Detection and Recovery of Adversarial Slow-Pose Drift in Offloaded Visual-Inertial Odometry
This addresses a security threat in edge-server VIO for VR applications, but it is incremental as it builds on existing VIO systems.
The paper tackles the problem of adversarial slow-pose drift in offloaded Visual-Inertial Odometry (VIO) for VR, presenting an unsupervised detection and recovery mechanism that reduces trajectory and pose error compared to a no-defense baseline.
Visual-Inertial Odometry (VIO) supports immersive Virtual Reality (VR) by fusing camera and Inertial Measurement Unit (IMU) data for real-time pose. However, current trend of offloading VIO to edge servers can lead server-side threat surface where subtle pose spoofing can accumulate into substantial drift, while evading heuristic checks. In this paper, we study this threat and present an unsupervised, label-free detection and recovery mechanism. The proposed model is trained on attack-free sessions to learn temporal regularities of motion to detect runtime deviations and initiate recovery to restore pose consistency. We evaluate the approach in a realistic offloaded-VIO environment using ILLIXR testbed across multiple spoofing intensities. Experimental results in terms of well-known performance metrics show substantial reductions in trajectory and pose error compared to a no-defense baseline.