research-article

Unsupervised domain adaptive person re-identification via camera penalty learning

Authors:

Jinlei ZhuAuthors Info & Claims

Multimedia Tools and Applications, Volume 80, Issue 10

Pages 15215 - 15232

https://doi.org/10.1007/s11042-021-10589-6

Published: 01 April 2021 Publication History

Abstract

Unsupervised domain adaptive (UDA) person re-identification (re-ID) aims to adapt the model trained on a labeled source domain to an unlabeled target domain. For pseudo-label-based UDA methods, pseudo label noise is the main problem for model degradation and cross-camera problem is one main factor to cause this noise. In this paper, a novel camera penalty learning (CPL) UDA person re-ID method is proposed to address this problem. The possibility of selecting wrong negative sample and positive sample is relatively high in conventional triplet loss due to cross-camera problem. To alleviate this problem, a camera-penalty-based triplet loss (PTL) is designed. It adds camera-ID-penalty to conventional triplet loss to reduce sample distance imbalance, thereby improving the quality of pseudo labels. In order to reduce the dependence on pseudo labels and improve the robustness, a camera-penalty-neighborhood loss (PNL) is designed and combined with the push loss (PL). The PNL minimizes the distance between one image and its camera-penalty-weighted neighbors. The PL maximizes the distances among all images. The proposed CPL model achieves considerable results of 87.4%/70.7% and 75.2%/59.0% Rank-1/ mAP on DukeMTMC-reID-to-Market-1501 and Market-1501-to-DukeMTMC-reID UDA tasks.

References

[1]

Bolle R M, Connell J H, Pankanti S, et al(2005) The relation between the ROC curve and the CMC.Fourth IEEE workshop on automatic identification advanced technologies (AutoID'05) 15-20

[2]

Chen Y, Zhu X, Gong S (2019) Instance-guided context rendering for cross-domain person re-identification. In: 2019 proceedings of the IEEE international conference on computer vision (ECCV) 232-242

[3]

Cheng D, Gong Y, Shi W, and Zhang S Person re-identification by the asymmetric triplet and identification loss function Multimed Tools Appl 2017 77 3 3533-3550

[4]

Fan H, Zheng L, Yan C, and Yang Y Unsupervised person re-identification: clustering and fine-tuning ACM Trans Multimed Comput Commun Appl 2018 14 4 1-18

[5]

Fu Y, Wei Y, Wang G, Zhou Y, Shi H, Huang T-S (2019) Self-similarity grouping: a simple unsupervised cross domain adaptation approach for person re-identification. In: 2019 proceedings of the IEEE international conference on computer vision (ECCV) 6112–6121

[6]

Ge Y-X, Zhu F, Zhao R, Li H-S (2020) Structured domain adaptation for unsupervised person re-identification. arXiv:200306650

[7]

Ge Y-X, Chen D-P, Li H-S (2020) Mutual mean-teaching: Pseudo label refinery for unsupervised domain adaptation on person re-identification. In: 2020 international conference on learning representations (ICLR).

[8]

Genç A and Ekenel HK Cross-dataset person re-identification using deep convolutional neural networks: effects of context and domain adaptation Multimed Tools Appl 2018 78 5 5843-5861

[9]

Hermans A, Beyer L, Leibe B (2017) In defense of the triplet loss for person re-identification. arXiv:170307737

[10]

Li W, Zhao R, Xiao T, Wang X (2014) DeepReID: deep filter pairing neural network for person re-identification. In: 2014 IEEE conference on computer vision and pattern recognition (CVPR) 152-159

[11]

Li K, Ding Z, Li K, Zhang Y, Fu Y (2018) Support neighbor loss for person re-identification. ACM Trans Multimed Comput Commun Appl 1492–1500

[12]

Li M, Zhu X, and Gong S Unsupervised Tracklet person re-identification IEEE Trans Pattern Anal Mach Intell 2020 42 7 1770-1782

[13]

Li Y-J, Lin C-S, Lin Y-B, Wang F Y-C (2019) Cross-dataset person re-identification via unsupervised pose disentanglement and adaptation. In: 2019 IEEE International Conference on Computer Vision (ICCV) 7918-7928

[14]

Lin Y-T., Dong X-Y, Zheng L, Yan Y, Yang Y (2019) A bottom-up clustering approach to unsupervised person re-identification. In: 2019 proceedings of the AAAI conference on artificial intelligence 33:8738-8745

[15]

Lin Y, Wu Y, Yan C, Xu M, and Yang Y Unsupervised person re-identification via cross-camera similarity exploration IEEE Trans Image Process 2020 29 5481-5490

[16]

Liu J, Li W, Pei H, Wang Y, Qu F, Qu Y, and Chen Y Identity preserving generative adversarial network for cross-domain person re-identification IEEE Access 2019 7 114021-114032

[17]

Pan X-G, Luo P, Shi J-P, Tang X-O (2018) Two at once: enhancing learning and generalization capacities via ibn-net. In: 2018 European conference on computer vision (ECCV) 484-500

[18]

Ren C-X, Liang B, Ge P, Zhai Y, and Lei Z Domain adaptive person re-identification via camera style generation and label propagation IEEE Trans Inf Forensics Secur 2020 15 1290-1302

[19]

Shen C, Qi G-J, Jiang R, Jin Z, Yong H, Chen Y, and Hua X-S Sharp attention network via adaptive sampling for person re-identification IEEE Trans Circuits Syst Vid Technol 2019 29 10 3016-3027

[20]

Sikdar A and Chowdhury AS Scale-invariant batch-adaptive residual learning for person re-identification Pattern Recogn Lett 2020 129 279-286

[21]

Song L, Wang C, Zhang L, Du B, Zhang Q, Huang C, Wang X (2020) Unsupervised domain adaptive re-identification: Theory and practice Pattern Recognition 102

[22]

Wang J-Y, Zhu X-T, Gong S-G, Li W (2018) Transferable joint attribute-identity deep learning for unsupervised person re-identification. In: 2018 proceedings of the IEEE conference on computer vision and pattern recognition (CVPR) 2275-2284

[23]

Wu A, Zheng W-S, Lai J-H (2019) Unsupervised person reidentification by camera-aware similarity consistency learning. In: 2019 proceedings of the IEEE international conference on computer vision (ECCV) 6922–6931

[24]

Xian Y and Hu H Enhanced multi-dataset transfer learning method for unsupervised person re-identification using co-training strategy IET Comput Vis 2018 12 8 1219-1227

[25]

Xie G-S, Zhang X-Y, Yan S, and Liu C-L Hybrid CNN and dictionary-based models for scene recognition and domain adaptation IEEE Trans Circuits Syst Vid Technol 2017 27 6 1263-1274

[26]

Xie G-S, Zhang Z, Liu L, Zhu F, Zhang X-Y, Shao L, and Li X SRSC: selective, robust, and supervised constrained feature representation for image classification IEEE Trans Neural Netw Learn Syst 2019 10 1-13

[27]

Yang F, Yan K, Lu S, Jia H, Xie D, Yu Z, Guo X, Huang F, Gao W (2020) Part-aware progressive unsupervised domain adaptation for person re-identification. IEEE Trans Multimed, 1

[28]

Yu H-Y, Zheng W-S, Wu A-C, et al (2019) Unsupervised person re-identification by soft multilabel learning. In: 2019 IEEE conference on computer vision and pattern recognition (CVPR) 2148-2157

[29]

Yu H-X, Wu A, and Zheng W-S Unsupervised person re-identification by deep asymmetric metric embedding IEEE Trans Pattern Anal Mach Intell 2020 42 4 956-973

[30]

Zhang X-Y, Cao J-W, Shen C-H, You M-Y (2019) Self-training with progressive augmentation for unsupervised cross-domain person re-identification. In: 2019 IEEE international conference on computer vision (ICCV) 8222-8231

[31]

Zhao F, Liao S-C, Xie G-S, Zhao J, Shao L (2020) Unsupervised domain adaptation with noise resistible mutual-training for person re-identification. In: 2020 European conference on computer vision (ECCV)

[32]

Zheng L, Shen L, Tian L, Wang S, Wang J, Tian Q (2015) Scalable person re-identification: a benchmark. In: 2015 IEEE international conference on computer vision (ICCV) 1116-1124

[33]

Zheng Z, Zheng L, Yang Y (2017) Unlabeled samples generated by GAN improve the person re-identification baseline in vitro. In: 2017 IEEE international conference on computer vision (ICCV) 3774-3782

[34]

Zheng Z, Zheng L, and Yang Y Pedestrian alignment network for large-scale person re-identification IEEE Trans Circuits Syst Vid Technol 2019 29 10 3037-3045

[35]

Zhong Z, Zheng L, Li S-Z, Yang Y (2018) Generalizing a person retrieval model hetero-and homogeneously. In: 2018 proceedings of the European conference on computer vision (ECCV) 172-188

[36]

Zhong Z, Zheng L, Luo Z-M, Li S-Z, Yang Y. Invariance Matters: Exemplar Memory for Domain Adaptive Person Re-identification. In: 2019 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) 598–607

[37]

Zhu J-Y, Park T, Isola P, et al (2017) Unpaired image-to-image translation using cycle-consistent adversarial networks. In: 2017 IEEE international conference on computer vision (ICCV) 2242-2251

Cited By

Wu JChang KLin I(2023)Generalizable person re-identification with part-based multi-scale networkMultimedia Tools and Applications10.1007/s11042-023-14718-182:25(38639-38666)Online publication date: 1-Oct-2023
https://dl.acm.org/doi/10.1007/s11042-023-14718-1
Li YZhu XSun JChen HLi Z(2023)Unsupervised person re-identification based on high-quality pseudo labelsApplied Intelligence10.1007/s10489-022-04270-053:12(15112-15126)Online publication date: 1-Jun-2023
https://dl.acm.org/doi/10.1007/s10489-022-04270-0

Index Terms

Unsupervised domain adaptive person re-identification via camera penalty learning
1. Computing methodologies
  1. Artificial intelligence
    1. Computer vision
      1. Computer vision problems
  2. Machine learning
    1. Learning paradigms
      1. Supervised learning

Index terms have been assigned to the content through auto-classification.

Recommendations

Unsupervised Person Re-Identification via Multi-Label Classification
Abstract
The challenge of unsupervised person re-identification (ReID) lies in learning discriminative features without true labels. Most of previous works predict single-class pseudo labels through clustering. To improve the quality of generated pseudo ...
Angular regularization for unsupervised domain adaption on person re-identification
Abstract
State-of-the-art Re-ID methods based on unsupervised domain adaption transferred the domain knowledge by pre-training model on labeled source domain and fine-tuned the pre-trained model with pseudo labels generated by clustering samples on ...
Joint Visual and Temporal Consistency for Unsupervised Domain Adaptive Person Re-identification
Computer Vision – ECCV 2020
Abstract
Unsupervised domain adaptive person Re-IDentification (ReID) is challenging because of the large domain gap between source and target domains, as well as the lackage of labeled data on the target domain. This paper tackles this challenge through ...

Comments

Information & Contributors

Information

Published In

cover image Multimedia Tools and Applications

Multimedia Tools and Applications Volume 80, Issue 10

Apr 2021

1575 pages

ISSN:1380-7501

Issue’s Table of Contents

© The Author(s), under exclusive licence to Springer Science+Business Media, LLC part of Springer Nature 2021.

Publisher

Kluwer Academic Publishers

United States

Publication History

Published: 01 April 2021

Accepted: 20 January 2021

Revision received: 30 September 2020

Received: 01 July 2020

Author Tags

Qualifiers

Research-article

Funding Sources

National Natural Science Foundation of China (CN)
Major Science and Technology Innovation Project of Shandong Province

Contributors

Other Metrics

View Article Metrics

Bibliometrics & Citations

Bibliometrics

Article Metrics

2
Total Citations
View Citations
0
Total Downloads

Downloads (Last 12 months)0
Downloads (Last 6 weeks)0

Reflects downloads up to 22 Jan 2025

Other Metrics

View Author Metrics

Citations

Cited By

Wu JChang KLin I(2023)Generalizable person re-identification with part-based multi-scale networkMultimedia Tools and Applications10.1007/s11042-023-14718-182:25(38639-38666)Online publication date: 1-Oct-2023
https://dl.acm.org/doi/10.1007/s11042-023-14718-1
Li YZhu XSun JChen HLi Z(2023)Unsupervised person re-identification based on high-quality pseudo labelsApplied Intelligence10.1007/s10489-022-04270-053:12(15112-15126)Online publication date: 1-Jun-2023
https://dl.acm.org/doi/10.1007/s10489-022-04270-0

View Options

View options

Media

Figures

Other

Tables

View Issue’s Table of Contents