Research map / Label-efficient and robust learning / Domain adaptation and generalization
Reliability Exploration with Self-Ensemble Learning for Domain Adaptive 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 |
|---|---|---|---|
| Add the temporally averaged model and contrastive loss (ablation) | RESL baseline · Duke to Market | mAP: 69.1 -> 79.0 | Helped |
| Add reliability-weighted loss (ablation) | RESL with temporally averaged model and contrastive loss · Duke to Market | mAP: 79.0 -> 79.5 | Helped |
| Add self-ensemble learning (ablation) | RESL with temporally averaged model and contrastive loss · Duke to Market | mAP: 79.0 -> 81.3 | Helped |
| Add reliable sample selection (ablation) | RESL with temporally averaged model and contrastive loss · Duke to Market | mAP: 79.0 -> 82.1 | Helped |
| Add self-ensemble learning (ablation) | RESL with temporally averaged model and contrastive loss · Market to Duke | mAP: 69.2 -> 70.6 | Helped |
| Add reliable sample selection (ablation) | RESL with temporally averaged model and contrastive loss · Market to Duke | mAP: 69.2 -> 71.2 | Helped |
| Use only one branch at inference instead of the full multi-branch model (ablation) | RESL · Duke to Market | mAP: 83.1 -> 80.7 | Hurt |
| Vary the number of branches K (ablation) | RESL · Duke to Market | performance | Mixed |
| Vary the reliability threshold β (ablation) | RESL · Duke to Market | performance | Mixed |
Closest papers by meaning
- Exploiting Sample Uncertainty for Domain Adaptive Person Re-Identification (AAAI 2021 · similarity 0.94)
- Group-aware Label Transfer for Domain Adaptive Person Re-identification (CVPR 2021 · similarity 0.93)
- Delving into Probabilistic Uncertainty for Unsupervised Domain Adaptive Person Re-identification (AAAI 2022 · similarity 0.93)
- Mutual Mean-Teaching: Pseudo Label Refinery for Unsupervised Domain Adaptation on Person Re-identification (ICLR 2020 · similarity 0.92)
- SECRET: Self-Consistent Pseudo Label Refinement for Unsupervised Domain Adaptive Person Re-identification (AAAI 2022 · similarity 0.91)
- Unsupervised Domain Adaptive Person Search via Dual Self-Calibration (AAAI 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.