Journal of Electrical and Computer Engineering Innovations (JECEI)
مقاله 8 ، دوره 14، شماره 2 ، مهر 2026، صفحه 391-402 1.58 M )
نوع مقاله: Original Research Paper
شناسه دیجیتال (DOI): 10.22061/jecei.2025.12094.853 نویسندگان
Morteza Akbari ؛ Seyyed Mohammad Razavi ؛ Sajad Mohamadzadeh*
تاریخ دریافت : 10 تیر 1404 ،
تاریخ بازنگری : 19 مهر 1404 ،
تاریخ پذیرش : 07 آبان 1404
چکیده Background and Objectives: Multi-object tracking in dense, multi-camera environments remains challenging due to occlusions, lighting variations, and fragmented trajectories. While existing methods rely on hierarchical two-step approaches or complex Bayesian filters, they often fail to fully exploit spatio-temporal correlations or to approach global consistency across cameras and frames. This study aims to address these limitations by proposing a novel graph-based deep learning model for continuous person tracking that independently optimizes spatial and temporal associations.Methods: The proposed model decomposes multi-camera tracking into two tasks: temporal association (linking objects across frames using velocity and time) and spatial association (aligning objects from multiple viewpoints). A spatio-temporal graph structure is constructed, with nodes representing detected objects and edges encoding relationships. Message Passing Networks (MPNs) iteratively update node and edge features, while a graph consensus fusion module merges spatial and temporal graphs for robust tracking. The model is trained using Focal Loss and evaluated on the Wildtrack and CAMPUS datasets.Results: The model achieves state-of-the-art performance, with a MOTA score of 85.5% on Wildtrack and 77.4–87.4% on CAMPUS subsets. Key improvements include a 100% MT (mostly tracked) rate and 0% ML (mostly lost) rate on CAMPUS, demonstrating exceptional robustness in occluded and crowded scenes. The IDF1 score (87.2%) highlights superior identity preservation. The decoupled design reduces graph size, which improves scalability.Conclusion: By decoupling spatial and temporal associations and leveraging graph-based optimization, the proposed model significantly enhances tracking accuracy and reliability in multi-camera settings. This work provides a framework for applications like surveillance and autonomous systems, with future potential for attention mechanisms and adaptive graph integration.کلیدواژهها
Person tracking ؛ Multi-camera environment ؛ Deep learning ؛ Spatio-temporal features ؛ graph neural networks
مراجع
[1] P. Chu, J. Wang, Q. You, H. Ling, Z. Liu, "Transmot: Spatial-temporal graph transformer for multiple object tracking," in Proc. the IEEE/CVF Winter Conference on applications of computer vision: 4870-4880, 2023.
[2] Y. Zhang, P. Sun, Y. Jiang, D. Yu, F. Weng, Z. Yuan, X. Wang, "Bytetrack: Multi-object tracking by associating every detection box," in Proc. European conference on computer vision: 1-121, 2022.
[3] D. M. Nguyen, R. Henschel, B. Rosenhahn, D. Sonntag, P. Swoboda, "LMGP: Lifted multicut meets geometry projections for multi-camera multi-object tracking," in Proc. the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 8866-8875, 2022.
[4] K. G. Quach, P. Nguyen, H. Le, T. D. Truong, C. N. Duong, M. T. Tran, K. Luu, "DyGLIP: A dynamic graph model with link prediction for accurate multi-camera multiple object tracking," in Proc. the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR): 13784-13793, 2021.
[5] G. Braso, L. Leal-Taixé, "Learning a neural solver for multiple object tracking," in Proc. the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020.
[6] Y. Xu, X. Liu, Y. Liu, S. C. Zhu, "Multi-view people tracking via hierarchical trajectory composition," in Proc. the IEEE Conference on Computer Vision and Pattern Recognition (CVPR): 4256–4265, 2016.
[7] H. Wang, Z. Li, Y. Li, K. Nai, M. Wen, "Sture: Spatial–temporal mutual representation learning for robust data association in online multi-object tracking," Comput. Vision Image Understanding, 220: 103433, 2022.
[8] Q. Xie, W. Zhou, G. L. Qi, Q. Tian, H. Li, "Progressive unsupervised person re-identification by tracklet association with spatio-temporal regularization," IEEE Trans. Multimedia, 23: 597-610, 2020.
[9] J. Xu, Y. Cao, Z. Zhang, H. Hu, "Spatial-temporal relation networks for multi-object tracking," in Proc. the IEEE/CVF International Conference on Computer Vision: 3988-3998, 2019.
[10] W. Chen, L. Cao, X. Chen, K. Huang, "A novel solution for multi-camera object tracking," in Proc. 2014 IEEE International Conference on Image Processing (ICIP): 2329-2333, 2014.
[11] J. Cao, X. Weng, R. Khirodkar, J. Pang, K. Kitani, "Observation-centric SORT: Rethinking SORT for robust multi-object tracking," in Proc. the IEEE/CVF Conference on Computer Vision and Pattern Recognition: 9686-9696, 2023.
[12] P. Chu, J. Wang, Q. You, H. Ling, Z. Liu, "TransMOT: Spatial-temporal graph transformer for multiple object tracking," in Proc. the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV): 4870-4880, 2023.
[13] Y. Zhang, P. Sun, Y. Jiang, D. Yu, F. Weng, Z. Yuan, X. Wang, "Bytetrack: Multi-object tracking by associating every detection box," in Proc. European conference on computer vision: 1-21, 2022.
[14] S. Sun, N. Akhtar, X. Song, H. Song, A. Mian, M. Shah, "Simultaneous detection and tracking with motion modelling for multiple object tracking," in Proc. the European Conference on Computer Vision (ECCV): 626-643, 2020.
[15] Z. Wang, L. Zheng, Y. Liu, Y. Li, S. Wang, "Towards real-time multi-object tracking," in Proc. the European Conference on Computer Vision (ECCV): 107-122, 2020.
[16] Y. Xu, A. Ošep,Y. Ban, R. Horaud, L. Leal-Taixé, X. Alameda-Pineda, "How to train your deep multi-object tracker," in Proc. the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020.
[17] X. Zhou, V. Koltun, P. Krähenbühl, "Tracking objects as points," in Proc. the European Conference on Computer Vision (ECCV): 474-490, 2020.
[18] Q. You, H. Jiang, "Real-time 3D deep multi-camera tracking," arXiv preprint arXiv:2003.11753, 2020.
[19] J. Ong, B. T. Vo, B. N., Vo, D. Y. Kim, S. Nordholm, "A Bayesian filter for multi-view 3D multi-object tracking with occlusion handling," IEEE Trans.Pattern Anal. Mach. Intell. (TPAMI), 44(5): 2246-2263, 2020.
[20] T. Chavdarova, P. Baqué, S. Bouquet, A. Maksai, C. Jose, T. Bagautdinov, F. Fleuret, "Wildtrack: A multi-camera HD dataset for dense unscripted pedestrian detection," in Proc. the IEEE Conference on Computer Vision and Pattern Recognition (CVPR): 5030–5039, 2018.
[21] Y. He, X. Wei, X. Hong, W. Shi, Y. Gong, "Multi-target multi-camera tracking by tracklet-to-target assignment," IEEE Trans. Image Process. (TIP), 29: 5191-5205, 2020.
[22] Y. Xu, X. Liu, L. Qin, S. C. Zhu, "Cross-view people tracking by scene-centered spatio-temporal parsing," in Proc. the AAAI Conference on Artificial Intelligence (AAAI), 2017.
[23] A. Kim, G. Braso, A. Ošep, L. Leal-Taixé, "PolarMOT: How far can geometric relations take us in 3D multi-object tracking?," in Proc. the European Conference on Computer Vision (ECCV): 41-58, 2022.
[24] A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A. N. Gomez, Ł. Kaiser, I. Polosukhin, "Attention is all you need," Adv. Neural Inf. Process. Syst., 30, 2017.
[25] M. Andriluka, S. Roth, B. Schiele, "People-tracking-by-detection and people-detection-by-tracking," in Proc. the IEEE Conference on Computer Vision and Pattern Recognition (CVPR): 1-8, 2008.
[26] Q. Xie, W. Zhou, G. J. Qi, Q. Tian, H. Li, "Progressive unsupervised person re-identification by tracklet association with spatio-temporal regularization," IEEE Trans. Multimedia, 23: 597-610, 2020.
[27] E. Luna, J. C. SanMiguel, J. M. Martínez, P. Carballeira, "Graph neural networks for cross-camera data association," IEEE Trans. Circuits Syst. Video Technol., 33(2): 589-601, 2022.
[28] T. Y. Lin, P. Goyal, R. Girshick, K. He, P. Dollár, "Focal loss for dense object detection," in Proc. the IEEE International Conference on Computer Vision (ICCV): 2999-3007, 2017.
[29] M. Engilberge, W. Liu, P. Fua, "Multi-view tracking using weakly supervised human motion prediction," in Proc. the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV): 1582-1592, 2023.
[30] Y. Hou, L. Zheng, "Multiview detection with shadow transformer (and view-coherent data augmentation)," in Proc. the 29th ACM International Conference on Multimedia (MM '21), 2021.
[31] Y. Hou, L. Zheng, S. Gould, "Multiview detection with feature perspective transformation," in Proc. the European Conference on Computer Vision (ECCV), 2020.
[32] K. Bernardin, R. Stiefelhagen, "Evaluating multiple object tracking performance: The clear mot metrics," EURASIP J. Image Video Process., 2008: 1-10, 2008.
[33] E. Ristani, F. Solera, R. Zou, R. Cucchiara, C. Tomasi, "Performance measures and a data set for multi-target, multi-camera tracking," in European Conference on Computer Vision: 17-35, 2016.
[34] K. Zhou, Y. Yang, A. Cavallaro, T. Xiang, "Omni-scale feature learning for person re-identification," in Proc. the IEEE/CVF International Conference on Computer Vision: 3702-3712, 2019.
[35] D. P. Kingma, J. Ba, "Adam: A method for stochastic optimization," in Proc. International Conference on Learning Representations (ICLR), 2015.
آمار
تعداد مشاهده مقاله: 263
تعداد دریافت فایل اصل مقاله: 10
