Automatic Adaptation of Video Analytics Pipelines with Observability Support
Abstract
The expansion of urban cameras has boosted traffic monitoring, enabling real-time event detection. However, the dynamic nature of cities requires Video Analytics (VA) systems to be adaptable to unexpected changes. In Edge Computing, where resources are limited, processing efficiency is crucial. This work proposes the automatic adaptation of VA pipelines with observability support. The approach adjusts resolution, FPS, and inference models to balance accuracy and performance. A MAPE-K cycle analyzes metrics and applies realtime adjustments, optimizing CPU and GPU usage and reducing bottlenecks. Experiments show that adaptation improves system efficiency and stability.
Keywords:
Video Analytics, Pipeline, Microservice, Observability
References
Faye, Y., Faticanti, F., Jain, S., and Bronzino, F. (2024). Videojam: Self-balancing architecture for live video analytics. In 2024 IEEE/ACM Symposium on Edge Computing (SEC), pages 149–163. IEEE.
Kosinska, J., Balis, B., Konieczny, M., Malawski, M., and Zielinski, S. (2023). Towards the observability of cloud-native applications: The overview of the state-of-the-art. IEEE Access.
Mi, L., Yuan, T., Wang, W., Dai, H., Sun, L., Zheng, J., Chen, G., and Fu, X. (2024). Accelerated neural enhancement for video analytics with video quality adaptation. IEEE/ACM Transactions on Networking.
Oliveira, D., Bhering, F., Obraczka, K., Passos, D., and Albuquerque, C. (2024). Uma arquitetura para roteamento dinamico de vídeos por multicaminhos em IoT. In Anais do XLII Simposio Brasileiro de Redes de Computadores e Sistemas Distribuídos, pages 545–558, Porto Alegre, RS, Brasil. SBC.
Padmanabhan, A., Agarwal, N., Iyer, A., Ananthanarayanan, G., Shu, Y., Karianakis, N., Xu, G. H., and Netravali, R. (2023). Gemel: Model merging for memory-efficient, real-time video analytics at the edge. In 20th USENIX Symposium on Networked Systems Design and Implementation (NSDI 23), pages 973–994.
Ravindran, A. A. (2023). Internet-of-things edge computing systems for streaming video analytics: Trails behind and the paths ahead. IoT, 4(4):486–513.
Redmon, J. (2016). You only look once: Unified, real-time object detection. In CVPR Proceedings. IEEE.
Sun, L., Wang, W., Yuan, T., Mi, L., Dai, H., Liu, Y., and Fu, X. (2024). Biswift: Bandwidth orchestrator for multi-stream video analytics on edge. In IEEE INFOCOM 2024-IEEE Conference on Computer Communications, pages 1181–1190. IEEE.
Thome, M., Prestes, A., Gomes, R., and Mota, V. (2020). Um arcabouço para detecção e alerta de anomalias de mobilidade urbana em tempo real. In Anais do XXXVIII Simposio Brasileiro de Redes de Computadores e Sistemas Distribuídos, pages 784–797, Porto Alegre, RS, Brasil. SBC.
Usman, M., Ferlin, S., Brunstrom, A., and Taheri, J. (2022). A survey on observability of distributed edge & container-based microservices. IEEE Access, 10:86904–86919.
Wang, X., Shen, M., and Yang, K. (2024). On-edge high-throughput collaborative inference for real-time video analytics. IEEE Internet of Things Journal.
Wang, Y., Liu, Z., Zhao, Y., Wang, X., and Qiu, C. (2023). Enabling real-time video analytics with adaptive sampling and detection-based tracking in edge computing. In GLOBECOM 2023 - 2023 IEEE Global Communications Conference, pages 3554–3559.
Wong, M., Ramanujam, M., Balakrishnan, G., and Netravali, R. (2024). MadEye: Boosting live video analytics accuracy with adaptive camera configurations. In 21st USENIX Symposium on Networked Systems Design and Implementation (NSDI 24), pages 549–568.
Xu, R., Razavi, S., and Zheng, R. (2023). Edge video analytics: A survey on applications, systems and enabling techniques. IEEE Communications Surveys & Tutorials.
Zhang, L., Zhang, Y., Wu, X., Wang, F., Cui, L., Wang, Z., and Liu, J. (2022). Batch adaptive streaming for video analytics. In IEEE INFOCOM 2022-IEEE Conference on Computer Communications, pages 2158–2167. IEEE.
Zhang, L., Zhong, Y., Liu, J., and Cui, L. (2023). Resource and bandwidth-aware video analytics with adaptive offloading. In 2023 IEEE 20th International Conference on Mobile Ad Hoc and Smart Systems (MASS), pages 107–115. IEEE.
Zhang, Q., Sun, H., Wu, X., and Zhong, H. (2019). Edge video analytics for public safety: A review. Proceedings of the IEEE, 107(8):1675–1696.
Kosinska, J., Balis, B., Konieczny, M., Malawski, M., and Zielinski, S. (2023). Towards the observability of cloud-native applications: The overview of the state-of-the-art. IEEE Access.
Mi, L., Yuan, T., Wang, W., Dai, H., Sun, L., Zheng, J., Chen, G., and Fu, X. (2024). Accelerated neural enhancement for video analytics with video quality adaptation. IEEE/ACM Transactions on Networking.
Oliveira, D., Bhering, F., Obraczka, K., Passos, D., and Albuquerque, C. (2024). Uma arquitetura para roteamento dinamico de vídeos por multicaminhos em IoT. In Anais do XLII Simposio Brasileiro de Redes de Computadores e Sistemas Distribuídos, pages 545–558, Porto Alegre, RS, Brasil. SBC.
Padmanabhan, A., Agarwal, N., Iyer, A., Ananthanarayanan, G., Shu, Y., Karianakis, N., Xu, G. H., and Netravali, R. (2023). Gemel: Model merging for memory-efficient, real-time video analytics at the edge. In 20th USENIX Symposium on Networked Systems Design and Implementation (NSDI 23), pages 973–994.
Ravindran, A. A. (2023). Internet-of-things edge computing systems for streaming video analytics: Trails behind and the paths ahead. IoT, 4(4):486–513.
Redmon, J. (2016). You only look once: Unified, real-time object detection. In CVPR Proceedings. IEEE.
Sun, L., Wang, W., Yuan, T., Mi, L., Dai, H., Liu, Y., and Fu, X. (2024). Biswift: Bandwidth orchestrator for multi-stream video analytics on edge. In IEEE INFOCOM 2024-IEEE Conference on Computer Communications, pages 1181–1190. IEEE.
Thome, M., Prestes, A., Gomes, R., and Mota, V. (2020). Um arcabouço para detecção e alerta de anomalias de mobilidade urbana em tempo real. In Anais do XXXVIII Simposio Brasileiro de Redes de Computadores e Sistemas Distribuídos, pages 784–797, Porto Alegre, RS, Brasil. SBC.
Usman, M., Ferlin, S., Brunstrom, A., and Taheri, J. (2022). A survey on observability of distributed edge & container-based microservices. IEEE Access, 10:86904–86919.
Wang, X., Shen, M., and Yang, K. (2024). On-edge high-throughput collaborative inference for real-time video analytics. IEEE Internet of Things Journal.
Wang, Y., Liu, Z., Zhao, Y., Wang, X., and Qiu, C. (2023). Enabling real-time video analytics with adaptive sampling and detection-based tracking in edge computing. In GLOBECOM 2023 - 2023 IEEE Global Communications Conference, pages 3554–3559.
Wong, M., Ramanujam, M., Balakrishnan, G., and Netravali, R. (2024). MadEye: Boosting live video analytics accuracy with adaptive camera configurations. In 21st USENIX Symposium on Networked Systems Design and Implementation (NSDI 24), pages 549–568.
Xu, R., Razavi, S., and Zheng, R. (2023). Edge video analytics: A survey on applications, systems and enabling techniques. IEEE Communications Surveys & Tutorials.
Zhang, L., Zhang, Y., Wu, X., Wang, F., Cui, L., Wang, Z., and Liu, J. (2022). Batch adaptive streaming for video analytics. In IEEE INFOCOM 2022-IEEE Conference on Computer Communications, pages 2158–2167. IEEE.
Zhang, L., Zhong, Y., Liu, J., and Cui, L. (2023). Resource and bandwidth-aware video analytics with adaptive offloading. In 2023 IEEE 20th International Conference on Mobile Ad Hoc and Smart Systems (MASS), pages 107–115. IEEE.
Zhang, Q., Sun, H., Wu, X., and Zhong, H. (2019). Edge video analytics for public safety: A review. Proceedings of the IEEE, 107(8):1675–1696.
Published
2025-05-19
How to Cite
SANTOS, Luan I. F.; GOMES, Francisco A. de A.; BONFIM, Michel S.; MAIA, José G. R.; TRINTA, Fernando A. M.; REGO, Paulo A. L..
Automatic Adaptation of Video Analytics Pipelines with Observability Support. In: BRAZILIAN SYMPOSIUM ON COMPUTER NETWORKS AND DISTRIBUTED SYSTEMS (SBRC), 43. , 2025, Natal/RN.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2025
.
p. 854-867.
ISSN 2177-9384.
DOI: https://doi.org/10.5753/sbrc.2025.6389.
