Region-based convolutional neural networks for occluded person re-identification

(1) * Atiqul Islam Mail (Swinburne University of Technology Sarawak Campus, Malaysia)
(2) Mark Tee Kit Tsun Mail (Swinburne University of Technology Sarawak Campus, Malaysia)
(3) Lau Bee Theng Mail (Swinburne University of Technology Sarawak Campus, Malaysia)
(4) Caslon Chua Mail (Swinburne University of Technology, Melbourne, Australia)
*corresponding author


In a variety of applications, including intelligent surveillance systems, targeted tracking, and assistive human-following robots, the ability to accurately identify individuals even when they are partially obscured is imperative. Such Continuous person tracking is complicated by the close similarity between the appearance of people and target occlusions. This study addresses this significant challenge by proposing a two-step, detection-first approach that uses a region-based convolutional neural network (R-CNN) as the re-identification (re-ID)solution. The model is specifically trained to detect occluded persons at different levels of occlusion before forwarding the image for the re-ID process. Three occluded-specific datasets are selected to evaluate the model's effectiveness in detecting occluded people. There are 379 distinct people in total, and each has five images obstructed from different angles. A sample of the data is taken to simulate various environment settings, and new data points are generated with different degrees of occlusion to assess how well the model performs under varying levels of obstruction. The findings demonstrate that the proposed person re-ID model is reliable in most circumstances, correctly re-identifying at 74% (Rank-1) and 90% (Rank-5). Although there is a decrease in accuracy as the number of distinctive people in the dataset increases, this does not significantly impact the tracking performance in various applications, which are expected to recognize a single person or a small group of individuals. Future works will explore refining similarity matching algorithms by delving into robust image comparison techniques, thereby addressing the challenges presented by occlusions. A critical aspect is to assess the model under diverse lighting conditions and investigate scenarios with multiple individuals in a frame. It is also beneficial to exploit high-resolution datasets, such as DukeMTMC-reID, and integrate finer contextual details, like clothing or carried objects. These collective efforts are essential for optimizing the model’s efficacy in practical applications and advancing person re-ID technologies.


Occlusion; R-CNN; Re-identification; Region re-ranking



Article metrics

Abstract views : 329 | PDF views : 71




Full Text



[1] A. Islam, M. K. T. Tee, and B. T. Lau, “Development of an Improved Occluded Person Re-Identification System Using Deep Learning,” in 2022 6th High Performance Computing and Cluster Technologies Conference (HPCCT), Jul. 2022, pp. 44–50, doi: 10.1145/3560442.3560449.

[2] M. Ye, J. Shen, G. Lin, T. Xiang, L. Shao, and S. C. H. Hoi, “Deep Learning for Person Re-Identification: A Survey and Outlook,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 44, no. 6, pp. 2872–2893, Jun. 2022, doi: 10.1109/TPAMI.2021.3054775.

[3] X.-T. Vo and K.-H. Jo, “Accurate Bounding Box Prediction for Single-Shot Object Detection,” IEEE Trans. Ind. Informatics, vol. 18, no. 9, pp. 5961–5971, Sep. 2022, doi: 10.1109/TII.2021.3138336.

[4] J. Miao, Y. Wu, P. Liu, Y. Ding, and Y. Yang, “Pose-Guided Feature Alignment for Occluded Person Re-Identification,” in 2019 IEEE/CVF International Conference on Computer Vision (ICCV), Oct. 2019, vol. 2019-Octob, pp. 542–551, doi: 10.1109/ICCV.2019.00063.

[5] D. Wu et al., “Random Occlusion Recovery for Person Re-identification,” J. Imaging Sci. Technol., vol. 63, no. 3, pp. 030405-1-030405–9, May 2019, doi: 10.2352/J.ImagingSci.Technol.2019.63.3.030405.

[6] X. Liu, Y. Jiang, P. Jain, and K.-H. Kim, “TAR: Enabling Fine-Grained Targeted Advertising in Retail Stores,” in Proceedings of the 16th Annual International Conference on Mobile Systems, Applications, and Services, Jun. 2018, pp. 323–336, doi: 10.1145/3210240.3210342.

[7] C. B. Nalty et al., “A Brief Survey on Person Recognition at a Distance,” in 2022 56th Asilomar Conference on Signals, Systems, and Computers, Oct. 2022, vol. 2022-Octob, pp. 145–152, doi: 10.1109/IEEECONF56349.2022.10051819.

[8] R. Hou, B. Ma, H. Chang, X. Gu, S. Shan, and X. Chen, “VRSTC: Occlusion-Free Video Person Re-Identification,” in 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Jun. 2019, vol. 2019-June, pp. 7176–7185, doi: 10.1109/CVPR.2019.00735.

[9] X. Zhang, Y. Yan, J.-H. Xue, Y. Hua, and H. Wang, “Semantic-Aware Occlusion-Robust Network for Occluded Person Re-Identification,” IEEE Trans. Circuits Syst. Video Technol., vol. 31, no. 7, pp. 2764–2778, Jul. 2021, doi: 10.1109/TCSVT.2020.3033165.

[10] C. Zhao, X. Lv, S. Dou, S. Zhang, J. Wu, and L. Wang, “Incremental Generative Occlusion Adversarial Suppression Network for Person ReID,” IEEE Trans. Image Process., vol. 30, pp. 4212–4224, 2021, doi: 10.1109/TIP.2021.3070182.

[11] L. Zheng, Y. Yang, and A. G. Hauptmann, “Person Re-identification: Past, Present and Future,” arXiv Computer Vision and Pattern Recognition, Oct. 10, pp. 1-20, 2016.

[12] Q. Leng, M. Ye, and Q. Tian, “A Survey of Open-World Person Re-Identification,” IEEE Trans. Circuits Syst. Video Technol., vol. 30, no. 4, pp. 1092–1108, Apr. 2020, doi: 10.1109/TCSVT.2019.2898940.

[13] A. Zahra, N. Perwaiz, M. Shahzad, and M. M. Fraz, “Person re-identification: A retrospective on domain specific open challenges and future trends,” Pattern Recognit., vol. 142, p. 109669, Oct. 2023, doi: 10.1016/j.patcog.2023.109669.

[14] R. Chalapathy, N. L. D. Khoa, and S. Chawla, “Robust Deep Learning Methods for Anomaly Detection,” in Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, Aug. 2020, pp. 3507–3508, doi: 10.1145/3394486.3406704.

[15] L. Liu et al., “Deep Learning for Generic Object Detection: A Survey,” Int. J. Comput. Vis., vol. 128, no. 2, pp. 261–318, Feb. 2020, doi: 10.1007/s11263-019-01247-4.

[16] J. Marin, D. Vazquez, A. M. Lopez, J. Amores, and L. I. Kuncheva, “Occlusion Handling via Random Subspace Classifiers for Human Detection,” IEEE Trans. Cybern., vol. 44, no. 3, pp. 342–354, Mar. 2014, doi: 10.1109/TCYB.2013.2255271.

[17] D. G. Lowe, “Distinctive Image Features from Scale-Invariant Keypoints,” Int. J. Comput. Vis., vol. 60, no. 2, pp. 91–110, Nov. 2004, doi: 10.1023/B:VISI.0000029664.99615.94.

[18] Y. Hu, S. Liao, Z. Lei, D. Yi, and S. Z. Li, “Exploring Structural Information and Fusing Multiple Features for Person Re-identification,” in 2013 IEEE Conference on Computer Vision and Pattern Recognition Workshops, Jun. 2013, pp. 794–799, doi: 10.1109/CVPRW.2013.119.

[19] G. Xie et al., “Pose-guided feature region-based fusion network for occluded person re-identification,” Multimed. Syst., vol. 29, no. 3, pp. 1771–1783, Jun. 2023, doi: 10.1007/s00530-021-00752-2.

[20] D. K. Dastur et al., “The B-vitamins in malnutrition with alcoholism: A model of intervitamin relationships,” Br. J. Nutr., vol. 36, no. 2, pp. 143–159, Sep. 1976, doi: 10.1017/S0007114500020158.

[21] J. Miao, Y. Wu, and Y. Yang, “Identifying Visible Parts via Pose Estimation for Occluded Person Re-Identification,” IEEE Trans. Neural Networks Learn. Syst., vol. 33, no. 9, pp. 4624–4634, Sep. 2022, doi: 10.1109/TNNLS.2021.3059515.

[22] Z. Zhao, R. Song, Q. Zhang, P. Duan, and Y. Zhang, “JoT-GAN: A Framework for Jointly Training GAN and Person Re-Identification Model,” ACM Trans. Multimed. Comput. Commun. Appl., vol. 18, no. 1s, pp. 1–18, Feb. 2022, doi: 10.1145/3491225.

[23] L. Wei, S. Zhang, W. Gao, and Q. Tian, “Person Transfer GAN to Bridge Domain Gap for Person Re-identification,” in 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Jun. 2018, pp. 79–88, doi: 10.1109/CVPR.2018.00016.

[24] C. Zhang, L. Zhu, S. Zhang, and W. Yu, “PAC-GAN: An effective pose augmentation scheme for unsupervised cross-view person re-identification,” Neurocomputing, vol. 387, pp. 22–39, Apr. 2020, doi: 10.1016/j.neucom.2019.12.094.

[25] T. He, X. Shen, J. Huang, Z. Chen, and X.-S. Hua, “Partial Person Re-identification with Part-Part Correspondence Learning,” in 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Jun. 2021, pp. 9101–9111, doi: 10.1109/CVPR46437.2021.00899.

[26] Y. Li, J. He, T. Zhang, X. Liu, Y. Zhang, and F. Wu, “Diverse Part Discovery: Occluded Person Re-identification with Part-Aware Transformer,” in 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Jun. 2021, pp. 2897–2906, doi: 10.1109/CVPR46437.2021.00292.

[27] X.-P. Lin and Y.-B. Yang, “An Adaptive Part-Based Model For Person Re-Identification,” in ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Jun. 2021, vol. 2021-June, pp. 1965–1969, doi: 10.1109/ICASSP39728.2021.9415086.

[28] Z. Yao, X. Wu, Z. Xiong, and Y. Ma, “A Dynamic Part-Attention Model for Person Re-Identification,” Sensors, vol. 19, no. 9, p. 2080, May 2019, doi: 10.3390/s19092080.

[29] L. Zhao, X. Li, Y. Zhuang, and J. Wang, “Deeply-Learned Part-Aligned Representations for Person Re-identification,” in 2017 IEEE International Conference on Computer Vision (ICCV), Oct. 2017, vol. 2017-Octob, pp. 3239–3248, doi: 10.1109/ICCV.2017.349.

[30] W.-S. Zheng, X. Li, T. Xiang, S. Liao, J. Lai, and S. Gong, “Partial Person Re-Identification,” in 2015 IEEE International Conference on Computer Vision (ICCV), Dec. 2015, pp. 4678–4686, doi: 10.1109/ICCV.2015.531.

[31] i-LIDS Team, “Imagery Library for Intelligent Detection Systems (i-LIDS) A Standard for Testing Video Based Detection Systems,” in Proceedings 40th Annual 2006 International Carnahan Conference on Security Technology, Oct. 2006, pp. 75–80, doi: 10.1109/CCST.2006.313432.

[32] M. Everingham, L. Van Gool, C. K. I. Williams, J. Winn, and A. Zisserman, “The PASCAL Visual Object Classes Challenge 2007 (VOC2007) Development Kit,” Pattern Analysis, Statistical Modelling and Computational Learning, Tech. Rep, pp. 1-45, 2012.

[33] G. Wang, Y. Yuan, X. Chen, J. Li, and X. Zhou, “Learning Discriminative Features with Multiple Granularities for Person Re-Identification,” in Proceedings of the 26th ACM international conference on Multimedia, Oct. 2018, pp. 274–282, doi: 10.1145/3240508.3240552.

[34] J. Dang, X. Tang, and S. Li, “HA-FPN: Hierarchical Attention Feature Pyramid Network for Object Detection,” Sensors, vol. 23, no. 9, p. 4508, May 2023, doi: 10.3390/s23094508.

[35] Y. Wang, S. Yang, S. Liu, and Z. Zhang, “Cross-Domain Person Re-identification: A Review,” in Lecture Notes in Electrical Engineering, vol. 653, Springer Science and Business Media Deutschland GmbH, 2021, pp. 153–160, doi: 10.1007/978-981-15-8599-9_19.

[36] S. Liao, Y. Hu, Xiangyu Zhu, and S. Z. Li, “Person re-identification by Local Maximal Occurrence representation and metric learning,” in 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Jun. 2015, vol. 07-12-June, pp. 2197–2206, doi: 10.1109/CVPR.2015.7298832.

[37] Y.-C. Chen, W.-S. Zheng, and J. Lai, “Mirror Representation for Modeling View-Specific Transform in Person Re-Identification,” in Proceedings of the Twenty-Fourth International Joint Conference on Artificial Intelligence (IJCAI 2015), 2016, pp. 3402–3408. [Online]. Available at:

[38] N. Perwaiz, M. M. Fraz, and M. Shahzad, “Hierarchical Refined Local Associations for Robust Person Re-Identification,” in 2019 International Conference on Robotics and Automation in Industry (ICRAI), Oct. 2019, pp. 1–6, doi: 10.1109/ICRAI47710.2019.8967389.

[39] L. Zhang, T. Xiang, and S. Gong, “Learning a Discriminative Null Space for Person Re-identification,” in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Jun. 2016, vol. 2016-Decem, pp. 1239–1248, doi: 10.1109/CVPR.2016.139.

[40] Y. Sun, L. Zheng, W. Deng, and S. Wang, “SVDNet for Pedestrian Retrieval,” in Proceedings of the IEEE International Conference on Computer Vision, 2018, pp. 3820–3828, doi: 10.1109/ICCV.2017.410.

[41] Z. Zhong, L. Zheng, G. Kang, S. Li, and Y. Yang, “Random Erasing Data Augmentation,” Proc. AAAI Conf. Artif. Intell., vol. 34, no. 07, pp. 13001–13008, Apr. 2020, doi: 10.1609/aaai.v34i07.7000.

[42] C. Yan, G. Pang, J. Jiao, X. Bai, X. Feng, and C. Shen, “Occluded Person Re-Identification with Single-scale Global Representations,” in 2021 IEEE/CVF International Conference on Computer Vision (ICCV), Oct. 2021, pp. 11855–11864, doi: 10.1109/ICCV48922.2021.01166.

[43] Q. Yang, P. Wang, Z. Fang, and Q. Lu, “Focus on the Visible Regions: Semantic-Guided Alignment Model for Occluded Person Re-Identification,” Sensors, vol. 20, no. 16, p. 4431, Aug. 2020, doi: 10.3390/s20164431.

[44] X. Zhong, M. Wang, W. Liu, J. Yuan, and W. Huang, “SCPNet: Self-constrained parallelism network for keypoint-based lightweight object detection,” J. Vis. Commun. Image Represent., vol. 90, p. 103719, Feb. 2023, doi: 10.1016/j.jvcir.2022.103719.

[45] L. He, J. Liang, H. Li, and Z. Sun, “Deep Spatial Feature Reconstruction for Partial Person Re-identification: Alignment-free Approach,” in 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Jun. 2018, pp. 7073–7082, doi: 10.1109/CVPR.2018.00739.

[46] H. Luo, W. Jiang, X. Fan, and C. Zhang, “STNReID: Deep Convolutional Networks With Pairwise Spatial Transformer Networks for Partial Person Re-Identification,” IEEE Trans. Multimed., vol. 22, no. 11, pp. 2905–2913, Nov. 2020, doi: 10.1109/TMM.2020.2965491.

[47] Y. Sun et al., “Perceive Where to Focus: Learning Visibility-Aware Part-Level Features for Partial Person Re-Identification,” in 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Jun. 2019, vol. 2019-June, pp. 393–402, doi: 10.1109/CVPR.2019.00048.

[48] L. He, X. Liao, W. Liu, X. Liu, P. Cheng, and T. Mei, “FastReID: A Pytorch Toolbox for General Instance Re-identification,” in Proceedings of the 31st ACM International Conference on Multimedia, Oct. 2023, pp. 9664–9667, doi: 10.1145/3581783.3613460.

[49] Z. Pang, J. Guo, W. Sun, Y. Xiao, and M. Yu, “Cross-domain person re-identification by hybrid supervised and unsupervised learning,” Appl. Intell., vol. 52, no. 3, pp. 2987–3001, Feb. 2022, doi: 10.1007/s10489-021-02551-8.

[50] H. Zhang, S. Wang, N. Wang, S. Liu, and Z. Zhang, “Efficiency Evaluation of Deep Model for Person Re-identification,” in Lecture Notes in Electrical Engineering, vol. 653, Springer Science and Business Media Deutschland GmbH, 2021, pp. 130–136, doi: 10.1007/978-981-15-8599-9_16.

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

International Journal of Advances in Intelligent Informatics
ISSN 2442-6571  (print) | 2548-3161 (online)
Organized by UAD and ASCEE Computer Society
Published by Universitas Ahmad Dahlan
E: (paper handling issues) (publication issues)

View IJAIN Stats

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0