Skip to content
InfoResearchPreprint

Adversarial Tracking: Attacks on Image Sensors for Object Tracking in Autonomous Vehicles

Published
Record updated
View JSON

Summary

This paper evaluates image-specific and image-agnostic adversarial attacks on YOLOv3 and YOLOv4 object detectors paired with SORT, Deep SORT, Strong SORT and OC-SORT trackers in the CARLA simulator. Under image-specific attacks, HOTA for YOLOv3-based trackers falls from 53.45-56.19% to 14.85-17.03%, with MOTA reaching -772.60% in the worst case. A single Universal Adversarial Perturbation can also reduce tracking accuracy without per-frame perturbations.