Cross-camera vehicle trajectory reconstruction in roadside surveillance networks with partial and minimal overlaps

Abstract

Cross-camera vehicle trajectory reconstruction is essential for roadside perception systems supporting traffic analysis, safety evaluation, and infrastructure monitoring. In practical deployments, roadside cameras often exhibit partial or minimal overlaps, providing limited spatiotemporal continuity and making cross-camera association challenging. This paper proposes a global trajectory reconstruction framework designed for such roadside surveillance networks. Single-camera trajectories are first extracted and projected into a unified road-plane coordinate system through calibration, enabling metric-level comparison across views. Appearance features and motion cues are then jointly exploited in a multi-stage association strategy that integrates spatiotemporal feasibility and visual similarity. Matched trajectories are stitched and refined using quality-aware alignment and short-gap compensation to generate continuous and physically plausible cross-camera trajectories. Real-world experiments with roadside cameras, RTK-equipped vehicles, and UAV-based trajectory data demonstrate that the proposed method achieves high association accuracy and reconstruction precision in both single- and multi-vehicle scenarios while maintaining low computational overhead.

Keywords

Cross-camera trajectory reconstruction / Roadside surveillance / Vehicle tracking / Partial-overlap camera networks / Smart Transportation

References

[1] Jain NK, Saini RK, Mittal P. A review on traffic monitoring system techniques. Soft computing: Theories and applications: Proceedings of SoCTA. 2017, 2018: 569-577

[2] Bewley A, Ge Z, Ott L, Ramos F, Upcroft B (eds) (2016) Simple online and realtime tracking. 2016 IEEE international conference on image processing (ICIP). IEEE

[3] Carion N, Massa F, Synnaeve G, Usunier N, Kirillov A, Zagoruyko S (eds) (2020) End-to-end object detection with transformers. European conference on computer vision. Springer

[4] Girshick R (ed) (2015) Fast r-cnn. Proceedings of the IEEE international conference on computer vision

[5] Redmon J, Divvala S, Girshick R, Farhadi A (eds) (2016) You only look once: unified, real-time object detection. Proceedings of the IEEE conference on computer vision and pattern recognition

[6] Wojke N, Bewley A, Paulus D (eds) (2017) Simple online and realtime tracking with a deep association metric. 2017 IEEE international conference on image processing (ICIP). IEEE

[7] Zhang Y, Sun P, Jiang Y, Yu D, Weng F, Yuan Z et al (eds) (2022) Bytetrack: multi-object tracking by associating every detection box. European conference on computer vision. Springer

[8] Zhao Y, Lv W, Xu S, Wei J, Wang G, Dang Q et al (eds) (2024) Detrs beat yolos on real-time object detection. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

[9] Heo J, Kwon Y. 3d vehicle trajectory extraction using dcnn in an overlapping multi-camera crossroad scene. Sensors. 2021, 21(23): 7879.

[10] Luna E, SanMiguel JC, Martínez JM, Escudero-Viñolo M. Online clustering-based multi-camera vehicle tracking in scenarios with overlapping FOVs. Multimedia Tools and Applications. 2022, 8157063-7083.

[11] Zhang H, Fang R, Li S, Miao Q, Fan X, Hu Jet al. . Multi-camera multi-vehicle tracking guided by highway overlapping fovs. Mathematics. 2024, 12(10): 1467.

[12] Castañeda JN, Jelaca V, Frías A, Pizurica A, Philips W, Cabrera RR et al (eds) (2011) Non-overlapping multi-camera detection and tracking of vehicles in tunnel surveillance. 2011 International Conference on Digital Image Computing: Techniques and Applications. IEEE

[13] Jiang W, Xiao C, Jin H, Zhu S, Lu Z (eds) (2013) Vehicle tracking with non-overlapping views for multi-camera surveillance system. 2013 ieee 10th international conference on high performance computing and communications & 2013 ieee international conference on embedded and ubiquitous computing. IEEE

[14] Qian Y, Yu L, Liu W, Hauptmann AG (eds) (2020) Electricity: an efficient multi-camera vehicle tracking system for intelligent city. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops

[15] Ren P, Lu K, Yang Y, Yang Y, Sun G, Wang W et al (eds) (2021) Multi-camera vehicle tracking system based on spatial-temporal filtering. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

[16] Wan J, Li L (eds) (2013) Distributed optimization for global data association in non-overlapping camera networks. 2013 Seventh International Conference on Distributed Smart Cameras (ICDSC). IEEE

[17] Mahdavi H (2013) Optical flow based moving object detection and tracking for traffic surveillance

[18] Meinhardt T, Kirillov A, Leal-Taixe L, Feichtenhofer C (eds) (2022) Trackformer: multi-object tracking with transformers. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

[19] Zeng F, Dong B, Zhang Y, Wang T, Zhang X, Wei Y (eds) (2022) Motr: end-to-end multiple-object tracking with transformer. European conference on computer vision. Springer

[20] Guiducci A. Camera calibration for road applications. Comput Vis Image Underst. 2000, 792250-266.

[21] Su G, Zeng Z, Song A, Zhao C, Shen F, Yuan Let al. . A general framework for reconstructing full-sample continuous vehicle trajectories using roadside sensing data. Appl Sci. 2023, 13(5): 3141.

[22] Ueshiba, Tomita (eds) (2003) Plane-based calibration algorithm for multi-camera systems via factorization of homography matrices. Proceedings Ninth IEEE International Conference on Computer Vision. IEEE

[23] Khan SD, Ullah H. A survey of advances in vision-based vehicle re-identification. Comput Vis Image Underst. 2019, 182: 50-63.

[24] Liu X, Liu W, Ma H, Fu H (eds) (2016) Large-scale vehicle re-identification in urban surveillance videos. 2016 IEEE international conference on multimedia and expo (ICME). IEEE

[25] Wu M, Qian Y, Wang C, Yang M (eds) (2021) A multi-camera vehicle tracking system based on city-scale vehicle re-id and spatial-temporal information. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

[26] Zheng Z, Ruan T, Wei Y, Yang Y, Mei T. Vehiclenet: learning robust visual representation for vehicle re-identification. IEEE Trans Multimedia. 2020, 23: 2683-2693.

[27] Piva S, Calbi A, Angiati D, Regazzoni CS (eds) (2005) A multi-feature object association framework for overlapped field of view multi-camera video surveillance systems. IEEE Conference on Advanced Video and Signal Based Surveillance, 2005. IEEE

[28] Olagoke AS, Ibrahim H, Teoh SS. Literature survey on multi-camera system and its application. IEEE Access. 2020, 8: 172892-172922.

[29] Liu C, Zhang Y, Luo H, Tang J, Chen W, Xu X et al (eds) (2021) City-scale multi-camera vehicle tracking guided by crossroad zones. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

[30] Liu Y, Zhang X, Zhang B, Zhang X, Wang S, Xu J (eds) (2022) Multi-camera vehicle tracking based on occlusion-aware and inter-vehicle information. Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition

[31] Nguyen TT, Nguyen HH, Sartipi M, Fisichella M. Multi-vehicle multi-camera tracking with graph-based tracklet features. IEEE Trans Multimedia. 2023, 26: 972-983.

[32] Hsu H-M, Huang T-W, Wang G, Cai J, Lei Z, Hwang J-N (eds) (2019) Multi-camera tracking of vehicles based on deep features re-id and trajectory-based camera link models. CVPR workshops

[33] Tang Z, Naphade M, Liu M-Y, Yang X, Birchfield S, Wang S et al (eds) (2019) Cityflow: a city-scale benchmark for multi-target multi-camera vehicle tracking and re-identification. Proceedings of the IEEE/CVF conference on computer vision and pattern recognition

[34] Chen X, Yin J, Tang K, Tian Y, Sun J. Vehicle trajectory reconstruction at signalized intersections under connected and automated vehicle environment. IEEE Trans Intell Transp Syst. 2022, 231017986-18000.

[35] Tong P, Li M, Li M, Huang J, Hua X (eds) (2021) Large-scale vehicle trajectory reconstruction with camera sensing network. Proceedings of the 27th Annual International Conference on Mobile Computing and Networking

[36] He S, Luo H, Wang P, Wang F, Li H, Jiang W (eds) (2021) Transreid: transformer-based object re-identification

[37] Liu X, Liu W, Mei T, Ma H (eds) (2016) A deep learning-based approach to progressive vehicle re-identification for urban surveillance. Springer

[38] Jocher G, Qiu J (2024) “Ultralytics YOLO11,” GitHub repository, version 11.0.0. https://github.com/ultralytics/ultralytics

Funding

Department of Transportation of Zhejiang Province(2024007)

RIGHTS & PERMISSIONS

The Author(s)