Author(s)

XingTao Xu¹, JiZhe Zhang², HaiTao Zhang^1*

Affiliation(s)

¹School of Computer Science (National Pilot Software Engineering School), Beijing University of Posts and Telecommunications, Beijing 100876, China.

²Kembl Petroleum Technology Co., Ltd., Beijing 100029, China.

Corresponding Author

HaiTao Zhang

ABSTRACT

The growing demand for real-time, multi-task video analytics at the edge has encountered challenges in resource-constrained environments, including redundant computations across tasks and poor adaptability to dynamic workloads. In this paper, we propose a performance scoring-driven framework for model scaling and scheduling in edge video analytics. The framework consists of two core modules: (1) the model performance scoring module evaluates the model performance from four dimensions—video complexity, task correlation, model performance, and system resource utilization. (2) The model scaling and scheduling module then calculates a comprehensive score based on these four evaluation metrics. Aiming at maximizing the comprehensive score, this module employs the particle swarm optimization algorithm for model scheduling and system resource allocation, and selects the optimal combination of detection models based on the current model and system states. Experimental results demonstrate that our framework outperforms state-of-the-art baselines, achieving superior performance under dynamic edge workloads.

KEYWORDS

Model scaling and scheduling; Edge computing; Scoring metrics

CITE THIS PAPER

HaiTao Zhang. Performance-scoring driven model scaling and scheduling for edge video analytics serving. Journal of Computer Science and Electrical Engineering. 2025, 7(3): 80-88. DOI: https://doi.org/10.61784/jcsee3059.

REFERENCES

[1] Bogdoll D, Nitsche M, Zollner J M. Anomaly detection in autonomous driving: A survey. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2022: 4488-4499.

[2] Tay N C, Connie T, Ong T S, et al. A review of abnormal behavior detection in activities of daily living. IEEE Access, 2023, 11: 5069-5088.

[3] Zhang Q, Sun H, Wu X, et al. Edge video analytics for public safety: A review. Proceedings of the IEEE, 2019, 107(8): 1675-1696.

[4] Xu R, Razavi S, Zheng R. Edge video analytics: A survey on applications, systems and enabling techniques. IEEE Communications Surveys & Tutorials, 2023, 25(4): 2951-2982.

[5] Tolstykh I, Chernyshov M, Kuprashevich M. CerberusDet: Unified Multi-Dataset Object Detection. arXiv preprint arXiv:2407.12632, 2024.

[6] Lin J, Yang P, Wu W, et al. Learning-based query scheduling and resource allocation for low-latency mobile-edge video analytics. IEEE Internet of Things Journal, 2023, 11(3): 4872-4887.

[7] Ma H, Li R, Zhang X, et al. Reliability-aware online scheduling for DNN inference tasks in mobile-edge computing. IEEE Internet of Things Journal, 2023, 10(13): 11453-11464.

[8] Wang C, Yang P, Hou J, et al. Dependence-aware multi-task scheduling for edge video analytics with accuracy guarantee. IEEE Internet of Things Journal, 2024, 11: 26970-26983.

[9] Ahmad S, Guan H, Friedman B D, et al. Proteus: A high-throughput inference-serving system with accuracy scaling. In: Proceedings of the 29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, 2024(1): 318-334.

[10] Yang Y, Shen H, Tian H. Scheduling workflow tasks with unknown task execution time by combining machine-learning and greedy-optimization. IEEE Transactions on Services Computing, 2024, 17(3): 1181-1195.

[11] Zhou X, Liang W, Yan K, et al. Edge-enabled two-stage scheduling based on deep reinforcement learning for internet of everything. IEEE Internet of Things Journal, 2022, 10(4): 3295-3304.

[12] Kennedy J, Eberhart R. Particle swarm optimization. In: Proceedings of ICNN'95-International Conference on Neural Networks, 1995, 4: 1942-1948.

[13] Jain M, Saihjpal V, Singh N, et al. An overview of variants and advancements of PSO algorithm. Applied Sciences, 2022, 12(17): 8392.

[14] Mei X, Zhou X, Xu F, et al. Human intrusion detection in static hazardous areas at construction sites: Deep learning–based method. Journal of Construction Engineering and Management, 2023, 149(1): 04022142.

[15] Chen B, Chen C, Hu J, et al. Computer vision and machine learning-based gait pattern recognition for flat fall prediction. Sensors, 2022, 22(20): 7960.

[16] Huang T, Han Q, Min W, et al. Loitering detection based on pedestrian activity area classification. Applied Sciences, 2019, 9(9): 1866.

[17] Song Q, Wang C, Jiang Z, et al. Rethinking counting and localization in crowds: A purely point-based framework. In: Proceedings of the IEEE/CVF International Conference on Computer Vision, 2021: 3365-3374.

[18] Wojke N, Bewley A, Paulus D. Simple online and realtime tracking with a deep association metric. In: Proceedings of the IEEE International Conference on Image Processing, 2017: 3645-3649.

[19] Shao J, Zhang X, Zhang J. Task-oriented communication for edge video analytics. IEEE Transactions on Wireless Communications, 2023, 23(5): 4141-4154.

[20] Dendorfer P, Osep A, Milan A, et al. Motchallenge: A benchmark for single-camera multiple target tracking. International Journal of Computer Vision, 2021, 129: 845-881.