Research map / Label-efficient and robust learning / Domain adaptation and generalization
Mutual Mean-Teaching: Pseudo Label Refinery for Unsupervised Domain Adaptation on Person Re-identification
Read the paperCheck 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 |
|---|---|---|---|
| Apply MMT instead of SSG | Person re-identification UDA · Market-to-Duke | mAP | Helped |
| Apply MMT instead of SSG | Person re-identification UDA · Duke-to-Market | mAP | Helped |
| Apply MMT instead of SSG | Person re-identification UDA · Market-to-MSMT | mAP | Helped |
| Apply MMT instead of SSG | Person re-identification UDA · Duke-to-MSMT | mAP | Helped |
| Add on-line refined soft pseudo-label supervision to hard pseudo-label training (ablation) | MMT · Duke-to-Market | mAP: 17.7% | Helped |
| Add on-line refined soft pseudo-label supervision to hard pseudo-label training (ablation) | MMT · Market-to-Duke | mAP: 14.9% | Helped |
| Add soft softmax-triplet loss (ablation) | MMT · Duke-to-Market | mAP: 5.3% | Helped |
| Add soft softmax-triplet loss (ablation) | MMT · Market-to-Duke | mAP: 3.6% | Helped |
| Add hard classification loss (ablation) | MMT · Duke-to-Market | mAP: 76.5 vs 38.0 | Helped |
| Add hard classification loss (ablation) | MMT · Market-to-Duke | mAP: 65.7 vs 27.5 | Helped |
| Add hard triplet loss (ablation) | MMT · Duke-to-Market | mAP: 76.2 vs 76.5 | No clear effect |
| Add hard triplet loss (ablation) | MMT · Market-to-Duke | mAP: 65.6 vs 65.7 | No clear effect |
Closest papers by meaning
- Group-aware Label Transfer for Domain Adaptive Person Re-identification (CVPR 2021 · similarity 0.93)
- SECRET: Self-Consistent Pseudo Label Refinement for Unsupervised Domain Adaptive Person Re-identification (AAAI 2022 · similarity 0.93)
- Domain Adaptive Attention Learning for Unsupervised Person Re-Identification (AAAI 2020 · similarity 0.93)
- Unsupervised Multi-Source Domain Adaptation for Person Re-Identification (CVPR 2021 · similarity 0.92)
- Unsupervised Person Re-Identification by Soft Multilabel Learning (CVPR 2019 · similarity 0.92)
- Reliability Exploration with Self-Ensemble Learning for Domain Adaptive Person Re-identification (AAAI 2022 · similarity 0.92)
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.