[1]张智,毕晓君.基于风格转换的无监督聚类行人重识别[J].智能系统学报,2021,16(1):48-56.[doi:10.11992/tis.202012014]
 ZHANG Zhi,BI Xiaojun.Clustering approach based on style transfer for unsupervised person re-identification[J].CAAI Transactions on Intelligent Systems,2021,16(1):48-56.[doi:10.11992/tis.202012014]
点击复制

基于风格转换的无监督聚类行人重识别(/HTML)
分享到:

《智能系统学报》[ISSN:1673-4785/CN:23-1538/TP]

卷:
第16卷
期数:
2021年1期
页码:
48-56
栏目:
学术论文—机器感知与模式识别
出版日期:
2021-01-05

文章信息/Info

Title:
Clustering approach based on style transfer for unsupervised person re-identification
作者:
张智1 毕晓君2
1. 哈尔滨工程大学 信息与通信工程学院,黑龙江 哈尔滨 150001;
2. 中央民族大学 信息工程学院,北京 100081
Author(s):
ZHANG Zhi1 BI Xiaojun2
1. College of Information and Communication Engineering, Harbin Engineering University, Harbin 150001, China;
2. School of Information Engineering, Minzu University of China, Beijing 100081, China
关键词:
机器视觉行人重识别无监督聚类风格转换生成对抗网络残差块跨域
Keywords:
machine visionpedestrian re-identificationunsupervisedclusteringstyle transformationgenerative adversarial networksresidual blockcross domain
分类号:
TP391
DOI:
10.11992/tis.202012014
摘要:
无监督行人重识别中源域与目标域间的巨大差异性是影响模型性能的最关键因素。基于聚类的无监督行人重识别方法挖掘目标域数据间的相似性,以此缓解该问题,但仍未消除域间差异性。本文提出一种基于风格转换的无监督聚类行人重识别方法。首先,针对基于聚类方法的模型存在受域间差异性影响的问题,将一种基于生成对抗网络的风格转换方法引入到聚类方法模型中,将源域数据转换为目标域风格数据,直接减小域间差异性,提升模型的识别性能。其次,针对风格转换模型的生成器存在转换尺度单一以及特征信息传递效率低的问题,使用一种新型残差块替换原始残差块并将其引入到生成器上采样和下采样中,形成多特征尺度转换以及信息传递效率高的生成器,提升风格转换效果,降低域间差异性,进一步提升整体模型的识别效果。在Market1501以及Duke-MTMC-reID数据集上对所提的算法进行实验,结果表明改进方法取得了更好的识别效果。
Abstract:
The substantial difference between the source and target domains is the most crucial factor affecting the performance of unsupervised person re-identification models. The clustering-based unsupervised person re-identification method alleviates the problem to a certain extent by mining the similarity between the target domain, but it does not fundamentally eliminate the discrepancy between the domains. This paper proposes a clustering approach based on cross-domain style transfer for unsupervised pedestrian re-identification. First, to avoid the difference between domains in clustering-based unsupervised person re-identification models, the across-domain style transfer method based on a generative adversarial network is introduced into the clustering process. It transfers the source domain data to the target domain style data, which directly reduces the difference between domains and improves the recognition effect of the model. Second, the generator of cross-domain style transfer model has a single transfer scale and low efficiency of characteristics information transfer. A new type of residual block is proposed to replace the original residual block; then, it is inserted into the generator to achieve up-sampling and down-sampling. The specific generator has more characteristics of the scale transfer, and it transmits information more effectively. The cross-domain style transfer model can better transfer the style of the source and target domains, further reduce the difference between the two domains, and improve the recognition effect of the overall model. Extensive experiments were implemented on Market1501 and Duke-MTMC-Reid datasets to examine the proposed method, and the results showed that the proposed improved method achieved a better recognition effect.

参考文献/References:

[1] 刘帅师, 程曦, 郭文燕, 等. 深度学习方法研究新进展[J]. 智能系统学报, 2016, 11(5):567-577
LIU Shuaishi, CHENG Xi, GUO Wenyan, et al. Progress report on new research in deep learning[J]. CAAI transactions on intelligent systems, 2016, 11(5):567-577
[2] 宋婉茹, 赵晴晴, 陈昌红, 等. 行人重识别研究综述[J]. 智能系统学报, 2017, 12(6):770-780
SONG Wanru, ZHAO Qingqing, CHEN Changhong, et al. Survey on pedestrian re-identification research[J]. CAAI transactions on intelligent systems, 2017, 12(6):770-780
[3] GENG Mengyue, WANG Yaowei, XIANG Tao, et al. Deep transfer learning for person re-identification[J]. arXiv preprint arXiv:1611.05244, 2016:00007480.
[4] LI Wei, ZHAO Rui, XIAO Tong, et al. Deepreid:deep filter pairing neural network for person re-identification[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Columbus, OH, USA, 2014:152-159.
[5] ZHONG Zhun, ZHENG Liang, LI Shaozi, et al. Generalizing a person retrieval model hetero- and homogeneously[C]//Proceedings of the 15th European Conference on Computer Vision (ECCV). Munich, Germany, 2018:176-192.
[6] PENG Peixi, XIANG Tao, WANG Yaowei, et al. Unsupervised cross-dataset transfer learning for person re-identification[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, NV, USA, 2016:1306-1315.
[7] DENG Weijian, ZHENG Liang, YE Qixiang, et al. Image-image domain adaptation with preserved self-similarity and domain-dissimilarity for person re-identification[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City, UT, USA, 2018:994-1003.
[8] FU Yang, WEI Yunchao, WANG Guanshuo, et al. Self-similarity grouping:a simple unsupervised cross domain adaptation approach for person re-identification[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision. Seoul, Korea (South), 2019:6111-6120.
[9] ZHU Junyan, PARK T, ISOLA P, et al. Unpaired image-to-image translation using cycle-consistent adversarial networks[C]//Proceedings of the IEEE International Conference on Computer Vision. Venice, Italy, 2017:2242-2251.
[10] HE Kaiming, ZHANG Xiangyu, REN Shaoqing, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, NV, USA, 2016:770-778.
[11] ESTER M, KRIEGEL H P, SANDER J, et al. A density-based algorithm for discovering clusters in large spatial databases with noise[C]//Proceedings of the Second International Conference on Knowledge Discovery and Data Mining. Portland, Oregon, USA, 1996:226-231.
[12] HERMANS A, BEYER L, LEIBE B. In defense of the triplet loss for person re-identification[J]. arXiv preprint arXiv:1703.07737, 2017.
[13] HE Kaiming, ZHANG Xiangyu, REN Shaoqing, et al. Identity mappings in deep residual networks[C]//14th European Conference on Computer Vision. Amsterdam, The Netherlands, 2016:630-645.
[14] HADSELL R, CHOPRA S, LECUN Y. Dimensionality reduction by learning an invariant mapping[C]//2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. New York, NY, USA, 2006:1735-1742.
[15] ZHENG Liang, SHEN Liyue, TIAN Lu, et al. Scalable person re-identification:a benchmark[C]//Proceedings of the IEEE International Conference on Computer Vision. Santiago, Chile, 2015:1116-1124.
[16] RISTANI E, SOLERA F, ZOU R, et al. Performance measures and a data set for multi-target, multi-camera tracking[C]//European Conference on Computer Vision. Amsterdam, The Netherlands, 2016:17-35.
[17] CHEN Yanbei, ZHU Xiatian, GONG Shaogang. Instance-guided context rendering for cross-domain person re-identification[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision. Seoul, Korea (South), 2019:232-242.
[18] ZHONG Zhun, ZHENG Liang, LUO Zhiming, et al. Invariance matters:exemplar memory for domain adaptive person re-identification[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Long Beach, CA, USA, 2019:598-607.
[19] YU Hongxing, ZHENG Weishi, WU Ancong, et al. Unsupervised person re-identification by soft multilabel learning[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. Long Beach, CA, USA, 2019:2143-2152.

相似文献/References:

[1]田国会,吉艳青,李晓磊.家庭智能空间下基于场景的人的行为理解[J].智能系统学报,2010,5(01):57.
 TIAN Guo-hui,JI Yan-qing,LI Xiao-lei.Human behaviors understanding based on scene knowledge in home intelligent space[J].CAAI Transactions on Intelligent Systems,2010,5(1):57.
[2]梁义辉,战强.一种面向无线图像传输的视觉平台[J].智能系统学报,2016,11(5):608.[doi:10.11992/tis.201512014]
 LIANG Yihui,ZHAN Qiang.A visual platform for wireless image transmission[J].CAAI Transactions on Intelligent Systems,2016,11(1):608.[doi:10.11992/tis.201512014]
[3]李霞丽,吴立成,樊艳明.易于硬件实现的压缩感知观测矩阵的研究与构造[J].智能系统学报,2017,12(03):279.[doi:10.11992/tis.201606037]
 LI Xiali,WU Licheng,FAN Yanming.Study and construction of a compressed sensing measurement matrix that is easy to implement in hardware[J].CAAI Transactions on Intelligent Systems,2017,12(1):279.[doi:10.11992/tis.201606037]
[4]宋婉茹,赵晴晴,陈昌红,等.行人重识别研究综述[J].智能系统学报,2017,12(06):770.[doi:10.11992/tis.201706084]
 SONG Wanru,ZHAO Qingqing,CHEN Changhong,et al.Survey on pedestrian re-identification research[J].CAAI Transactions on Intelligent Systems,2017,12(1):770.[doi:10.11992/tis.201706084]
[5]郭晓峰,王耀南,周显恩,等.中国象棋机器人棋子定位与识别方法[J].智能系统学报,2018,13(04):517.[doi:10.11992/tis.201709020]
 GUO Xiaofeng,WANG Yaonan,ZHOU Xianen,et al.Chess-piece localization and recognition method for Chinese chess robot[J].CAAI Transactions on Intelligent Systems,2018,13(1):517.[doi:10.11992/tis.201709020]
[6]安果维,王耀南,周显恩,等.基于显著性检测的双目测距系统[J].智能系统学报,2018,13(06):913.[doi:10.11992/tis.201712005]
 AN Guowei,WANG Yaonan,ZHOU Xianen,et al.Binocular distance measurement system based on saliency detection[J].CAAI Transactions on Intelligent Systems,2018,13(1):913.[doi:10.11992/tis.201712005]
[7]赵立明,龙大周,徐晓东,等.工业机器人加工轨迹双目3D激光扫描成像修正方法[J].智能系统学报,2021,16(4):690.[doi:10.11992/tis.202008008]
 ZHAO Liming,LONG Dazhou,XU Xiaodong,et al.Binocular 3D laser scanning imaging-based industrial robot machining trajectory correction method[J].CAAI Transactions on Intelligent Systems,2021,16(1):690.[doi:10.11992/tis.202008008]
[8]朱齐丹,李小铜,郑天昊.舰载机位姿实时视觉测量算法研究[J].智能系统学报,2021,16(6):1045.[doi:10.11992/tis.202103014]
 ZHU Qidan,LI Xiaotong,ZHENG Tianhao.Research on real-time vision measurement algorithm of shipborne aircraft pose[J].CAAI Transactions on Intelligent Systems,2021,16(1):1045.[doi:10.11992/tis.202103014]
[9]谢家阳,王行健,史治国,等.动态云台摄像机无人机检测与跟踪算法[J].智能系统学报,2021,16(5):858.[doi:10.11992/tis.202103032]
 XIE Jiayang,WANG Xingjian,SHI Zhiguo,et al.Drone detection and tracking in dynamic pan-tilt-zoom cameras[J].CAAI Transactions on Intelligent Systems,2021,16(1):858.[doi:10.11992/tis.202103032]

备注/Memo

备注/Memo:
收稿日期:2020-12-07。
作者简介:张智,硕士研究生,主要研究方向为深度学习、图像处理;毕晓君,教授,博士生导师,主要研究方向为信息智能处理、数字图像处理、智能优化算法及机器学习。主持国家自然科学基金面上项目2项、科技部国际合作项目面上项目1 项、教育部博士点基金项目1 项、工业和信息化部海洋工程装备科研项目子项目1 项、民品横向课题1 项,获国家专利8项。发表学术论文170 余篇,出版学术专著3部
通讯作者:毕晓君. E-mail:bixiaojun@hrbeu.edu.cn
更新日期/Last Update: 2021-02-25