arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2103.15890cs.CVcs.LG
收藏

学习领域不变表示的广义人再识别

Learning Domain Invariant Representations for Generalizable Person Re-Identification

Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);School of Artificial Intelligence, University of Chinese Academy of Sciences(中国科学院大学人工智能学院);University of Chinese Academy of Sciences(中国科学院大学)

Yi-Fan Zhang, Zhang Zhang, Da Li, Zhen Jia, Liang Wang, Tieniu Tan

摘要:

可推广的人的再识别(ReID)是近年来计算机视觉领域的研究热点。在这项工作中,我们构建了一个结构因果模型之间的身份标签,身份特定的因素(衣服/鞋子的颜色等),和特定领域的因素(背景,观点等)。根据因果关系分析,我们提出了一个新的领域不变的表示学习的可推广的人重新识别(DIR-ReID)框架。具体来说,我们首先提出解开特定身份和特定领域的特征空间,在此基础上,我们提出了一个有效的算法实现后门调整,本质上是作为一个因果干预对SCM。大量的实验表明,DIR-ReID在大规模领域泛化ReID基准测试中的性能优于最先进的方法。

英文摘要:

Generalizable person Re-Identification (ReID) has attracted growing attention in recent computer vision community. In this work, we construct a structural causal model among identity labels, identity-specific factors (clothes/shoes color etc), and domain-specific factors (background, viewpoints etc). According to the causal analysis, we propose a novel Domain Invariant Representation Learning for generalizable person Re-Identification (DIR-ReID) framework. Specifically, we first propose to disentangle the identity-specific and domain-specific feature spaces, based on which we propose an effective algorithmic implementation for backdoor adjustment, essentially serving as a causal intervention towards the SCM. Extensive experiments have been conducted, showing that DIR-ReID outperforms state-of-the-art methods on large-scale domain generalization ReID benchmarks.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑