基于区域特征对齐与k倒排编码的行人再识别方法

doi:10.19678/j.issn.1000-3428.0050802

计算机工程 ›› 2019, Vol. 45 ›› Issue (3): 207-211. doi: 10.19678/j.issn.1000-3428.0050802

基于区域特征对齐与k倒排编码的行人再识别方法

库浩华^a,周萍^b,蔡晓东^a,杨海燕^a,梁晓曦^a

桂林电子科技大学 a.信息与通信学院; b.电子工程与自动化学院,广西桂林 541004

收稿日期:2018-03-15 出版日期:2019-03-15 发布日期:2019-03-15
作者简介:库浩华(1993—),男,硕士研究生,主研方向为图像处理;周萍,教授;蔡晓东(通信作者),教授、博士;杨海燕,副教授、博士研究生;梁晓曦,硕士研究生。
基金资助:
“认知无线电与信息处理”省部共建教育部重点实验室基金(CRKL160102);广西重点研发计划(桂科AB16380264)。

Person Re-identification Method Based on Regional Feature Alignment and k-reciprocal Encoding

KU Haohua^a,ZHOU Ping^b,CAI Xiaodong^a,YANG Haiyan^a,LIANG Xiaoxi^a

a.School of Information and Communication; b.School of Electronic Engineering and Automation, Guilin University of Electronic Technology,Guilin,Guangxi 541004,China

Received:2018-03-15 Online:2019-03-15 Published:2019-03-15

摘要/Abstract

摘要：

在行人再识别过程中,由于行人姿态变化会导致图像之间对应位置存在身体区域不对齐的问题,从而降低识别准确率。为此,设计一种新的行人再识别方法。利用卷积神经结构计算行人图像的响应图,根据响应图中的极值点定位行人身体节点,并以此划分特征区域,将提取的各个区域的特征进行融合得到特征表示。在比对距离度量上通过引入k倒排近邻使更多的正样本包含在近邻中,在杰卡德距离中将k倒排近邻集编码成向量以减少计算量,使得越近的邻域获得越大的权重。实验结果表明,相比于对整幅行人图像提取特征方法与单独使用马氏距离的方法,该方法能有效提高行人再识别的准确率。

关键词: 身体节点, 身体区域, k倒排近邻, 杰卡德距离, 行人再识别

Abstract:

In the process of person re-identification,pose variations lead to a problem of misalignment between spatial areas of images,which results in low recognition rate.So this paper proposes a new person re-identification method.Firstly,a convolution neural structure is utilized to calculate responding maps of person images,and the body joints are located according to the extreme points in the responding maps.Secondly,body sub-regions are divided according to the positions of body joints,and they are aligned before further feature extraction.Finally,the features of body sub-regions are fused for identification.By introducing the k-reciprocal nearest neighbor method,more positive samples can be included in the nearest neighbors.With the Jaccard distance,the computation cost is reduced by encoding the k-reciprocal nearest neighbor sets into a vector,which assigns larger weights to closer neighbors.Experimental results show that compared with the feature extraction from the whole person image and using Mahalanobis distance alone,the proposed method can improve the accuracy of person re-identification significantly.

Key words: body joints, body sub-regions, k-reciprocal nearest neighbor, Jaccard distance, person re-identification

中图分类号:

TP753

库浩华,周萍,蔡晓东,杨海燕,梁晓曦. 基于区域特征对齐与k倒排编码的行人再识别方法[J]. 计算机工程, 2019, 45(3): 207-211.

KU Haohua,ZHOU Ping,CAI Xiaodong,YANG Haiyan,LIANG Xiaoxi. Person Re-identification Method Based on Regional Feature Alignment and k-reciprocal Encoding[J]. Computer Engineering, 2019, 45(3): 207-211.

http://www.ecice06.com/CN/Y2019/V45/I3/207

参考文献

［1］俞婧,仇春春,王恬,等.基于距离匹配的行人再识别技术综述［J］.微处理机,2016,37(3):77-80.
［2］OJALA T,PIETIKAINEN M,MAENPAA T.Multi-resolution gray-scale and rotation invariant texture classification with local binary patterns［J］.Classification with Local Binary Patterns,2002,24(7):971-987.
［3］FARENZENA M,BAZZANI L,PERINA A,et al.Person re-identification by symmetry-driven accumulation of local features［C］//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2010:2360-2367.
［4］CHENG D S,CRISTANI M,STOPPA M,et al.Custom pictorial structures for re-identification［C］//Proceedings of the British Machine Vision Conference.Washington D.C.,USA:IEEE Press,2011:1-11.
［5］LI W,ZHAO R,XIAO T,et al.DeepReID:deep filter pairing neural network for person re-identification［C］//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.USA:IEEE Press,2014:152-159.
［6］AHMED E,JONES M J,MARKS T K.An improved deep learning architecture for person re-identification［C］//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2015:3908-3916.
［7］SZEGEDY C,LIU W,JIA Y Q,et al.Going deeper with convolutions［C］//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2015:1-9.
［8］CHENG D,GONG Y H,ZHOU S P,et al.Person re-identification by multi-channel parts-based CNN with improved triplet loss function［C］//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2016:1335-1344.
［9］XIAO T,LI H,OUYANG W,et al.Learning deep feature representations with domain guided dropout for person re-identification［C］//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2016:1249-1258.
［10］WU S X,CHEN Y C,LI X,et al.An enhanced deep feature representation for person re-identification［C］//Proceedings of IEEE Winter Conference on Applications of Computer Vision.Washington D.C.,USA:IEEE Press,2016:1-8.
［11］ZHAO H Y,TIAN M Q,SUN S Y,et al.Spindle net:person re-identification with human body region guided feature decomposition and fusion［C］//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2017:907-915.
［12］ZHONG Z,ZHENG L,CAO D L,et al.Re-ranking person re-identification with k-reciprocal encoding［C］//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2017:3652-3661.
［13］宋宗涛,陈岳林,蔡晓东.基于多分块三重损失计算的行人识别方法［J］.电子与信息学报,2016,38(6):1528-1535.
［14］QIN D F,GAMMETER S,BOSSARD L,et al.Hello neighbor:accurate object retrieval with k-reciprocal nearest neighbors［C］//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Washington D.C.,USA:IEEE Press,2011:777-784.
［15］ZHENG L,SHEN L Y,TIAN L,et al.Scalable person re-identification:a benchmark［C］//Proceedings of IEEE Conference on Computer Vision.Washington D.C.,USA:IEEE Press,2015:1116-1124.

[1]	唐佳敏, 韩华, 黄丽. 行人再识别中基于无监督学习的粗细粒度特征提取[J]. 计算机工程, 2022, 48(4): 269-275,283.
[2]	周东明, 张灿龙, 唐艳平, 李志欣. 联合语义分割与注意力机制的行人再识别模型[J]. 计算机工程, 2022, 48(2): 201-206.
[3]	姜国权, 肖禛禛, 霍占强. 融合RGB与灰度图像特征的行人再识别方法[J]. 计算机工程, 2021, 47(4): 226-233,240.
[4]	郑烨, 赵杰煜, 王翀, 张毅. 基于姿态引导对齐网络的局部行人再识别[J]. 计算机工程, 2020, 46(5): 247-253.
[5]	陶飞, 成科扬, 张建明, 汤宇豪. 基于姿态与并行化属性学习的行人再识别方法[J]. 计算机工程, 2020, 46(3): 246-253.
[6]	吴彦丞, 陈鸿昶, 李邵梅, 高超. 基于行人属性异质性的行人再识别神经网络模型[J]. 计算机工程, 2018, 44(10): 196-203.

选择文件类型/文献管理软件名称

选择包含的内容

基于区域特征对齐与k倒排编码的行人再识别方法

Person Re-identification Method Based on Regional Feature Alignment and k-reciprocal Encoding

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 6

编辑推荐

Metrics

本文评价

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

基于区域特征对齐与k倒排编码的行人再识别方法

Person Re-identification Method Based on Regional Feature Alignment and k-reciprocal Encoding

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 6

编辑推荐

Metrics

本文评价