电子学报2025，Vol.53Issue(6)：1829-1846,18.DOI:10.12263/DZXB.20250110
基于渐进式混合对比学习的无监督领域自适应行人再识别
Unsupervised Domain Adaptive Person Re-Identification Based on Progressive Hybrid Contrastive Learning
摘要
Abstract
Unsupervised domain adaptive(UDA)person re-identification(Re-ID)seeks to leverage labeled source do-main data to address the task of unsupervised Re-ID in unlabeled target domain data.Recently,contrastive learning has at-tracted attention in this field.However,current methods suffer from small differences in positive sample pairs and overlook biases in negative proxy sampling.To resolve these challenges,this paper presents a progressive hybrid contrastive learn-ing(PHCL)method.In each training epoch,the PHCL method divides the unlabeled dataset into clustered samples with pseudo-labels and un-clustered independent instances through two steps:clustering and progressive refinement.Based on the clustering results,PHCL implements contrastive learning at two different levels:to learn intra-category similarity through bringing together similar samples within the same cluster(target domain)or identity label(source domain)and ex-plores inter-instance discrimination by applying repulsion among un-clustered individual instances.Moreover,the PHCL method generates positive proxies for anchor samples through nearest neighbor mining,increasing the differences among positive sample pairs to learn richer semantic information.Additionally,the PHCL method performs debiasing in the nega-tive proxy sampling process,mitigating the adverse impact of false negative proxies on model training.Experimental results show that the PHCL method achieves mean average precision(mAP)of 85.9％and 42.3％on the Market-1501 and MSMT17 datasets,respectively,which are improvements of 4.3 percentage points and 13.5 percentage points over the base-line model.These results validate the efficacy of the PHCL method for UDA Re-ID.关键词
无监督领域自适应(UDA)行人再识别(Re-ID)/对比学习/伪标签/最近邻挖掘/去偏差Key words
unsupervised domain adaptive(UDA)person re-identification(Re-ID)/contrastive learning/pseudo la-bels/nearest neighbor mining/debiasing分类
信息技术与安全科学引用本文复制引用
赵宇,舒巧媛..基于渐进式混合对比学习的无监督领域自适应行人再识别[J].电子学报,2025,53(6):1829-1846,18.基金项目
重庆市自然科学基金面上项目(No.CSTB2022NSCQ-MSX0404) （No.CSTB2022NSCQ-MSX0404）
重庆市教育委员会科学技术研究项目(No.KJQN202401624) General Program of Chongqing Natural Science Foundation(No.CSTB2022NSCQ-MSX0404) （No.KJQN202401624）
Science and Technology Research Program of Chongqing Municipal Education Commission(No.KJQN202401624) （No.KJQN202401624）
