Multilingual Text-to-Image Person Retrieval via Bidirectional Relation Reasoning and Aligning
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914104419024896 |
|---|---|
| author | Cao, Min Zhou, Xinyu Jiang, Ding Du, Bo Ye, Mang Zhang, Min |
| author_facet | Cao, Min Zhou, Xinyu Jiang, Ding Du, Bo Ye, Mang Zhang, Min |
| contents | Text-to-image person retrieval (TIPR) aims to identify the target person using textual descriptions, facing challenge in modality heterogeneity. Prior works have attempted to address it by developing cross-modal global or local alignment strategies. However, global methods typically overlook fine-grained cross-modal differences, whereas local methods require prior information to explore explicit part alignments. Additionally, current methods are English-centric, restricting their application in multilingual contexts. To alleviate these issues, we pioneer a multilingual TIPR task by developing a multilingual TIPR benchmark, for which we leverage large language models for initial translations and refine them by integrating domain-specific knowledge. Correspondingly, we propose Bi-IRRA: a Bidirectional Implicit Relation Reasoning and Aligning framework to learn alignment across languages and modalities. Within Bi-IRRA, a bidirectional implicit relation reasoning module enables bidirectional prediction of masked image and text, implicitly enhancing the modeling of local relations across languages and modalities, a multi-dimensional global alignment module is integrated to bridge the modality heterogeneity. The proposed method achieves new state-of-the-art results on all multilingual TIPR datasets. Data and code are presented in https://github.com/Flame-Chasers/Bi-IRRA. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_17685 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Multilingual Text-to-Image Person Retrieval via Bidirectional Relation Reasoning and Aligning Cao, Min Zhou, Xinyu Jiang, Ding Du, Bo Ye, Mang Zhang, Min Computer Vision and Pattern Recognition Artificial Intelligence Text-to-image person retrieval (TIPR) aims to identify the target person using textual descriptions, facing challenge in modality heterogeneity. Prior works have attempted to address it by developing cross-modal global or local alignment strategies. However, global methods typically overlook fine-grained cross-modal differences, whereas local methods require prior information to explore explicit part alignments. Additionally, current methods are English-centric, restricting their application in multilingual contexts. To alleviate these issues, we pioneer a multilingual TIPR task by developing a multilingual TIPR benchmark, for which we leverage large language models for initial translations and refine them by integrating domain-specific knowledge. Correspondingly, we propose Bi-IRRA: a Bidirectional Implicit Relation Reasoning and Aligning framework to learn alignment across languages and modalities. Within Bi-IRRA, a bidirectional implicit relation reasoning module enables bidirectional prediction of masked image and text, implicitly enhancing the modeling of local relations across languages and modalities, a multi-dimensional global alignment module is integrated to bridge the modality heterogeneity. The proposed method achieves new state-of-the-art results on all multilingual TIPR datasets. Data and code are presented in https://github.com/Flame-Chasers/Bi-IRRA. |
| title | Multilingual Text-to-Image Person Retrieval via Bidirectional Relation Reasoning and Aligning |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2510.17685 |