许潇月
江南大学轻工过程先进控制教育部重点实验室 无锡 214000
基金项目:国家自然科学基金(61573168)
详细信息
作者简介:陈莹:女,1976年生,教授,博士生导师,研究方向为模式识别、信息融合
许潇月:女,1994年生,硕士生,研究方向为行人再识别
通讯作者:陈莹 chenying@jiangnan.edu.cn
中图分类号:TN911.73; TP391计量
文章访问数:2495
HTML全文浏览量:926
PDF下载量:49
被引次数:0
出版历程
收稿日期:2019-03-18
修回日期:2019-05-24
网络出版日期:2019-07-03
刊出日期:2020-02-19
Matrix Metric Learning for Person Re-identification Based on Bidirectional Reference Set
Ying CHEN,,Xiaoyue XU
Key Laboratory of Advanced Control Light Process, Jiangnan University, Wuxi 214000, China
Funds:The National Natural Science Foundation of China (61573168)
摘要
摘要:针对行人再识别中由于外观差异不显著导致特征描述不准确的问题,该文提出一种基于双向参考集矩阵度量学习(BRM2L)的行人再识别算法。首先通过互近邻算法获得每个摄像头下的互近邻参考集,为保证参考集的鲁棒性,联合考虑各摄像头下的互近邻参考集获得双向参考集。通过双向参考集挖掘出困难样本进行特征描述,从而得到准确的外观差异描述。最后利用该特征描述进行更有效的矩阵度量学习。在多个公开数据集上的实验结果证明了该算法比现有算法具有更好的行人再识别性能。
关键词:行人再识别/
外观差异/
矩阵度量/
互近邻/
双向参考集
Abstract:To solve the problem of inaccurate feature representation caused by indistinctive appearance difference in person re-identification domain, a new Matrix Metric Learning algerithm based on Bidirectional Reference (BRM2L) set is proposed. Firstly, reciprocal-neighbor reference sets in different camera views are respectively constructed by the reciprocal-neighbor scheme. To ensure the robustness of reference sets, the reference sets in different camera views are jointly considered to generate the Bidirectional Reference Set (BRS). With hard samples which are mined by the BRS to represent feature descriptors, accurate appearance difference representations could be obtained. Finally, these representations are utilized to conduct more effective matrix metric learning. Experimental results on several public datasets demonstrate the superiority of the proposed method.
Key words:Person re-identification/
Appearance difference/
Matrix metric/
Reciprocal neighbor/
Bidirectional reference set
PDF全文下载地址:
https://jeit.ac.cn/article/exportPdf?id=efa5ffad-0803-48b1-907b-99be81668ed6