Fooling Detection Alone is Not Enough: First Adversarial Attack against Multiple Object Tracking

TL;DR

This paper introduces a novel adversarial attack method called tracker hijacking that effectively fools multiple object tracking in autonomous driving, achieving nearly 100% success with minimal frame attacks, highlighting safety concerns.

Contribution

It is the first to study adversarial attacks on the entire visual perception pipeline in autonomous driving, specifically targeting multiple object tracking with a new technique.

Findings

Tracker hijacking achieves nearly 100% success with 3 attacked frames.

Existing attacks require over 98% detection success to affect tracking.

Attacking a single frame can cause significant safety hazards.

Abstract

Recent work in adversarial machine learning started to focus on the visual perception in autonomous driving and studied Adversarial Examples (AEs) for object detection models. However, in such visual perception pipeline the detected objects must also be tracked, in a process called Multiple Object Tracking (MOT), to build the moving trajectories of surrounding obstacles. Since MOT is designed to be robust against errors in object detection, it poses a general challenge to existing attack techniques that blindly target objection detection: we find that a success rate of over 98% is needed for them to actually affect the tracking results, a requirement that no existing attack technique can satisfy. In this paper, we are the first to study adversarial machine learning attacks against the complete visual perception pipeline in autonomous driving, and discover a novel attack technique, tracker hijacking, that can effectively fool MOT using AEs on object detection. Using our technique, successful AEs on as few as one single frame can move an existing object in to or out of the headway of an autonomous vehicle to cause potential safety hazards. We perform evaluation using the Berkeley Deep Drive dataset and find that on average when 3 frames are attacked, our attack can have a nearly 100% success rate while attacks that blindly target object detection only have up to 25%.