V.V. Khryashchev1, A.L. Priorov2, D.V. Matveev3, Yu.A. Lukashevich4

1–4 Yaroslavl State University named after P.G. Demidov (Yaroslavl, Russia)

1 v.khryashchev@uniyar.ac.ru, 2 andcat@yandex.ru, 3 diman@uniyar.ac.ru, 4 lukashevich.yuriy@gmail.com

The task of detecting and tracking people on video data is relevant in many areas of computer vision, in particular in the field of sports for collecting statistics about players. The aim of the work is to detect and accompany athletes in the video data stream using deep machine learning methods. The results of a study of the JDE algorithm for detecting and tracking targets on video data, namely athletes on video recordings from competitions, are presented. The developed convolutional neural network has been trained and tested on the NVIDIA DGX-1 supercomputer. To analyze the quality of the model, the MOTA indicator was used, which directly correlates with how the human eye tracks the target objects in the video stream. The quality of the JDE algorithm is evaluated on videos with excerpts of basketball team games held in the gym of the P.G. Demidov YarSU. The obtained research results can be used in various sports to collect statistics about players in order to improve the effectiveness of their performances.

Khryashchev V.V., Priorov A.L., Matveev D.V., Lukashevich Yu.A. Using deep machine learning techniques to detect and track athletes in the video data stream. Neurocomputers. 2023. V. 25. № 6. Р. 37-46. DOI: https://doi.org/10.18127/j19998554-202306-04 (In Russian)

1. Budaev E.S., Mikhailova S.S., Evdokimova I.S., Khalmakshinov E.A. Development of a neural network model for detecting ob-jects in a video stream. Neurocomputers. 2023. V. 25. № 4. Р. 54–64. DOI 10.18127/j19998554-202304-07 (In Russian)
2. Minaev E.Yu., Kutikova V.V., Nikonorov A.V. Tracking objects in a video stream based on convolutional neural networks and fractal analysis. Proceedings of the IV International Conference and Youth School "Information Technologies and Nanotechnology" (ITNT-2018). Samara: New technology. 2018. P. 2792–2798. (In Russian)
3. Goodfellow Ya., Benjio I., Courville A. Deep learning. M.: DMK Press. 2017. 652 p. (In Russian)
4. Nikolenko S., Kadurin A., Arkhangelskaya E. Deep learning. Dive into the world of neural networks. St. Petersburg: Peter. 2018. 480 p. (In Russian)
5. Yu F., Li W., Li Q., Liu Y., Shi X., Yan J. Poi: Multiple object tracking with high performance detection and appearance feature. Lecture Notes in Computer Science. 2016. V. 9914. P. 36–42. DOI 10.1007/978-3-319-48881-3_3.
6. Komissarenko N. 3 methods of detecting objects with Deep Learning: RCON, Fast RUN and Master-CN. [Electronic resource] – Access mode: https://medium.com/@bigdataschool/3-метода-детектирования-объектов-c-deep-learning-r-cnn-fast-r-cnn-и-faster-r-cnn-acdf6380fd33, date of reference 05.07.2023. (In Russian)
7. Wang Z., Zheng L., Liu Y., Wang S. Towards Real-Time Multi-Object Tracking. Lecture Notes in Computer Science. 2020. V. 12356. P. 107–122. DOI 10.1007/978-3-030-58621-8_7.
8. Dollar P., Wojek C., Schiele B., Perona P. Pedestrian detection: A benchmark. IEEE Conference on Computer Vision and Pattern Recognition. 2009. P. 304–311.
9. Xiao T., Li S., Wang B., Lin L., Wang X. Joint Detection and Identification Feature Learning for Person Search. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2017. P. 3376–3385. DOI 10.1109/CVPR.2017.360.
10. Zheng L., Zhang H., Sun S., Chandraker M., Yang Y., Tian Q. Person Re-identification in the Wild. IEEE Conference on Computer Vision and Pattern Recognition. 2017. P. 3346–3355. DOI 10.1109/CVPR.2017.357.
11. Milan A., Leal-Taixe L., Reid I., Roth S., Schindler K. MOT16: A Benchmark for Multi-Object Tracking. [Electronic resource] – Access mode: https://arxiv.org/pdf/1603.00831.pdf, date of reference 05.07.2023.
12. Redmon J., Farhadi A. YOLOv3: An Incremental Improvement. [Electronic resource] – Access mode: https://arxiv.org/pdf/1804. 02767v1.pdf, date of reference 05.07.2023.
13. Lin T.-Y., Dollar P., Girshick R., He K., Hariharan B., Belongie S. Feature Pyramid Networks for Object Detection. [Electronic resource] – Access mode: https://arxiv.org/pdf/1612.03144.pdf, date of reference 05.07.2023.
14. Pointer Ya. Programming with Pwtorch. St. Petersburg: Peter. 2020. 256 p. (In Russian)
15. Vlasov A.I., Zavyalov N.V., Selivanov K.V., Skalchenkov I.I. Application of neural networks in detecting PCB defects. Neurocomputers. 2022. V. 24. № 6. Р. 5–19. DOI 10.18127/j19998554-202206-01. (In Russian)
16. McMahan B. Getting to know PyTorch: Deep learning in natural language processing. St. Petersburg: Peter. 2020. 250 p. (In Russian)
17. Sanders J., Candrot E. CUDA technology in examples. Introduction to graphics processor programming. Moscow: DMK Press. 2013. 232 p. (In Russian)
18. Gonzalez R., Woods R. Digital image processing. M.: Technosphere. 2005. 621 p. (In Russian)
19. Shchelkunov A.E., Kovalev V.V., Morev K.I., Sidko I.V. Metrics for evaluating automatic tracking algorithms. Taganrog: News of the Southern Federal University. Technical sciences. 2020. № 1. P. 233–245. (In Russian)