Skip to main content
SpringerLink
Log in

Study of cross-domain person re-identification based on DCGAN

  • 1213: Computational Optimization and Applications for Heterogeneous Multimedia Data
  • Published:
Multimedia Tools and Applications Aims and scope Submit manuscript

Abstract

Person re-identification(re-ID) techniques have been rapidly improving with the development of deep neural networks, and the accuracy of fully supervised re-ID models is already very high. However, when person re-identification models with supervised learning are used directly in unlabeled target scenes, the accuracy is greatly reduced. This is due to the large disparity between the domains of different datasets, such as resolution, lighting changes and occlusion. At the same time, the existing re-ID methods based on domain transfer have the problem of blurred images due to big gap between domains. Therefore, in this paper, a DCGAN (Deep Convolutional Generative Adversarial Network) is added to the unsupervised cross-domain re-ID model, which can allow the target domain distribution to be adequately fitted to the labeled domain distribution, so that the re-ID model trained in source domain can be used on the target domain without large fluctuations due to domain shift. By comparing with the UMDL(Unsupervised Multi-task Dictionary Learning) method, rank-1 and rank-5 are improved by 9.61% and 17.81%, respectively, when the model trained from Market-1501 is used in the GRID dataset, and by 9.52% and 14.14%, respectively, when compared with the GAN(Generative Adversarial Networks)-based method under the same experiment.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Algorithm 1
Algorithm 2
Fig. 3
Fig. 4

Similar content being viewed by others

References

  1. Bousmalis K, Trigeorgis G, Silberman N et al (2016) Domain separation networks. Advances in neural information processing systems, pp 343–351

  2. Chen G, Lu Y, Lu J et al (2020) Deep credible metric learning for unsupervised domain adaptation person re-identification. In: ECCV, Glasgow, UK, pp 643–659

  3. Chen Y, Zhu X, Gong S (2019) Instance-guided context rendering for cross-domain person re-identification. In: Proceeding ICCV, Seoul, Korea, pp 232–242

  4. Dai Y, Liu J, Sun Y et al (2021) IDM:, An Intermediate Domain Module for Domain Adaptive Person Re-ID, arXiv:2108.02413

  5. Fan L, Li T, Fang R et al (2020) Learning Longterm Representations for Person ReIdentification Using Radio Signals. In: Proceeding CVPR, Seattle, SEA, USA, pp 10699–10709

  6. Fang W, Zhang FH, Sheng VS, Ding YW (2018) Method for improving CNN-based image recognition using DCGAN. CMC: Comput Mater Contin 57:167–178

    Google Scholar 

  7. Fu K, Peng J, Zhang H, Wang X, Jiang F (2020) Image super-resolution based on generative adversarial networks: a brief review. Comput Mater Contin 64(3):1977–1997

    Google Scholar 

  8. He L, Liang J, Li H et al (2018) Deep spatial feature reconstruction for partial person reidentification: Alignment-free approach. In: Proceeding CVPR, Salt Lake City, SLC USA, pp 7073–7082

  9. He L, Wang Y, Liu W et al (2019) Foreground-aware pyramid reconstruction for alignment-free occluded person re-identification. In: Proceeding CVPR, State of California, CA, USA, pp 8450–8459

  10. He K, Zhang X, Ren S et al (2016) Deep residual learning for image recognition. In: Proceeding CVPR, Las Vegas, LV USA, pp 770–778

  11. Hermans A, Beyer L, Leibe B (2017) In defense of the triplet loss for person re-identification, arXiv:1703.07737

  12. Huang Y, Wu Q, Xu JS et al (2019) SBSGAN: Suppression Of inter-domain background shift for person re-identification. In: Proceeding CVPR, State of California, CA, USA, pp 9527–9536

  13. Isola P, Zhu JY, Zhou T, et al. (2017) Image-to-image translation with conditional adversarial networks. In: Proceeding CVPR, Hawaii, HI, USA, pp 1125–1134

  14. Loy CC, Xiang T, Gong S (2009) Multi-camera activity correlation analysis. In: Proceeding CVPR, Miami, Florida, USA, pp 1988–1995

  15. Mekhazni D, Bhuiyan A, Ekladious G, et al. (2020) Unsupervised Domain Adaptation in the Dissimilarity Space for Person Re-identification. In: ECCV, Glasgow, UK, pp 159–174

  16. Miao J, Wu Y, Liu P et al (2019) Pose-guided feature alignment for occluded person reidentification. In: Proceeding CVPR, State of California, CA, USA, pp 542–551

  17. Peng P, Xiang T, Wang Y et al (2016) Unsupervised cross-dataset transfer learning for person re-identification. In: Proceeding CVPR, Las Vegas, LV, USA, pp 1306–1315

  18. Shu R, Bui HH, Narui H et al (2018) A dirt-t approach to unsupervised domain adaptation, arXiv:1802.08735

  19. Song C, Huang Y, Ouyang W et al (2018) Mask-guided contrastive attention model for person re-identification. In: Proceeding CVPR, Salt Lake City, SLC, USA, pp 1179–1188

  20. Su K, Li J, Fu H (2011) Smart city and the applications. In: ICECC Ningbo, China, pp 1028–1031

  21. Su C, Li J, Zhang S et al (2017) Pose-driven deep convolutional model for person reidentification. In: Proceeding CVPR, Hawaii, HI, USA, pp 3960–3969

  22. Su C, Zhang S, Xing J et al (2016) Deep attributes driven multi-camera person reidentification, European conference on computer vision, vol 2016. Springer, Cham, pp 475–491

    Google Scholar 

  23. Sun Y, Xu Q, Li Y et al (2019) Perceive where to focus: Learning visibility-aware part-level features for partial person re-identification. In: Proceeding CVPR, State of California, CA, USA, pp 393–402

  24. Sun Y, Zheng L, Deng W, Wang S (2017) Svdnet for pedestrian retrieval. In: Proceeding ICCV, Venice Italy, pp 3800–3808

  25. Tian M, Yi S, Li H, Li S, Zhang X, Shi J, Yan J, Wang X (2018) Eliminating background-bias for robust person reidentification. In: Proc. CVPR, Salt Lake City, SLC, USA, pp 5794–5803

  26. Tzeng E, Hoffman J, Darrell T et al (2015) Simultaneous deep transfer across domains and tasks. In: Proceeding CVPR, Boston, BSN, USA, pp 4068–4076

  27. Wang T, Gong S, Zhu X et al (2014) Person re-identification by video ranking. In: ECCV, pp 688–703

  28. Wang G, Yang S, Liu H, et al. (2020) High-Order Information matters: learning relation and topology for occluded person Re-Identification. In: Proceeding CVPR, Seattle, SEA, USA, pp 6449–6458

  29. Wang J, Zhu X, Gong S et al (2018) Transferable joint attribute-identity deep learning for unsupervised person re-identification. In: Proceeding CVPR, Salt Lake City, SLC, USA, pp 2275–2284

  30. Wei L, Zhang S, Gao W, et al. (2018) Person transfer gan to bridge domain gap for person re-identification. In: Proceeding CVPR, Salt Lake City, SLC, USA, pp 79–88

  31. Wei L, Zhang S, Gao W et al (2018) Person transfer gan to bridge domain gap for person re-identification. In: Proceeding CVPR, Salt Lake City, SLC, USA, pp 79–88

  32. Yang F, Zhong Z, Luo Z et al (2021) Joint noise-tolerant learning and meta camera shift adaptation for unsupervised person Re-Identification. In: Proc CVPR, pp 4855–4864

  33. Zhai Y, Lu S, Ye Q et al (2020) Ad-cluster: Augmented discriminative clustering for domain adaptive person re-identification. In: Proceeding CVPR, Seattle, SEA, USA, pp 9021–9030

  34. Zhao S, Gao C, Zhang J et al (2020) Do Not Disturb Me:, Person Re-identification Under the Interference of Other Persons, arXiv:2008.06963

  35. Zhao F, Liao S, Xie GS et al (2020) Unsupervised domain adaptation with noise resistible mutual-training for person re-identification. In: ECCV, Glasgow, UK, pp 1–18

  36. Zheng M, Karanam S, Wu Z et al (2019) Re-identification with consistent attentive siamese networks. In: Proceeding CVPR, State of California, CA, USA, pp 5735–5744

  37. Zheng L, Shen L, Tian L, et al. (2015) Scalable person re-identification: A benchmark. In: Proceeding CVPR, Boston, BSN, USA, pp 1116–1124

  38. Zheng Z, Yang X, Yu Z et al (2019) Joint discriminative and generative learning for person re-identification. In: Proceeding CVPR, State of California, CA, USA, pp 2138–2147

  39. Zheng Z, Zheng L, Yang Y (2017) Unlabeled samples generated by gan improve the person re-identification baseline in vitro. In: Proceeding CVPR, Hawaii, HI, USA, pp 3754–3762

  40. Zheng Z, Zheng L, Yang Y (2017) Unlabeled samples generated by gan improve the person re-identification baseline in vitro. In: Proceeding ICCV, Venice, Italy, pp 3754–3762

  41. Zhong Z, Zheng L, Luo Z et al (2019) Invariance matters: Exemplar memory for domain adaptive person re-identification. In: Proceeding CVPR, State of California, CA, USA, pp 598–607

  42. Zhong Z, Zheng L, Zheng Z et al (2018) Camera style adaptation for person re-identification. In: Proceeding CVPR, Salt Lake City, SLC, USA, pp 5157–5166

  43. Zhu JY, Park T, Isola P, et al. (2017) Unpaired image-to-image translation using cycleconsistent adversarial networks. In: Proc. CVPR, Hawaii, HI, USA, pp 2223–2232

Download references

Acknowledgements

This work was supported by the National Natural Science Foundation of China (Grant No.42075007), the Open Grants of the State Key Laboratory of Severe Weather(No.2021LASW-B19), the Open Project of Provincial Key Laboratory for Computer Information Processing Technology under Grant KJS1935, Soochow University, and the Priority Academic Program Development of Jiangsu Higher Education Institutions, Graduate Scientific Research Innovation Program of Jiangsu Province under Grant no. KYCX21_1015.

Author information

Authors and Affiliations

  1. School of Computer, Software, Engineering Research Center of Digital Forensics, Ministry of Education, Nanjing University of Information Science, Technology, Nanjing, China

    Wei Fang, Weinan Yi & Lin Pang

  2. State Key Laboratory of Severe Weather, Chinese Academy of Meteorological Sciences, Beijing, China

    Wei Fang

  3. Provincial Key Laboratory for Computer Information Processing Technology, Soochow University, Suzhou, China

    Wei Fang

  4. Texas Tech University, Lubbock, USA

    Victor S. Sheng

Authors
  1. Wei Fang
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Weinan Yi
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Lin Pang
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Victor S. Sheng
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Weinan Yi.

Ethics declarations

Conflict of Interests

The Authors declares that there is no conflict of interest.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Springer Nature or its licensor holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Fang, W., Yi, W., Pang, L. et al. Study of cross-domain person re-identification based on DCGAN. Multimed Tools Appl 81, 36551–36565 (2022). https://doi.org/10.1007/s11042-022-13526-3

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11042-022-13526-3

Keywords

Search

Navigation