Research map / Label-efficient and robust learning / Domain adaptation and generalization
Domain Adaptive Person Search
Junjie Li, Yichao Yan, Guanshuo Wang, Fufu Yu, Qiong Jia, Shouhong Ding
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 domain alignment module to direct transfer (ablation) | SeqNet baseline · CUHK-SYSU target | mAP: 52.5 -> 62.2 | Improved |
| add dynamic clustering to direct transfer (ablation) | SeqNet baseline · CUHK-SYSU target | mAP: 52.5 -> 70.9 | Improved |
| add DAM to dynamic clustering (ablation) | DAPS · CUHK-SYSU target | mAP: 70.9 -> 72.6 | Improved |
| add hybrid hard case mining (ablation) | DAM+DC · CUHK-SYSU target | mAP: 72.6 -> 73.2 | Improved |
| add hybrid hard case mining (ablation) | DAM+DC · PRW target | mAP: 32.7 -> 34.5 | Improved |
| add detection training on target domain (ablation) | DAM+DC · CUHK-SYSU target | AP: 63.2 -> 71.1 | Improved |
| add all modules (DAM, DC, HM, DTD) (ablation) | SeqNet baseline · CUHK-SYSU target | mAP: 52.5 -> 77.6 | Improved |
| task-sensitive instead of normal instance alignment (ablation) | DAM · PRW target | mAP: 21.7 -> 30.9 | Improved |
| task-sensitive instead of normal instance alignment (ablation) | DAM · CUHK-SYSU target | mAP: 58.2 -> 62.2 | Improved |
| dynamic update instead of static proposal boxes (ablation) | DAPS · PRW target | mAP: 25.3 -> 32.7 | Improved |
| dynamic update instead of static proposal boxes (ablation) | DAPS · CUHK-SYSU target | mAP: 64.0 -> 72.6 | Improved |
| use ground-truth boxes instead of pseudo boxes (ablation) | DAPS · CUHK-SYSU target | mAP: 72.6 -> 73.6 | Mixed |
Closest papers by meaning
- Localization-Anchored Instance Discrimination for Domain Adaptive Person Search (AAAI 2026 · similarity 0.94)
- Mixture of Submodules for Domain Adaptive Person Search (CVPR 2025 · similarity 0.90)
- Doubly Contrastive Learning for Source-Free Domain Adaptive Person Search (AAAI 2025 · similarity 0.88)
- Joint Detection and Identification Feature Learning for Person Search (CVPR 2017 · similarity 0.88)
- Divide and Conquer: Hybrid Pre-training for Person Search (AAAI 2024 · similarity 0.88)
- 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.
