四川大学学报（自然科学版）2025，Vol.62Issue(2)：399-411,13.DOI:10.19907/j.0490-6756.240155
基于全局和局部令牌变换的行人再识别算法
Person re-identification algorithm based on global and local token transformer
摘要
Abstract
The field of person Re-Identification holds significant research and application value in the domains of intelligent security,intelligent city and intelligent transportation.However,considering the current re-search and application requirements,it is still challenging to implement person Re-Identification models for real-world scenarios.In this work,a Global and Local Token Transformer(GLTT)framework is proposed to effectively capture the complex image variations of person targets,enabling the acquisition of discrimina-tive and robust person features.Firstly,a Global Token Transformer(GTT)module is introduced into the GLTT framework.Considering that a single class token is difficult to deal with complex image variations,the GTT module uses multiple class tokens to learn multiple global features from different semantic spaces and improve the global robustness of person Re-Identification models.Then,since the local details contain crucial person information,a Local Token Transformer(LTT)module is designed to dynamically select semanti-cally relevant patch tokens by fusing self-attention weights.In addition,the LTT module contributes to infor-mation interaction between the selected patch tokens and class tokens,and therefore enhances the local dis-crimination of person re-identification models.Finally,a simple yet effective approach named Class Token Regularization(CTR)method is proposed to enhance the representation capability of multiple class token fea-tures by ensuring non-overlapping feature spaces for each class token.The experimental results demonstrate that the proposed GLTT framework achieves superior Re-Identification performance on Market1501,CUHK03,DukeMTMC,and MSMT17 datasets,thereby validating the discrimination and robustness of the proposed framework.关键词
行人再识别/全局令牌变换/局部令牌变换/类别令牌正则化Key words
Person re-identification/Global token transformer/Local token transformer/Class token regular-ization分类
信息技术与安全科学引用本文复制引用
吴婷婷,王苹宇,陈洪刚..基于全局和局部令牌变换的行人再识别算法[J].四川大学学报（自然科学版）,2025,62(2):399-411,13.基金项目
国家自然科学基金(62301346) （62301346）
四川省科技计划资助项目(2024NSFSC1424) （2024NSFSC1424）
成都市技术创新研发项目(2024-YF05-00652-SN) （2024-YF05-00652-SN）
四川大学引进人才科研启动经费资助项目(YJ202326) （YJ202326）
成都市科技成果转化示范项目(2023-YF09-00019-SN) （2023-YF09-00019-SN）
四川大学-中国核动力研究设计院联合创新基金 （）
