Minimizing the Pretraining Gap: Domain-aligned Text-Based Person Retrieval
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866918415702163456 |
|---|---|
| author | Yang, Shuyu Wang, Yaxiong Li, Yongrui Zhu, Li Zheng, Zhedong |
| author_facet | Yang, Shuyu Wang, Yaxiong Li, Yongrui Zhu, Li Zheng, Zhedong |
| contents | In this work, we focus on text-based person retrieval, which identifies individuals based on textual descriptions. Despite advancements enabled by synthetic data for pretraining, a significant domain gap, due to variations in lighting, color, and viewpoint, limits the effectiveness of the pretrain-finetune paradigm. To overcome this issue, we propose a unified pipeline incorporating domain adaptation at both image and region levels. Our method features two key components: Domain-aware Diffusion (DaD) for image-level adaptation, which aligns image distributions between synthetic and real-world domains, e.g., CUHK-PEDES, and Multi-granularity Relation Alignment (MRA) for region-level adaptation, which aligns visual regions with descriptive sentences, thereby addressing disparities at a finer granularity. This dual-level strategy effectively bridges the domain gap, achieving state-of-the-art performance on CUHK-PEDES, ICFG-PEDES, and RSTPReid datasets. The dataset, model, and code are available at https://github.com/Shuyu-XJTU/MRA. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_10195 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Minimizing the Pretraining Gap: Domain-aligned Text-Based Person Retrieval Yang, Shuyu Wang, Yaxiong Li, Yongrui Zhu, Li Zheng, Zhedong Computer Vision and Pattern Recognition In this work, we focus on text-based person retrieval, which identifies individuals based on textual descriptions. Despite advancements enabled by synthetic data for pretraining, a significant domain gap, due to variations in lighting, color, and viewpoint, limits the effectiveness of the pretrain-finetune paradigm. To overcome this issue, we propose a unified pipeline incorporating domain adaptation at both image and region levels. Our method features two key components: Domain-aware Diffusion (DaD) for image-level adaptation, which aligns image distributions between synthetic and real-world domains, e.g., CUHK-PEDES, and Multi-granularity Relation Alignment (MRA) for region-level adaptation, which aligns visual regions with descriptive sentences, thereby addressing disparities at a finer granularity. This dual-level strategy effectively bridges the domain gap, achieving state-of-the-art performance on CUHK-PEDES, ICFG-PEDES, and RSTPReid datasets. The dataset, model, and code are available at https://github.com/Shuyu-XJTU/MRA. |
| title | Minimizing the Pretraining Gap: Domain-aligned Text-Based Person Retrieval |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2507.10195 |