Bayesian classification algorithm in the person re-identification task

Purpose of research. Development and experimental evaluation of a Bayesian classification algorithm for the person re-identification task using images from multiple surveillance cameras. The study aims to improve identification accuracy through integrating features derived from facial and silhouette images.

Methods. The proposed algorithm utilizes a Bayesian classification model based on multivariate normal distributions of features. These features are extracted by neural encoders built on the Vision Transformer architecture and trained using the ArcFace loss function. Integration of modality-specific features is performed by computing logarithmic posterior probabilities of class membership. The effectiveness of the method was evaluated using the open CUHK03 dataset, quantitative analysis via ROC curves, and feature space visualization using the t-SNE method.

Results. The algorithm demonstrated high classification performance: precision of 95.65% on CUHK03, up to 97.7% on Market-1501, and 89.2% on MARS. ROC analysis confirmed strong class separability, while t-SNE visualizations showed compact and well-defined clusters. The algorithm is deterministic, robust to noise, and scalable to larger datasets.

Conclusion. The developed Bayesian classification algorithm has proven its effectiveness and feasibility for person re-identification tasks in intelligent video surveillance systems. Its advantages include high accuracy, interpretability, and potential for integrating additional features. Future research should focus on incorporating extra attributes and evaluating algorithm performance on significantly larger and more diverse datasets..

1. Ye M., Shen J., Lin G., Xiang T., Shao L., Hoi S.C.H. Deep learning for person reidentification: A survey and outlook. IEEE Trans. Pattern Anal. Mach. Intell. 2021; 44(6): 2872–2893.

2. Ahmed E., Jones M., Marks T.K. An improved deep learning architecture for person re-identification. In: Proc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), 2015. P. 3908–3916.

3. Varior R.R., Haloi M., Wang G. Gated siamese convolutional neural network architecture for human re-identification. In: Proc. European Conf. on Computer Vision (ECCV), 2016. P. 791–808.

4. Cho Y., Kim J., Kim W. J., Jung J., Yoon S. Generalizable Person Re-identification via Balancing Alignment and Uniformity. arXiv preprint arXiv:2411.11471. 2024.

5. Luo H., Gu Y., Liao X., Lai S., Jiang W. Bag of tricks and a strong baseline for person re-identification. In: Proc. IEEE Conf. on Computer Vision and Pattern Recognition Workshops (CVPRW), 2019.P. 1487–1495.

6. Yadav A., Vishwakarma D. K. Deep learning algorithms for person re-identification: state-of-the-art and research challenges. Multimedia Tools and Applications. 2023; 83: 22005–22054.

7. Krivenko M.P. Bayesian classification of multidimensional data series. Sistemy i sredstva informatiki. 2020; 30(1): 34–45. (In Russ.).

8. Saburov V.S. Bayesian classifier in machine learning. Shag v nauku. 2024; (1): 78–81. (In Russ.).

9. Moghaddam B., Jebara T., Pentland A. Bayesian face recognition. Pattern Recognition. 2000; 33(11): 1771–1782.

10. Chen D., Cao X., Wang L., Wen F., Sun J. Bayesian face revisited: A joint formulation. In: Proc. European Conf. on Computer Vision (ECCV), 2012. P. 566–579.

11. Hermans A., Beyer L., Leibe B. (2017). In Defense of the Triplet Loss for Person Re-Identification. arXiv preprint arXiv:1703.07737.

12. Liao S., Hu Y., Zhu X., Li S.Z. Person re-identification by local maximal occurrence representation and metric learning. In: Proc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR). 2015. P. 2197–2206.

13. Liong V.E., Lu J., Ge Y. Regularized Bayesian metric learning for person reidentification. In: Proc. ECCV Workshops, 2014. Part III, LNCS 8927. P. 209–224.

14. Zhong Z., Zheng L., Cao D., Li S. Re-ranking person re-identification with kreciprocal encoding. In: Proc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), 2017. P. 1318–1327.

15. Bobyr M.V., Khrapova N.I. Information-analytical system for detecting object movements on pedestrian crossings. Ontologiya proektirovaniya. 2024; 14(4): 531–541. (In Russ.).

16. Bobyr M.V., Milostnaya N.A., Khrapova N.I. An approach to pedestrian motion detection using histogram of oriented gradients. Elektronnye biblioteki = Elektronnye biblioteki, 2024, 27(4): 429–447. (In Russ.).

17. Rusakov K.D. Person re-identification algorithm based on deep convolutional neural networks. Upravlenie bol'shimi sistemami: sbornik trudov. 2025; (110). (In Russ.).

18. Yu C., Gao C., Wang J., Yu G., Shen C., Sang N. BiSeNet V2: Bilateral network with guided aggregation for real-time semantic segmentation. Int. J. Comput. Vis., 2020.

19. Ignatieva S.A., Bogush R.P. Person re-identification using machine learning based on surveillance systems data. In: Iskusstvennyi intellekt v Belarusi: materialy II Foruma = In: Artificial Intelligence in Belarus: Proceedings of the II Forum. Minsk; 2023. P. 112–119. (In Russ.).

20. Ye M., Shen J., Lin G., Xiang T., Shao L., Hoi S. C. H. Deep Learning for Person Re-identification: A Survey and Outlook. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2021. 44(6): 2872–2893.

21. Liu W., Chang X., Chen L., Yang Y. Semi-supervised Bayesian attribute learning for person re-identification. In: Proc. AAAI Conf. on Artificial Intelligence, 2018. P. 680–687.

22. Zhao H., Ouyang W., Li X., Wang X. Spindle net: Person re-identification with human body region guided feature decomposition and fusion. In: Proc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), 2017. P. 907–915.