基于双分支注意力特征融合的跨域行人重识别

打开文本图片集

关键词：行人重识别；无监督域适应；注意力机制；特征融合

中图分类号：TP391 文献标志码：A 文章编号：1671-6841（2025）06-0051-07

DOI： 10.13705/j. issn.1671-6841.2024031

Abstract： With unsupervised cross domain pedestrian re-identification technology，labeled information could be transfered from the source domain to the target domain to cope with unlabeled situations，clustering methods were to achieve unsupervised domain adaptation，so that to achieve cross domain pedestrian re-identification.However，clustering based on solely global featureswas susceptible to noise generated by interdomain differences，and single network structure training could lead to eror amplification and affect model performance. A dual-branch attntion-based fusion algorithm was proposed to extract & fuse invariant and specific features to enhance target domain generalization and to reduce clustering noise. At the same time，a symmetric network architecture was introduced for synchronous collaborative training， form a mutually supervised learning mechanism to efectively suppress overfiting problems. Experiments showed that on the Market-1501 and DukeMTMC-ReID datasets，the algorithm significantly improved the （20 mAP and Rank accuracy of unsupervised cross domain pedestrian re-identification.

Key words： person re-identification；unsupervised domain adaptation；attention mechanism；feature fusion

0 引言

随着卷积神经网络的发展，行人重识别领域取得了令人鼓舞的成果，已广泛应用于智能安防领域之中[2]。（剩余11637字）