开放式行人再识别研究进展综述  

Review on Research Progress of Open-World Person Re-identification

作　　者：夏道勋 郭方[1] 刘浩杰 夏勇 XIA Daoxun;GUO Fang;LIU Haojie;XIA Yong(School of Big Data and Computer Science,Guizhou Normal University,Guiyang 550025,China;The Engineering Laboratory for Applied Technology of Big Data in Education in Guizhou Province,Guizhou Normal University,Guiyang 550025,China;School of Computer Science,Northwestern Polytechnical University,Xi'an 710129,China)

机构地区：[1]贵州师范大学大数据与计算机科学学院,贵阳550025 [2]贵州师范大学贵州省教育大数据应用技术工程实验室,贵阳550025 [3]西北工业大学计算机学院,西安710129

出　　处：《数据采集与处理》2021年第3期449-467,共19页Journal of Data Acquisition and Processing

基　　金：国家自然科学基金(61762023)资助项目;贵州省自然基金项目新苗培育专项(黔科合平台人才[2017]5726)资助项目;贵州师范大学博士启动基金(2017)资助项目。

摘　　要：开放式行人再识别是在一个未知的空间环境中,候选行人库中并不一定包含有待检索的行人,被视为图像检索的子问题,是比封闭式行人再识别更具挑战和更为实用的应用研究。本文首先总结了开放式行人再识别的发展历程,与封闭式行人再识别的异同,开放式行人再识别建模过程和数据集的比较分析;然后重点总结了开放式行人再识别技术的研究方法,它们分别是以数据为驱动的研究方法、以效率为驱动的研究方法和以实际应用为驱动的研究方法;最后对开放式行人再识别技术的研究难点和未来发展进行了较为详细的分析和探讨。Open-world person re-identification(Re-ID)is a task that the query person may not occur in the gallery set in an agnostic space environment.Compared with the closed-world person Re-ID,open-world Re-ID is seen as a subproblem of image retrieval,and is a more challenging and practical application research.We firstly compare and analyze the different pedestrian datasets and describe the development of existing open-set Re-ID works,the discrepancies between closed-and open-world Re-ID,and the modeling process of open-world Re-ID.Then,we focus on summarizing the research methods of openworld Re-ID,which are data-driven research methods,efficiency-driven research methods and applicationdriven research methods.Finally,the issues of the research and future directions for open-world Re-ID are discussed.

关 键 词：开放式行人再识别 目标检测 度量学习 深度学习 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]