AN INTELLIGENT VISION-BASED APPROACH FOR REAL-TIME MULTI-OBJECT TRACKING
Keywords
DOI:
https://doi.org/10.34238/tnu-jst.14521Abstract
This paper presents a real-time multi-object tracking system that integrates the YOLOv10 detector with the DeepSORT tracker. The two-branch architecture links per-frame object detection (bounding boxes, class labels, and confidence scores) to temporally consistent identity assignment, enabling end-to-end video analytics. We prepare video data via frame extraction and standard preprocessing/augmentation; YOLOv10 is trained with a conventional configuration and then coupled with DeepSORT for association and track life-cycle management. Evaluation uses Precision, Recall, Tracking Accuracy, IDF1, ID Switches, and FPS (Frame Per Second) to jointly assess detection quality, identity stability, and throughput. On 720p videos running on GPU (Graphics Processing Unit), the pipeline operates stably at approximately 7–8 FPS while maintaining a favorable balance between detection accuracy and identity consistency under crowded scenes with brief occlusions and inter-object crossings. These findings indicate the practical utility of the proposed system for surveillance, intelligent transportation, and behavior analysis, and suggest clear avenues for future work, including parameter optimization, lightweight appearance modeling, and expanded training data to further improve robustness and generalization.Downloads
References
H.T. Duong, V.T. Le, and V. T. Hoang, “Deep Learning-Based Anomaly Detection in Video Surveillance: A Survey,” Sensors, vol. 23, no. 11, Jan. 2023, doi: 10.3390/s23115024.
G. Ciaparrone, F. Luque Sánchez, S. Tabik, L. Troiano, R. Tagliaferri, and F. Herrera, “Deep learning in video multi-object tracking: A survey,” Neurocomputing, vol. 381, pp. 61-88, Mar. 2020, doi: 10.1016/j.neucom.2019.11.023.
N. Wojke, A. Bewley, and D. Paulus, “Simple Online and Realtime Tracking with a Deep Association Metric,” arXiv: arXiv:1703.07402, Mar. 21, 2017, doi: 10.48550/arXiv.1703.07402.
X. Zhou, V. Koltun, and P. Krähenbühl, “Tracking Objects as Points,” arXiv: arXiv:2004.01177, Aug. 21, 2020, doi: 10.48550/arXiv.2004.01177.
Y. Zhang, C. Wang, X. Wang, W. Zeng, and W. Liu, “FairMOT: On the Fairness of Detection and Re-Identification in Multiple Object Tracking,” Int. J. Comput. Vis., vol. 129, no. 11, pp. 3069-3087, Nov. 2021, doi: 10.1007/s11263-021-01513-4.
E. Ristani, F. Solera, R. S. Zou, R. Cucchiara, and C. Tomasi, “Performance Measures and a Data Set for Multi-Target, Multi-Camera Tracking,” arXiv: arXiv:1609.01775, Sept. 19, 2016, doi: 10.48550/arXiv.1609.01775.
Y. Zhang et al., “ByteTrack: Multi-Object Tracking by Associating Every Detection Box,” arXiv: arXiv:2110.06864, Apr. 07, 2022, doi: 10.48550/arXiv.2110.06864.
L. Leal-Taixé, A. Milan, I. Reid, S. Roth, and K. Schindler, “MOTChallenge 2015: Towards a Benchmark for Multi-Target Tracking,” arXiv: arXiv:1504.01942, Apr. 08, 2015, doi: 10.48550/arXiv.1504.01942.
P. Dendorfer et al., “MOT20: A benchmark for multi object tracking in crowded scenes,” arXiv: arXiv:2003.09003, Mar. 19, 2020, doi: 10.48550/arXiv.2003.09003.
N. Aharon, R. Orfaig, and B.-Z. Bobrovsky, “BoT-SORT: Robust Associations Multi-Pedestrian Tracking,” arXiv: arXiv:2206.14651, July 07, 2022, doi: 10.48550/arXiv.2206.14651.
A. Wang et al., “YOLOv10: Real-Time End-to-End Object Detection,” arXiv: arXiv:2405.14458, Oct. 30, 2024, doi: 10.48550/arXiv.2405.14458.
A. Bewley, Z. Ge, L. Ott, F. Ramos, and B. Upcroft, “Simple Online and Realtime Tracking,” in 2016 IEEE International Conference on Image Processing (ICIP), Sept. 2016, pp. 3464-3468, doi: 10.1109/ICIP.2016.7533003.
R. E. Kalman, “A New Approach to Linear Filtering and Prediction Problems,” J. Basic Eng., vol. 82, no. 1, pp. 35-45, Mar. 1960, doi: 10.1115/1.3662552.
H. W. Kuhn, “The Hungarian method for the assignment problem,” Nav. Res. Logist. Q., vol. 2, no. 1-2, pp. 83-97, 1955, doi: 10.1002/nav.3800020109.
K. Bernardin and R. Stiefelhagen, “Evaluating Multiple Object Tracking Performance: The CLEAR MOT Metrics,” EURASIP J. Image Video Process., vol. 2008, no. 1, May 2008, doi: 10.1155/2008/246309.
