Quality of tracking methods with object reidentification
https://doi.org/10.21822/2073-6185-2024-51-3-103-109
Abstract
Objective. This paper presents a study of object tracking algorithms for analyzing sports competitions with a minimum permissible number of frames per second without critical loss of quality of operational systems. The purpose of the study is to determine the optimal tracking method and reidentification model with an increase in the time interval between frames. Method. The study used the Re-Identification MSMT17, Market1501, DukeMTMCreID datasets; specialized neural networks OsNet, ResNet and MobileNet. Tracking algorithms BotSort, Bytetrack, OcSort, DeepOcSort, StrongSort. The NOTA and IDF1 quality metrics were used. Result. The metrics of the experiments are generally quite low, which is due to the nonlinearity of athletes' movement, unlike pedestrians' movement, and a large number of changes in their posture. Conclusion. Depending on the degree of information compression, the optimal tracking methods when using pre-trained reidentification models are ByteTrack and OcSort.
About the Authors
N. A. Maltceva
Ural Federal University
Russian Federation
Russian Federation
Natalya A. Maltceva, Engineer
19 Mira St., Ekaterinburg 620014
A. A. Polozov
Ural Federal University
Russian Federation
Russian Federation
Andrey A. Polozov, Dr. Sci. (Pedagogical), Prof., Department of Information Technologies and Control Systems
19 Mira St., Ekaterinburg 620014
N. V. Papulovskaya
Ural Federal University
Russian Federation
Russian Federation
Natalya V. Papulovskaya; Cand. Sci. (Pedagogical); Assoc. Prof., Department of Information Technologies and Control Systems
19 Mira St., Ekaterinburg 620014
S. L. Goldshtein
Ural Federal University
Russian Federation
Russian Federation
Sergey L. Goldshtein, Dr. Sci. (Eng), Prof., Prof., Department of Technical Physics
19 Mira St., Ekaterinburg 620014
References
1. H. Masson, A. Bhuiyan, L. Th. Nguyen-Meidine, M. Javan, P. Siva, I. B. Ayed, E. Granger, “Exploiting Prunability for Person Re-Identification”, 2021; arXiv:1907.02547.
2. L. Wu, Y. Wang, J. Gao, D. Tao, “Deep Co-attention based Comparators For Relative Representation Learning in Person Re-identification”, 2018; arXiv:1804.11027.
3. L. Wei, S. Zhang, W. Gao, Qi Tian, “Person Transfer GAN to Bridge Domain Gap for Person Re-Identification”, 2018; arXiv:1711.08565.
4. E. Ristani, F. SoleraRoger, Z. Cucchiara, and C. Tomasi, “Performance measures and a data set for multitarget, multi-camera tracking,”in ECCV Workshop on Benchmarking Multi-Target Tracking, 2016;17 – 35.
5. K. Zhou, Y. Yang, A. Cavallaro, T. Xiang, “Learning Generalisable Omni-Scale Representations for Person Re-Identification”, 2021; arXiv:1910.06827.
6. L. Zheng, S. Wang, L. Shen, L. Tian, J. Bu, Q. Tian, “Person Re-identification Meets Image Search”, 2015; arXiv:1502.02171.
7. W. Li, R. Zhao,T. Xiao, X.G. Wang, ”DeepReID: Deep Filter Pairing Neural Network for Person Re-Identification”, 2014, DOI:10.1109/CVPR.2014.27.
8. K. He, X. Zhang, Shaoqing Ren, Jian Sun, “Deep Residual Learning for Image Recognition”, 2015; arXiv:1512.03385.
9. M. Sandler, A. Howard, M. Zhu, A. Zhmoginov, L.-C. Chen, “MobileNetV2: Inverted Residuals and Linear Bottlenecks”, 2018; arXiv:1801.04381.
10. K. Zhou, T. Xiang, “Torchreid: A Library for Deep Learning Person Re-Identification in Pytorch”, 2019; arXiv:1910.10093.
11. N. Aharon, R. Orfaig, B.-Z. Bobrovsky, “BoT-SORT: Robust Associations Multi-Pedestrian Tracking”, 2022; arXiv:2206.14651.
12. Y. Zhang, P. Sun, Y. Jiang, D. Yu, F. Weng, Z. Yuan, P. Luo, W. Liu, X. Wang. “ByteTrack: Multi-Object Tracking by Associating Every Detection Box”, 2022; arXiv:2110.06864v3.
13. J. Cao, J. Pang, X. Weng, R. Khirodkar, K. Kitani, “Observation-Centric SORT: Rethinking SORT for Robust Multi-Object Tracking”, 2023; arXiv:2203.14360
14. N. Wojke, A. Bewley, D. Paulus, “Simple online and realtime tracking with a deep association metric”, ICIP. IEEE, 2017;3645 – 3649.
15. G.Maggiolino, A. Ahmad, J. Cao, K. Kitani, “Deep OC-SORT: Multi-Pedestrian Tracking by Adaptive Re-Identification”, 2023; arXiv:2302.11813v1
16. Y.Du, Z.Zhao, Y.Song, Y.Zhao, F.Su, T.Gong, H. Meng, “StrongSORT: Make DeepSORT Great Again”, 2023; arXiv:2202.13514v2
17. J. Luiten, A.Osep, P, Dendorfer, · P. Torr, · A. Geiger, · L. Leal-Taixe, · B. Leibe1, “HOTA: A Higher Order Metric for Evaluating Multi-object Tracking. International Journal of Computer Vision”, International Journal of Computer Vision, 2021; 548 – 578.
Review
For citations:
Maltceva N.A., Polozov A.A., Papulovskaya N.V., Goldshtein S.L. Quality of tracking methods with object reidentification. Herald of Dagestan State Technical University. Technical Sciences. 2024;51(3):103-109. (In Russ.) https://doi.org/10.21822/2073-6185-2024-51-3-103-109
Views:
185
Email this article 
