引入垂直全局表观约束的中层特征行人再识别

doi:10.11959/j.issn.1000-0801.2017023

电信科学 ›› 2017, Vol. 33 ›› Issue (4): 114-125.doi: 10.11959/j.issn.1000-0801.2017023

引入垂直全局表观约束的中层特征行人再识别

王亚东,张荣(),蒋桧慧,郭立君

宁波大学信息科学与工程学院，浙江宁波 315211

出版日期:2017-04-01 发布日期:2017-04-26
作者简介:王亚东（1990-），男，宁波大学信息科学与工程学院硕士生，主要研究方向为计算机视觉与模式识别。|张荣（1974-），女，博士，宁波大学副教授，主要研究方向为数字取证与信息安全。|蒋桧慧（1993-），女，宁波大学信息科学与工程学院硕士生，主要研究方向为计算机视觉与模式识别。|郭立君（1970-），男，博士，宁波大学教授，主要研究方向为计算机视觉与模式识别、移动互联网及其应用。
基金资助:
国家自然科学基金资助项目(61175026);浙江省自然科学基金资助项目(LY17F030002);宁波大学胡岚博士基金资助项目(ZX2013000319)

Person re-identification using mid-level features with vertical global appearance constraint

Yadong WANG,Rong ZHANG(),Huihui JIANG,Lijun GUO

College of Information Science and Engineering,Ningbo University,Ningbo 315211,China

Online:2017-04-01 Published:2017-04-26
Supported by:
The National Natural Science Foundation of China(61175026);Zhejiang Provincial Natural Science Foundation of China(LY17F030002);Zhejiang Open Foundation of the Most Important Subjects(xkxl1516);Zhejiang Open Foundation of the Most Important Subjects(xkxl1521);Dr.Hu Lan Foundation of Ningbo University(ZX2013000319)

摘要/Abstract

摘要：

行人再识别问题中，包含语义信息的中层特征能够提供更强的判别力。由于中层特征也采用局部匹配方式，与底层特征一样存在由于不同行人部分表观区域比较相似而产生误匹配问题。考虑到行人几乎都处于站立姿态，同一行人在垂直方向上的表观序列比不同行人的更相似，提出了在中层特征的基础上引入行人垂直全局表观约束，并融合底层稠密块匹配的识别方法。实验结果表明，算法在最具挑战的公用VIPeR数据库和CUHK01数据库上，均取得了比现有方法更高的命中率。

关键词: 行人再识别, 中层特征, 垂直表观约束

Abstract:

Mid-level features with the semantic information can provide stronger discrimination in the person re-identification than low-level features.But like the low-level features,the mid-level features also use local feature matching methods and easily leads mismatch problem when different pedestrians have similar appearance features in some local areas.Considering the same pedestrians are more similar than different pedestrians in the vertical direction since pedestrians are almost always in a standing position,global vertical appearance constraints was introduced.Furtherly,a method for person re-identification was proposed which fuses the low-level densely patch-matching and the mid-level features with the global vertical appearance constraints.Experimental results show that the proposed algorithm can achieve a higher hit rate than the existing methods on the most challenging public VIPeR database and CUHK01 database.

Key words: person re-identification, mid-level feature, apparent constraint

中图分类号:

TP391

王亚东,张荣,蒋桧慧,郭立君. 引入垂直全局表观约束的中层特征行人再识别[J]. 电信科学, 2017, 33(4): 114-125.

Yadong WANG,Rong ZHANG,Huihui JIANG,Lijun GUO. Person re-identification using mid-level features with vertical global appearance constraint[J]. Telecommunications Science, 2017, 33(4): 114-125.

图/表 10

图1

图2

图3

图4

图5

图6

表1

图7

表2

图8

参考文献 25

[1]	VEZZANI R , BALTIERI D , CUCCHIARA R . People reidentification in surveillance and forensics:a survey[J]. ACM Computing Surveys, 2013,46(2):1-37.
[2]	WANG X , ZHAO R . Person re-identification:system design and evaluation overview[M]. Berlin:Springer-Verlag, 2014:351-370.
[3]	黄凯奇，陈晓棠，康运锋，等. 智能视频监控技术综述[J]. 计算机学报， 2015,38(6):1093-1118.
	HUANG K Q , CHEN X T , KANG Y F , et al. Intelligent visual surveillance:a review[J]. Chinese Journal of Computers, 2015,38(6):1093-1118.
[4]	ZHAO R , OUYANG W , WANG X . Unsupervised salience learning for person re-identification[C]// CVPR, June 25-27,2013,Portland,Oregon. New Jersey: IEEE Press, 2013:3586-3593.
[5]	ZHAO R , OUYANG W , WANG X . Person re-identification by salience matching[C]// IEEE International Conference on Computer Vision, December 1-8,2013,Sydney,Australia. New Jersey: IEEE Press, 2013:2528-2535.
[6]	陈普强，郭立君，张荣，等. 基于全局空间约束块匹配的目标人体识别[J]. 计算机研究与发展， 2015(3):596-605.
	CHEN P Q , GUO L J , ZHANG R , et al. Patch matching with global spatial constraints for person re-identification[J]. Journal of Computer Research and Development, 2015(3):596-605.
[7]	FARENZENA M , BAZZANI L , PERINA A , et al. Person re-identification by symmetry-driven accumulation of local features[C]// CVPR, June 13-18,2013,San Francisco,California,USA. New Jersey: IEEE Press, 2010:2360-2367.
[8]	CHENG D S , CRISTANI M , STOPPA M , et al. Custom pictorial structures for re-identification[C]// BMVC, August 22-September 2,2011,Dundee,USA. New Jersey: IEEE Press, 2011,1(2):6.
[9]	PROSSER B , ZHENG W S , GONG S , et al. Person re-identification by support vector ranking[C]// BMVC, April 4-9,2010,New York,USA. New Jersey: IEEE Press, 2010:6.
[10]	SINGH S , GUPTA A , EFROS A A . Unsupervised discovery of mid-level discriminative patches[M]. Berlin: Springer Heidelberg, 2012:73-86.
[11]	JAIN A , GUPTA A , RODRIGUEZ M , et al. Representing videos using mid-level discriminative patches[C]// CVPR, June 25-27,2013,Portland,Oregon. New Jersey: IEEE Press, 2013:2571-2578.
[12]	LAYNE R , HOSPEDALES T M , GONG S , et al. Person re-identification by attributes[C]// BMVC, April 11-17,2012,Surrey,USA New Jersey: IEEE Press, 2012:8.
[13]	LIU X , SONG M , ZHAO Q , et al. Attribute-restricted latent topic model for person re-identification[J]. Pattern Recognition, 2012,45(12):4204-4213.
[14]	ZHAO R , OUYANG W , WANG X , et al. Learning mid-level filters for person re-identification[C]// IEEE Conference on Computer Vision and Pattern Recognition, June 24-27,2014,Columbus,Ohio,USA.New Jersey: IEEE Press, 2014:144-151.
[15]	MA K , BEN-ARIE J . Vector array based multi-view face detection with compound exemplars[C]// 2012 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), October 11-13,2012,Providence,USA.New Jersey: IEEE Press, 2012:3186-3193.
[16]	HU Y , LIAO S , LEI Z , et al. Exploring structural information and fusing multiple features for person re-identification[C]// CVPR, June 25-27,2013,Oregon,Portland.New Jersey: IEEE Press, 2013:794-799.
[17]	ZHENG W S , GONG S , XIANG T . Reidentification by relative distance comparison[J]. IEEE transactions on Pattern Analysis and Machine Intelligence, 2013,35(3):653-668.
[18]	LI Z , CHANG S , LIANG F , et al. Learning locally-adaptive decision functions for person verification[C]// Conference on Computer Vision and Pattern Recognition, June 25-27,2013,Portland,Oregon.New Jersey: IEEE Press, 2013:3610-3617.
[19]	ZHANG W , WANG X , ZHAO D , et al. Graph degree linkage:agglomerative clustering on a directed graph[C]// European Conference on Computer Vision, April 11-14,2012,Berlin,German.Berlin: Springer, 2012:428-441.
[20]	YANG Y , RAMANNAN D . Articulated pose estimation with flexible mixtures-of-parts[C]// IEEE Conference on Computer Vision and Pattern Recognition, June 20-25,2011,Colorado Springs,USA.New Jersey: IEEE Press, 2011:1385-1392.
[21]	PEDAGADI S , ORWELL J , VELASTIN S , et al. Local fisher discriminant analysis for pedestrian re-identification[C]// CVPR, June 25-27,2013:3318-3325.
[22]	K?STINGER M , HIRZER M , WOHLHART P , et al. Large scale metric learning from equivalence constraints[C]// Conference on Computer Vision and Pattern Recognition,Providence, June 16-21,2012,Providence,RI,USA.New Jersey: IEEE Press, 2012:2288-2295.
[23]	ZHENG W S , GONG S , XIANG T . Person re-identification by probabilistic relative distance comparison[C]// Conference on Computer Vision and Pattern recognition, June 20-25,2011,Colorado Springs,USA.New Jersey: IEEE Press, 2011:649-656.
[24]	MIGNON A , JURIE F . PCCA:a new approach for distance learning from sparse pairwise constraints[C]// Conference on Computer Vision and Pattern Recognition, June 16-21,2012,Providence,RI,USA.New Jersey: IEEE Press, 2012:2666-2672.
[25]	LI W , ZHAO R , WANG X . Human reidentification with transferred metric learning[C]// Asian Conference on Computer Vision, March 1-5,2012,Berlin,German.New Jersey: IEEE Press, 2012:31-44.

方法	r=1	r=5	r=10	r=20
SDALF	19.87%	38.89%	49.37%	65.73%
eSDC	26.31%	46.60%	58.86%	72.76%
PRDC	15.66%	38.42%	53.86%	70.09%
PCCA	19.27%	48.89%	64.91%	80.28%
KISSME	19.60%	48%	62.20%	77%
LF	24.18%	52%	67.12%	82%
SalMatch	30.16%	52.31%	65.54%	79.14%
LADF	29.34%	61.04%	75.98%	88.10%
LMF	29.11%	52.34%	65.95%	79.87%
本文算法	30.04%	53.27%	66.21%	79.50%
LMF+LADF	43.29%	73.04%	84.87%	93.70%
本文算法+LADF	44.38%	74.14%	85.56%	93.97%

方法	r=1	r=5	r=10	r=20
SDALF	9.90%	22.57%	30.33%	41.03%
eSDC	19.67%	32.71%	40.29%	50.58%
LMNN	13.45%	31.33%	42.25%	54.11%
ITML	15.98%	35.22%	45.60%	59.81%
GenericMetric	20%	43.58%	56.04%	69.27%
SalMatch	28.45%	45.85%	55.67%	67.95%
LMF	34.30%	55.06%	64.95%	74.94%
本文算法	35.85%	55.80%	65.89%	76.39%

引入垂直全局表观约束的中层特征行人再识别

Person re-identification using mid-level features with vertical global appearance constraint

在线阅读

PDF下载

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 10

参考文献 25

相关文章 15

Metrics

推荐阅读 0

[1]	金宏辉, 简志华, 杨曼, 吴超. 采用圆周局部三值模式纹理特征的合成语音检测方法[J]. 电信科学, 2023, 39(6): 85-95.
[2]	马辉, 王瑞琴, 杨帅. 一种渐进式增长条件生成对抗网络模型[J]. 电信科学, 2023, 39(6): 105-113.
[3]	卢敏, 胡娟, 张先超, 丁伟健, 乐光学. 基于用户多特征融合的个性化推荐模型[J]. 电信科学, 2023, 39(5): 101-115.
[4]	张永, 刘纪奎, 柯文龙. 基于并行可分离卷积和标签平滑正则化的脑电情感识别[J]. 电信科学, 2023, 39(5): 116-128.
[5]	邓琨, 蒋庆丰, 刘星妍. 融合节点分析与边分析的复杂网络社区识别算法[J]. 电信科学, 2023, 39(4): 87-100.
[6]	冶莉娟, 王亦婷, 朱励程. 基于细胞自动机模型电力网络攻击预测技术[J]. 电信科学, 2023, 39(4): 173-179.
[7]	韩一士, 徐雨欣, 卢甜甜. 一种基于耦合网络的RD-IHSAT网络谣言传播模型[J]. 电信科学, 2023, 39(2): 118-131.
[8]	徐嘉, 简志华, 金宏辉, 吴超, 游林, 吴迎笑. 基于中心对称局部二值模式的合成伪装语音检测方法[J]. 电信科学, 2023, 39(1): 72-78.
[9]	任华健, 郝秀兰, 徐稳静. 融合递增词汇选择的深度学习中文输入法[J]. 电信科学, 2022, 38(12): 56-64.
[10]	周薇娜, 刘露. 复杂场景下多尺度船舶实时检测方法[J]. 电信科学, 2022, 38(10): 67-78.
[11]	金楠, 王瑞琴, 陆悦聪. 基于艾宾浩斯遗忘曲线和注意力机制的推荐算法[J]. 电信科学, 2022, 38(10): 89-97.
[12]	杨帅, 王瑞琴, 马辉. 基于多通道的边学习图卷积网络[J]. 电信科学, 2022, 38(9): 95-104.
[13]	赵东明. 电信运营商知识图谱技术体系研究及应用实践[J]. 电信科学, 2022, 38(8): 151-162.
[14]	于佳祺, 简志华, 徐嘉, 游林, 汪云路, 吴超. 基于联合特征与随机森林的伪装语音检测[J]. 电信科学, 2022, 38(6): 91-99.
[15]	申情, 郭文宾, 楼俊钢, 余强国. 考虑多层次潜在特征的个性化推荐模型[J]. 电信科学, 2022, 38(2): 71-83.