Detection and Recovery of Adversarial Slow-Pose Drift in Offloaded Visual-Inertial Odometry
Fuente:
arXiv
Guardado en:
| Autores principales: | , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866909778158026752 |
|---|---|
| author | Saha, Soruya Absur, Md Nurul Debroy, Saptarshi |
| author_facet | Saha, Soruya Absur, Md Nurul Debroy, Saptarshi |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_07130 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Detection and Recovery of Adversarial Slow-Pose Drift in Offloaded Visual-Inertial Odometry Saha, Soruya Absur, Md Nurul Debroy, Saptarshi Computer Vision and Pattern Recognition Multimedia 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. |
| title | Detection and Recovery of Adversarial Slow-Pose Drift in Offloaded Visual-Inertial Odometry |
| topic | Computer Vision and Pattern Recognition Multimedia |
| url | https://arxiv.org/abs/2509.07130 |