题名 | SiamCLIM: Text-Based Pedestrian Search Via Multi-Modal Siamese Contrastive Learning |
作者 | |
发表日期 | 2023 |
会议名称 | 30th IEEE International Conference on Image Processing, ICIP 2023 |
会议录名称 | 2023 IEEE International Conference on Image Processing: Proceedings
![]() |
ISBN | 9781728198354 |
页码 | 1800-1804 |
会议日期 | OCT 08-11, 2023 |
会议地点 | Kuala Lumpur, MALAYSIA |
出版地 | NEW YORK, USA |
出版者 | IEEE |
摘要 | Text-based pedestrian search (TBPS) aims at retrieving target persons from the image gallery through descriptive text queries. Despite remarkable progress in recent state-of-the-art approaches, previous works still struggle to efficiently extract discriminative features from multi-modal data. To address the problem of cross-modal fine-grained text-to-image, we proposed a novel Siamese Contrastive Language-Image Model (SiamCLIM). The model implements textual description and target-person interaction through deep bilateral projection, and siamese network structure to capture the relationship between text and image. Experiments show that our model significantly outperforms the state-of-the-art methods on cross-modal fine-grained matching tasks. We conduct the downstream task experiments on the benchmark dataset CUHK-PEDES and the experimental results demonstrate that our model is state-of-the-art and outperforms the current methods by 11.55%, 11.02%, and 7.76% in terms of top-1, top-5, and top-10 accuracy, respectively. © 2023 IEEE. |
关键词 | contrastive learning multi-modal Siamese Network Text-based person search text-image |
DOI | 10.1109/ICIP49359.2023.10222660 |
URL | 查看来源 |
收录类别 | CPCI-S |
语种 | 英语English |
WOS研究方向 | Computer Science |
WOS类目 | Computer Science, Artificial Intelligence ; Computer Science, Theory & Methods |
WOS记录号 | WOS:001106821001175 |
Scopus入藏号 | 2-s2.0-85180805993 |
引用统计 | |
文献类型 | 会议论文 |
条目标识符 | https://repository.uic.edu.cn/handle/39GCC9TT/11536 |
专题 | 理工科技学院 |
通讯作者 | Zhang, Hui |
作者单位 | 1.Department of Computer Science and Technology, BNU-HKBU United International College 2.University of Edinburgh, United Kingdom 3.Department of Computer Science, Hong Kong Baptist University 4.University of Alberta, Canada |
第一作者单位 | 北师香港浸会大学 |
通讯作者单位 | 北师香港浸会大学 |
推荐引用方式 GB/T 7714 | Huang, Runlin,Wu, Shuyang,Jie, Leipinget al. SiamCLIM: Text-Based Pedestrian Search Via Multi-Modal Siamese Contrastive Learning[C]. NEW YORK, USA: IEEE, 2023: 1800-1804. |
条目包含的文件 | 条目无相关文件。 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论