Multi-granularity feature fusion for person re-identification

Liang ZHANG; Jin CHE; Qi Yang

doi:10.3788/YJYXS20203506.0555

您当前的位置：

首页 >

文章列表页 >

Multi-granularity feature fusion for person re-identification

Image Processing | 更新时间：2020-07-13

- Multi-granularity feature fusion for person re-identification
- Chinese Journal of Liquid Crystals and Displays Vol. 35, Issue 6, Pages: 555-563(2020)
- 作者机构：
  
  1.宁夏大学物理与电子电气工程学院, 宁夏银川 750021
  2.宁夏大学沙漠信息智能感知重点实验室, 宁夏银川 750021
- 作者简介：
- 基金信息：
  
  National Natural Science Foundation of China(61861037)
- DOI：10.3788/YJYXS20203506.0555
  CLC： TP391.4
- Received：04 November 2019，
  
  Revised：25 November 2019，
  
  Accepted：25 November 2019，
  
  Published：05 June 2020
- 稿件说明：
移动端阅览
Liang ZHANG, Jin CHE, Qi Yang. Multi-granularity feature fusion for person re-identification[J]. Chinese journal of liquid crystals and displays, 2020, 35(6): 555-563.
DOI：

Liang ZHANG, Jin CHE, Qi Yang. Multi-granularity feature fusion for person re-identification[J]. Chinese journal of liquid crystals and displays, 2020, 35(6): 555-563. DOI： 10.3788/YJYXS20203506.0555.

摘要

结合全局特征和局部特征是提高行人再识别精度的一种途径。现有的算法通常从人体特定的语义区域提取特征，由于没有将人体结构考虑在内，增加了学习难度，在差异较大的场景下效率和鲁棒性较差。为了较好地解决上述问题，本文提出一种融合了全局特征、局部特征以及人体结构特征的多粒度特征融合的行人再识别算法。本算法不引入任何人体结构先验知识，在特征提取方面，采用均值池化和最大池化对特征图加权得到强辨识性的全局特征。对特征图切片得到局部特征，在原有局部特征的基础上，引入局部相对特征作为人体结构特征。在度量方面，采用三元组损失与ID损失在不同尺度下的多级监督机制。在Market1501、DukeMTMC-reID的实验表明，算法的Rank-1指标相比于部分卷积基线（PCB）方法提升了1.3%、3.9%，平均精度均值（mAP）提升了5.1%、9.8%。

Abstract

Combining global features and local features is one way to improve person re-identification accuracy. Existing algorithms usually extract features from specific semantic regions of the human body. Since the human body structure is not taken into account

the learning difficulty is increased

and the efficiency and robustness are poor in the scenes with large differences. In order to solve the above problems

this paper proposes a person re-identification algorithm based on multi granularity feature fusion

which combines global feature

local feature and human structure feature. The algorithm does not introduce any prior knowledge of human body structure. In terms of feature extraction

the average pooling and maximum pooling are used to weight the feature map to obtain strong global features. Local features are obtained by slicing the feature map. Based on the original local features

local relative features are introduced as human structural features. In terms of metrics

a multi-level supervision mechanism with triple loss and ID loss at different scales is used. Experiments on Market1501 and DukeMTMC-reID show that the Rank-1 index of the algorithm is 1.3% and 3.9% higher than Part-based Convolutional Baseline(PCB) method

and the mean Average Precision(mAP) is 5.1% and 9.8% higher than PCB method.

关键词

Keywords

references

LIAO S C, HU Y, ZHU X Y, et al . Person re-identification by local maximal occurrence representation and metric learning[C]// Proceedings of 2015 IEEE Conference on Computer Vision and Pattern Recognition ( CVPR ). Boston, MA: IEEE, 2015: 2197-2206.

陈亮雨 , 李卫疆 . 多形状局部区域神经网络结构的行人再识别 . 中国图象图形学报 , 2019 . 24 ( 11 ): 1932 - 1941 . DOI: 10.11834/jig.190042 http://doi.org/10.11834/jig.190042 DOI: 10.11834/jig.190042 http://doi.org/10.11834/jig.190042 .

L Y CHEN , W J LI . Multishape part network architecture for person re-identification . Journal of Image and Graphics , 2019 . 24 ( 11 ): 1932 - 1941 . DOI: 10.11834/jig.190042 http://doi.org/10.11834/jig.190042 DOI: 10.11834/jig.190042 http://doi.org/10.11834/jig.190042 .

蒋建国 , 杨宁 , 齐美彬 , 等 . 区域块分割与融合的行人再识别 . 中国图象图形学报 , 2019 . 24 ( 4 ): 513 - 522 . http://d.old.wanfangdata.com.cn/Periodical/zgtxtxxb-a201904003 http://d.old.wanfangdata.com.cn/Periodical/zgtxtxxb-a201904003 .

J G JIANG , N YANG , M B QI , 等 . Person re-identification with region block segmentation and fusion . Journal of Image and Graphics , 2019 . 24 ( 4 ): 513 - 522 . http://d.old.wanfangdata.com.cn/Periodical/zgtxtxxb-a201904003 http://d.old.wanfangdata.com.cn/Periodical/zgtxtxxb-a201904003 .

朱福庆 , 孔祥维 , 付海燕 , 等 . 两路互补对称CNN结构的行人再识别 . 中国图象图形学报 , 2018 . 23 ( 7 ): 1052 - 1060 . http://d.old.wanfangdata.com.cn/Periodical/zgtxtxxb-a201807012 http://d.old.wanfangdata.com.cn/Periodical/zgtxtxxb-a201807012 .

F Q ZHU , X W KONG , H Y FU , 等 . Two-stream complementary symmetrical CNN architecture for person re-identification . Journal of Image and Graphics , 2018 . 23 ( 7 ): 1052 - 1060 . http://d.old.wanfangdata.com.cn/Periodical/zgtxtxxb-a201807012 http://d.old.wanfangdata.com.cn/Periodical/zgtxtxxb-a201807012 .

陈兵 , 查宇飞 , 李运强 , 等 . 基于卷积神经网络判别特征学习的行人重识别 . 光学学报 , 2018 . 38 ( 7 ): 0720001 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=gxxb201807032 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=gxxb201807032 .

B CHEN , Y F ZHA , Y Q LI , 等 . Person re-identification based on convolutional neural network discriminative feature learning . Acta Optica Sinica , 2018 . 38 ( 7 ): 0720001 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=gxxb201807032 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=gxxb201807032 .

ZHENG L, YANG Y, HAUPTMANN A G. Person re-identification: past, present and future[J]. arXiv , 2016: 1610.02984.

LIN Y T, ZHENG L, ZHENG Z D, et al . Improving person re-identification by attribute and identity learning[J]. ArXiv, 2017: 1703.07220.

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 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) . Las Vegas, NV: IEEE, 2016: 1335-1344.

HERMANS A, BEYER L, LEIBE B. In defense of the triplet loss for person re-identification[J]. ArXiv, 2017: 1703.07737.

Y F SUN , Q XU , Y L LI , 等 . Perceive where to focus:learning visibility-aware part-level features for partial person re-identification . ArXiv , 2019 . 1904 http://cn.bing.com/academic/profile?id=fc6c295ef4a02d218eb4977cf4ddc515&encoded=0&v=paper_preview&mkt=zh-cn http://cn.bing.com/academic/profile?id=fc6c295ef4a02d218eb4977cf4ddc515&encoded=0&v=paper_preview&mkt=zh-cn .

VARIOR R R, SHUAI B, LU J W, et al . A Siamese long short-term memory architecture for human re-identification[C]// Proceedings of the 14 th European Conference on Computer Vision . Amsterdam, The Netherlands: Springer, 2016: 135-153.

SUN Y F, ZHENG L, YANG Y et al . Beyond part models: person retrieval with refined part pooling (and a strong convolutional baseline)[C]// Proceedings of the 15 th European Conference on Computer Vision . Munich, Germany: Springer, 2018: 501-518.

ZHANG X, LUO H, FAN X, et al . AlignedReID: surpassing human-level performance in person re-identification[J]. ArXiv, 2017: 1711.08184.

ZHENG L, HUANG Y J, LU H C, et al . Pose invariant embedding for deep person re-identification[J]. ArXiv, 2017: 1701.07732.

Z D ZHENG , L ZHENG , Y YANG . Pedestrian alignment network for large-scale person re-identification . IEEE Transactions on Circuits and Systems for Video Technology , 2019 . 29 ( 10 ): 3037 - 3045 . DOI: 10.1109/TCSVT.2018.2873599 http://doi.org/10.1109/TCSVT.2018.2873599 .

SARFRAZ M S, SCHUMANN A, EBERLE A, et al . A pose-sensitive embedding for person re-identification with expanded cross neighborhood re-ranking[C]// Proceedings of 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition . Salt Lake City, UT: IEEE, 2018: 420-429.

SU H Y, WANG J H, TANG S Y, et al . Part-aligned bilinear representations for person re-identification[C]// Proceedings of the 15 th European Conference on Computer Vision . Munich, Germany: Springer, 2018: 418-437.

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 2017 IEEE Conference on Computer Vision and Pattern Recognition ( CVPR ). Honolulu, HI: IEEE, 2017: 907-915.

L H WEI , S L ZHANG , H T YAO , 等 . GLAD:global-local-alignment descriptor for scalable person re-identification . IEEE Transactions on Multimedia , 2019 . 21 ( 4 ): 986 - 999 . DOI: 10.1109/TMM.2018.2870522 http://doi.org/10.1109/TMM.2018.2870522 .

SU C, LI J N, ZHANG S L, et al . Pose-driven deep convolutional model for person re-identification[C]// Proceedings of 2017 IEEE International Conference on Computer Vision (ICCV) . Venice: IEEE, 2017: 3980-3989.

LI W, ZHU X T, GONG S G. Harmonious attention network for person re-identification[C]// Proceedings of 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition . Salt Lake City, UT: IEEE, 2018: 2285-2294.

CHEN Y B, ZHU X T, GONG S G, et al . Person re-identification by deep learning multi-scale representations[C]// Proceedings of 2017 IEEE International Conference on Computer Vision Workshops (ICCVW) . Venice: IEEE, 2017: 2590-2600.

WANG Y, WANG L Q, YOU Y R et al . Resource aware person re-identification across multiple resolutions[C]// Proceedings of 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition . Salt Lake City, UT: IEEE, 2018: 8042-8051.

贠卫国 , 史其琦 , 王民 . 基于深度卷积神经网络的多特征融合的手势识别 . 液晶与显示 , 2019 . 34 ( 4 ): 417 - 422 . http://d.old.wanfangdata.com.cn/Periodical/yjyxs201904013 http://d.old.wanfangdata.com.cn/Periodical/yjyxs201904013 .

W G YUN , Q Q SHI , M WANG . Multi-feature fusion gesture recognition based on deep convolutional neural networks . Chinese Journal of Liquid Crystals and Displays , 2019 . 34 ( 4 ): 417 - 422 . http://d.old.wanfangdata.com.cn/Periodical/yjyxs201904013 http://d.old.wanfangdata.com.cn/Periodical/yjyxs201904013 .

HE K M, ZHANG X Y, REN S Q, et al . Deep residual learning for image recognition[C]// Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition ( CVPR ). Las Vegas, NV: IEEE, 2016: 770-778.

A HERMANS , L BEYER , B LEIBE . In defense of the triplet loss for person re-identification . ArXiv , 2017 . 1703 http://cn.bing.com/academic/profile?id=81d34407458c559797e4074bd4ff9657&encoded=0&v=paper_preview&mkt=zh-cn http://cn.bing.com/academic/profile?id=81d34407458c559797e4074bd4ff9657&encoded=0&v=paper_preview&mkt=zh-cn .

SZEGEDY C, VANHOUCKE V, IOFFE S, et al . Rethinking the inception architecture for computer vision[C]// Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition ( CVPR ). Las Vegas, NV: IEEE, 2016: 2818-2826.

Z ZHONG , L ZHENG , G KANG , 等 . Random erasing data augmentation . ArXiv , 2017 . 1708 http://cn.bing.com/academic/profile?id=f4fb58e3a617595969caa86f61345503&encoded=0&v=paper_preview&mkt=zh-cn http://cn.bing.com/academic/profile?id=f4fb58e3a617595969caa86f61345503&encoded=0&v=paper_preview&mkt=zh-cn .

赵泉华 , 张洪云 , 李玉 . 采用非规则标识点过程的LiDAR点云数据目标提取 . 光学精密工程 , 2018 . 26 ( 5 ): 1201 - 1210 . http://d.old.wanfangdata.com.cn/Periodical/gxjmgc201805022 http://d.old.wanfangdata.com.cn/Periodical/gxjmgc201805022 .

Q H ZHAO , H Y ZHANG , Y LI . Target extraction from LiDAR point cloud data using irregular geometry marked point process . Optics and Precision Engineering , 2018 . 26 ( 5 ): 1201 - 1210 . http://d.old.wanfangdata.com.cn/Periodical/gxjmgc201805022 http://d.old.wanfangdata.com.cn/Periodical/gxjmgc201805022 .

金志刚 , 李静昆 . 基于对象性和多层线性模型的协同显著性检测 . 光学精密工程 , 2019 . 27 ( 8 ): 1845 - 1853 . http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=gxjmgc201908020 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=gxjmgc201908020 .

Z G JIN , J K LI . Co-saliency detection based on objectness and multi-layer linear model . Optics and Precision Engineering , 2019 . 27 ( 8 ): 1845 - 1853 . http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=gxjmgc201908020 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=gxjmgc201908020 .

雷金周 , 曾令斌 , 叶南 . 工业机器人单目视觉对准技术研究 . 光学精密工程 , 2018 . 26 ( 3 ): 733 - 741 . http://d.old.wanfangdata.com.cn/Periodical/gxjmgc201803027 http://d.old.wanfangdata.com.cn/Periodical/gxjmgc201803027 .

J Z LEI , L B ZENG , N YE . Research on industrial robot alignment technique with monocular vision . Optics and Precision Engineering , 2018 . 26 ( 3 ): 733 - 741 . http://d.old.wanfangdata.com.cn/Periodical/gxjmgc201803027 http://d.old.wanfangdata.com.cn/Periodical/gxjmgc201803027 .

L LIN , X L WANG , W YANG , 等 . Discriminatively trained and-or graph models for object shape detection . IEEE Transactions on Pattern Analysis and Machine Intelligence , 2015 . 37 ( 5 ): 959 - 972 . DOI: 10.1109/TPAMI.2014.2359888 http://doi.org/10.1109/TPAMI.2014.2359888 .

ZHENG L, SHEN L Y, TIAN L, et al . Scalable person re-identification: a benchmark[C]// Proceedings of 2015 IEEE International Conference on Computer Vision ( ICCV ). Santiago: IEEE, 2015: 1116-1124.

SUN Y F, ZHENG L, DENG W J, et al . SVDNet for pedestrian retrieval[C]// Proceedings of 2017 IEEE International Conference on Computer Vision (ICCV) . Venice: IEEE, 2017: 3820-3828.

Views

413

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Camouflaged object segmentation based on edge enhancement and feature fusion

Attention and cross-scale fusion for vehicle and pedestrian detection

Image rain removal algorithm based on multi-cascade progressive convolution structure

Progress of learning-based computer-generated holography

Edge-guided light field image saliency detection

Related Author

LI Mingyan

WU Chuan

ZHU Ming

LI Jian-dong

LI Jia-qi

QU Hai-cheng

ZHANG Yong

GUO Jie-long

Related Institution

Changchun Institute of Optics， Fine Mechanics and Physics， Chinese Academy of Sciences

University of Chinese Academy of Sciences

College of Software， Liaoning Technical University

College of Mining， Liaoning Technical University

College of Electrical Engineering and Automation， Fuzhou University

AI问答

Address：No.3888 Dong Nanhu Road, Changchun, Jilin, China 130033 Postal code：130033
Tel：0431-86176059 Email：yjxs@ciomp.ac.cn
Technical support is provided by Beijing Founder electronics co., LTD 吉ICP备11002662号-17 京公网安备11010802024621
It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.

⁰