Characterization of Mobile Augmented Reality Tasks Supported by Edge Computing Resources

• Karlla Chaves Rodrigues UFG
• Kleber Vieira Cardoso UFG
• Sand Luz Correa UFG

DOI: https://doi.org/10.5753/sbrc.2026.19355

Resumo

Mobile Augmented Reality (MAR) applications rely on computationally intensive and latency-sensitive tasks such as Simultaneous Localization and Mapping (SLAM) and object detection, which challenge mobile devices. Offloading these workloads to edge servers is promising, yet realistic workload and traffic models remain scarce, especially under heterogeneous CPU–GPU infrastructures. This paper presents an empirical characterization of computational demand, memory usage, and IP traffic generated by SLAM and YOLObased object detection using an enhanced MR-Leo prototype (eMR-Leo). Experiments are conducted under controlled conditions with USB tethering to isolate application-level traffic. Results show that SLAM is CPU-bound with limited gains from GPU acceleration, while object detection is highly parallel and benefits significantly from GPU offloading. Object detection latency is reduced from 148 ms to approximately 4 ms per frame (approximately 98% reduction), enabling real-time performance. Based on these measurements, we derive statistical workload models for both tasks, covering instruction counts, memory usage, and traffic patterns, supporting realistic simulation and performance evaluation of edge-assisted MAR systems.

Referências

3GPP (2022). Study on XR (extended reality) evaluations for NR, V17.0.0. [Online], Available at [link].

Cadena, C. et al. (2016). Past, Present, and Future of Simultaneous Localization and Mapping: Toward the Robust-Perception Age. IEEE Transactions on Robotics, 32(6):1309–1332.

Chatzopoulos, D. et al. (2017). Mobile Augmented Reality Survey: From Where We Are to Where We Go. IEEE Access, 5:6917–6950.

Chen, Z. et al. (2017). An empirical study of latency in an emerging class of edge computing applications for wearable cognitive assistance. In Proc. of the Second ACM/IEEE Symposium on Edge Computing. Association for Computing Machinery.

Espindola, G. et al. (2025). Demonstrating the Advantages of Computational Offloading of XR Services via WebAssembly. In 2025 IEEE Network Operations and Management Symposium, pages 1–3.

Hammad, N. et al. (2023). V-Light: Leveraging Edge Computing For The Design of Mobile Augmented Reality Games. In Proc. of the 18th International Conference on the Foundations of Digital Games. Association for Computing Machinery.

Milgram, P. and Kishino, F. (1994). A taxonomy of mixed reality visual displays. IEICE Trans. Inf. Syst., 77(12).

Morín, D. G. et al. (2022). Toward the Distributed Implementation of Immersive Augmented Reality Architectures on 5G Networks. IEEE Communications Magazine, 60(2):46–52.

Morín, D. G. et al. (2024). An eXtended Reality Offloading IP Traffic Dataset and Models. IEEE Transactions on Mobile Computing, 23(6):6820–6834.

Mur-Artal, R. and Tardós, J. D. (2017). ORB-SLAM2: An Open-Source SLAM System for Monocular, Stereo, and RGB-D Cameras. IEEE Transactions on Robotics, 33(5):1255–1262.

Muzzini, F. et al. (2023). Brief Announcement: Optimized GPU-accelerated Feature Extraction for ORB-SLAM Systems. In Proc. of the 35th ACM Symposium on Parallelism in Algorithms and Architectures. Association for Computing Machinery.

Pereira, N. et al. (2021). ARENA: The Augmented Reality Edge Networking Architecture. In 2021 IEEE International Symposium on Mixed and Augmented Reality (ISMAR), pages 479–488.

Redmon, J. et al. (2016). You only look once: Unified, real-time object detection. In Proc. of the IEEE conference on computer vision and pattern recognition, pages 779–788.

Siriwardhana, Y. et al. (2021). A Survey on Mobile Augmented Reality With 5G Mobile Edge Computing: Architectures, Applications, and Technical Aspects. IEEE Communications Surveys & Tutorials, 23(2):1160–1192.

Toczé, K. et al. (2020). Characterization and modeling of an edge computing mixed reality workload. Journal of Cloud Computing, 9(46).

Wang, X. et al. (2025). Empowering Edge Intelligence: A Comprehensive Survey on On-Device AI Models. ACM Comput. Surv., 57(9).

Zhang, W. et al. (2022). EdgeXAR: A 6-DoF Camera Multi-Target Interaction Framework for MAR with User-Friendly Latency Compensation. Proc. ACM Hum.-Comput. Interact., 6(EICS).

Zou, Z. et al. (2023). Object Detection in 20 Years: A Survey. Proceedings of the IEEE, 111(3):257–276.