二次聚类的无监督行人重识别方法

doi:10.3778/j.issn.1002-8331.2207-0469

摘要/Abstract

摘要： 针对当前无监督行人重识别方法因受到硬件差异、光照变化等客观因素的影响，导致同一行人图像出现较大反差，随之易带来样本错误伪标签生成的问题，使得现有无监督行人重识别方法还有待进一步提升的空间。为了解决此问题，提出了一种基于二次重聚类的无监督行人重识别（unsupervised person re-identification based on quadratic clustering）方法。该方法主要包括全局二次聚类的无监督学习模块和基于聚类结果的有监督学习模块。具体来说，前者基于全局二次聚类分别对相机ID和行人身份ID进行无监督分析，解决了同一行人在不同摄像机视角下的统一成像风格问题；后者则采用有监督学习方式改进了内存字典的初始化与更新方式，解决了模型在训练中偏移的问题。通过此双模块的协同训练以共同抑制跨摄像头间采集的图像所产生错误伪标签的问题。所提出的算法分别在Market-1501、DukeMTMC-ReID、MSMT17、Person和VeRi-776数据集上进行实验，取得了mAP=81.2%和rank-1=91.2%、mAP=68.4%和rank-1=78.7%、mAP=31.1%和rank-1=60.4%、mAP=88.3%和rank-1=93.6%的性能，对比当前最先进的方法，分别提高了2.4、1.8、6.0、2.5和4.3个百分点的rank-1准确率。

关键词: 行人重识别, 无监督学习, 二次聚类, 协同训练

Abstract: In view of the influence of objective factors such as hardware differences and illumination changes, the current unsupervised person re-identification method leads to a large contrast in the image of the same person, which is easy to cause the problem of wrong pseudo-labels generation of samples, which makes the existing unsupervised person re-identification method. There is still room for further improvement in the identification method. To solve this problem, this paper proposes an unsupervised person re-identification based on quadratic clustering method. This method mainly includes global quadratic clustering module and supervised learning module based on quadratic clustering results. Specifically, the former performs unsupervised analysis of camera ID and pedestrian ID based on global quadratic clustering, which solves the problem of unified imaging style of the same pedestrian under different camera perspectives; the latter uses supervised learning to improve memory. The initialization and update method of the dictionary solves the problem of model offset during training. Through the co-training of this dual module, it can jointly suppress the problem of false labels generated by images collected across cameras. The algorithm proposed in this paper is tested on Market-1501, DukeMTMC-ReID, MSMT17, Person and VeRi-776 datasets, respectively, and achieves mAP=81.2% and rank-1=91.2%, mAP=68.4% and rank-1=78.7%, mAP=31.1% and rank-1=60.4%, mAP=88.3% and rank-1=93.6%, compared with the current state-of-the-art methods, they have improved by 2.4, 1.8, 6.0, 2.5 and 4.3 percentage points rank-1 accuracy.

Key words: person re-identification, unsupervised learning, quadratic clustering, collaborative training

熊明福, 肖应雄, 陈佳, 胡新荣, 彭涛. 二次聚类的无监督行人重识别方法[J]. 计算机工程与应用, 2024, 60(1): 227-235.

XIONG Mingfu, XIAO Yingxiong, CHEN Jia, HU Xinrong, PENG Tao. Unsupervised Person Re-Identification Based on Quadratic Clustering[J]. Computer Engineering and Applications, 2024, 60(1): 227-235.

参考文献

[1] 杨永胜, 邓淼磊, 李磊, 等. 基于深度学习的行人重识别综述[J]. 计算机工程与应用, 2022, 58(9): 51-66.
YANG Y S, DENG M L, LI L, et al. Overview of pedestrian re-identification based on deep learning[J]. Computer Engineering and Applications, 2022, 58(9): 51-66.
[2] 叶钰, 王正, 梁超, 等. 多源数据行人重识别研究综述[J]. 自动化学报, 2020, 46(9): 1869-1884.
YE Y, WANG Z, LIANG C, et al. A survey on multi-source person re-identification[J]. Acta Automatica Sinica, 2020, 46(9): 1869-1884.
[3] 简彩仁, 翁谦, 夏靖波. 系数增强最小二乘回归子空间聚类法[J]. 计算机工程与应用, 2022, 58(20): 73-78.
JIAN C R, WENG Q, XIA J B. Coefficient enhanced least square regression subspace clustering method[J]. Computer Engineering and Applications, 2022, 58(20): 73-78.
[4] 柳恩涵, 张锐, 赵硕, 等. 一种基于视频预测的红外行人目标跟踪方法[J]. 哈尔滨工业大学学报, 2020, 52(10): 192-200.
LIU E H, ZHANG R, ZHAO S, et al. Infrared pedestrian target tracking method based on video prediction[J]. Journal of Harbin Institute of Technology, 2020, 52(10): 192-200.
[5] LIN X, REN P, YEH C H, et al. Unsupervised person re-identification: a systematic survey of challenges and solutions[J]. arXiv:2109.06057, 2021.
[6] LIN Y, DONG X, ZHENG L, et al. A bottom-up clustering approach to unsupervised person re-identification[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2019: 8738-8745.
[7] WANG D, ZHANG S. Unsupervised person re-identification via multi-label classification[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020: 10981-10990.
[8] DYER C. Notes on noise contrastive estimation and negative sampling[J]. arXiv:1410.8251, 2014.
[9] YANG F, ZHONG Z, LUO Z, et al. Joint noise-tolerant learning and meta camera shift adaptation for unsupervised person re-identification[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021: 4855-4864.
[10] DENG W, ZHENG L, YE Q, et al. Image-image domain adaptation with preserved self-similarity and domain-dissimilarity for person re-identification[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018.
[11] GE Y X, CHEN D P, LI H S. Mutual mean-teaching: pseudo label refinery for unsupervised domain adaptation on person re-identification[J]. arXiv:2001.01526, 2020.
[12] PATRA B, MONIZ J R A, GARG S, et al. Bilingual lexicon induction with semi-supervision in non-isometric embedding spaces[J]. arXiv:1908.06625, 2019.
[13] 陈璠, 彭力. 多层级重叠条纹特征融合的行人重识别[J]. 计算机科学与探索, 2021, 15(9): 1753-1761.
CHEN F, PENG L. Person re-identification based on multi-level feature fusion with overlapping stripes[J]. Journal of Frontiers of Computer Science & Technology, 2021, 15(9): 1753-1761.
[14] YU J, OH H. Unsupervised person re-identification via multi-label prediction and classification based on graph-structural insight[J]. arXiv:2106.08798, 2021.
[15] SUN Y, ZHENG L, YANG Y, et al. Beyond part models: person retrieval with refined part pooling (and a strong convolutional baseline)[C]//European Conference on Computer Vision. Cham: Springer, 2017.
[16] LUO H, JIANG W, GU Y, et al. A strong baseline and batch normalization neck for deep person re-identification[J]. IEEE Transactions on Multimedia, 2020, 22(10): 2597-2609.
[17] CHEN H, WANG Y, LAGADEC B, et al. Joint generative and contrastive learning for unsupervised person re-identification[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2021: 2004-2013.
[18] HAN B, YAO Q, YU X, et al. Co-teaching: robust training of deep neural networks with extremely noisy labels[C]//Advances in Neural Information Processing Systems, 2018.
[19] ESTER M. 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, 1996.
[20] MACQUEEN J. Some Methods for? classification and analysis of multivariate observations[C]//Proceedings of the 5th Berkeley Symposium on Mathematical Statistics and Probability, 1967: 281-297.
[21] SCHROFF F, KALENICHENKO D, PHILBIN J. Facenet: a unified embedding for face recognition and clustering[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2015: 815-823.
[22] HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 2016: 770-778.
[23] ZHENG L, SHEN L, TIAN L, et al. Scalable person re-identification: a benchmark[C]//Proceedings of the IEEE International Conference on Computer Vision, 2015: 1116-1124.
[24] 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. Cham: Springer, 2016: 17-35.
[25] ZHONG Z, ZHENG L, KANG G, et al. Random erasing data augmentation[C]//Proceedings of the AAAI Conference on Artificial Intelligence, 2020: 13001-13008.
[26] SOHN K. Improved deep metric learning with multi-class N-pair loss objective[C]//Neural Information Processing Systems, 2016.
[27] TONG X, TIAN X, YI Y, et al. Learning from massive noisy labeled data for image classification[C]//2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2015.
[28] LI M X, ZHU X T, GONG S G. Unsupervised person re-identification by deep learning tracklet association[J]. ACM Transactions on Multimedia Computing Communications & Applications, 2018, 14(4): 1-18.
[29] DAI Z, WANG G, ZHU S, et al. Cluster contrast for unsupervised person re-identification[J]. arXiv:2103.11568, 2021.
[30] HERMANS A, BEYER L, LEIBE B. In defense of the triplet loss for person re-identification[J]. arXiv:1703.07737, 2017.
[31] OORD A, LI Y, VINYALS O. Representation learning with contrastive predictive coding[J]. arXiv:1807.03748, 2018.
[32] WEI L, ZHANG S, WEN G, et al. Person transfer GAN to bridge domain gap for person re-identification[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018.
[33] SUN X, ZHENG L. Dissecting person re-identification from the viewpoint of viewpoint[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2019: 608-617.
[34] LIU X, WU L, MA H, et al. Large-scale vehicle re-identification in urban surveillance videos[C]//IEEE International Conference on Multimedia and Expo (ICME), 2016.
[35] DENG J, DONG W, SOCHER R, et al. ImageNet: a large-scale hierarchical image database[C]//2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR 2009), 20-25 June 2009, Miami, Florida, USA, 2009.
[36] 向朝参, 李耀宇, 冯亮, 等. 基于深度强化学习的智联网汽车感知任务分配[J]. 计算机学报, 2022, 45(5): 918-934.
XIANG C S, LI Y Y, FENG L, et al. Near-optimal vehicular crowdsensing task allocation empowered by deep reinforcement learning[J]. Chinese Journal of Computers, 2022, 45(5): 918-934.
[37] YU H X, ZHENG W S, WU A, et al. Unsupervised person re-identification by soft multilabel learning[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019.
[38] ZHAO F, LIAO S, XIE G S, et al. Unsupervised domain adaptation with noise resistible mutual-training for person re-identification[C]//European Conference on Computer Vision. Cham: Springer, 2020.
[39] ZHAI Y P, YE Q X, LU S J, et al. Multiple expert brainstorming for domain adaptive person re-identification[C]//European Conference on Computer Vision. Cham: Springer, 2020.
[40] ZENG K, NING M, WANG Y, et al. Hierarchical clustering with hard-batch triplet loss for person re-identification[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2020: 13657-13665.
[41] LI J, ZHANG S. Joint visual and temporal consistency for unsupervised domain adaptive person re-identification[C]//European Conference on Computer Vision. Cham: Springer, 2020: 483-499.
[42] QIN H, XIE W, LI Y, et al. PTGAN: a proposal-weighted two-stage GAN with attention for hyperspectral target detection[C]//2021 IEEE International Geoscience and Remote Sensing Symposium, 2021: 4428-4431.
[43] ZHONG Z, ZHENG L, LUO Z, et al. Invariance matters: exemplar memory for domain adaptive person re-identification[C]//2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019.
[44] FU Y, WEI Y, WANG G, 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, 2019: 6112-6121.
[45] GE Y, ZHU F, CHEN D, et al. Self-paced contrastive learning with hybrid memory for domain adaptive object re-id[C]//Advances in Neural Information Processing Systems, 2020: 11309-11321.