Research map / Label-efficient and robust learning / Noisy labels
Large-Scale Pre-Training for Person Re-Identification With Noisy Labels
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 |
|---|---|---|---|
| Pre-train MGN on LUPerson-NL with PNL instead of unsupervised pre-training on LUPerson | MGN · CUHK03 | mAP: 74.7 -> 80.4 | Helped |
| Pre-train MGN on LUPerson-NL with PNL instead of unsupervised pre-training on LUPerson | MGN · Market1501 | mAP: 91.0 -> 91.9 | Helped |
| Pre-train MGN on LUPerson-NL with PNL instead of unsupervised pre-training on LUPerson | MGN · DukeMTMC | mAP: 82.1 -> 84.3 | Helped |
| Pre-train MGN on LUPerson-NL with PNL instead of unsupervised pre-training on LUPerson | MGN · MSMT17 | mAP: 65.7 -> 68.0 | Helped |
| Add prototype-based contrastive learning and prototype label rectification (ablation) | PNL with supervised classification · MSMT17, small-scale 100% | mAP: 62.7 -> 65.8 | Helped |
| Add label-guided contrastive learning instead of using only supervised classification (ablation) | PNL with supervised classification · MSMT17, small-scale 100% | mAP: 62.7 -> 67.1 | Helped |
| Add instance-wise contrastive learning to supervised classification (ablation) | PNL · MSMT17, small-scale 100% | mAP: 62.7 -> 66.5 | Helped |
| Use label-guided contrastive learning instead of instance-wise contrastive learning (ablation) | PNL with supervised classification and prototypes · MSMT17, small-scale 100% | mAP: 67.4 -> 68.0 | Helped |
| Add label correction (ablation) | PNL with ce + pro · MSMT17 | mAP: 64.8 -> 65.8 | Helped |
| Add label correction (ablation) | PNL with ce + pro + lgc · MSMT17 | mAP: 66.7 -> 68.0 | Helped |
| Replace supervised classification labels with instance-wise contrastive learning alone (ablation) | PNL · MSMT17, small-scale 100% | mAP: 62.7 -> 65.3 | Helped |
Closest papers by meaning
- Unsupervised Pre-Training for Person Re-Identification (CVPR 2021 · similarity 0.91)
- Exploit the Unknown Gradually: One-Shot Video-Based Person Re-Identification by Stepwise Learning (CVPR 2018 · similarity 0.90)
- Tracklet Self-Supervised Learning for Unsupervised Person Re-Identification (AAAI 2020 · similarity 0.88)
- Asymmetric Co-Teaching for Unsupervised Cross-Domain Person Re-Identification (AAAI 2020 · similarity 0.88)
- Camera-Aware Proxies for Unsupervised Person Re-Identification (AAAI 2021 · similarity 0.88)
- Unsupervised Person Re-Identification by Soft Multilabel Learning (CVPR 2019 · 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.