: 72.7 -> 74.8 | Helped |
| combine primary and fine-grained features (ablation) | DG-Net · MSMT17 | mAP: 49.7 -> 52.3 | Helped |
Closest papers by meaning
- Unsupervised Pre-Training for Person Re-Identification (CVPR 2021 · similarity 0.90)
- Resource Aware Person Re-Identification Across Multiple Resolutions (CVPR 2018 · similarity 0.90)
- Transferable Joint Attribute-Identity Deep Learning for Unsupervised Person Re-Identification (CVPR 2018 · similarity 0.89)
- Unsupervised Cross-Dataset Transfer Learning for Person Re-Identification (CVPR 2016 · similarity 0.89)
- Unsupervised Person Re-Identification by Soft Multilabel Learning (CVPR 2019 · similarity 0.89)
- From Poses to Identity: Training-Free Person Re-Identification via Feature Centralization (CVPR 2025 · similarity 0.88)
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.