Transient Adversarial 3D Projection Attacks on Object Detection in Autonomous Driving

TL;DR

This paper introduces a novel transient 3D projection attack on object detection systems in autonomous driving, demonstrating high success rates in deceiving models like YOLOv3 and Mask R-CNN under certain conditions.

Contribution

It presents a new attack method using 3D projection with optimization, expanding the scope of adversarial attacks in autonomous driving.

Findings

Achieves up to 100% attack success rate in indoor tests

Effective under low ambient light conditions

Deceives popular object detection models

Abstract

Object detection is a crucial task in autonomous driving. While existing research has proposed various attacks on object detection, such as those using adversarial patches or stickers, the exploration of projection attacks on 3D surfaces remains largely unexplored. Compared to adversarial patches or stickers, which have fixed adversarial patterns, projection attacks allow for transient modifications to these patterns, enabling a more flexible attack. In this paper, we introduce an adversarial 3D projection attack specifically targeting object detection in autonomous driving scenarios. We frame the attack formulation as an optimization problem, utilizing a combination of color mapping and geometric transformation models. Our results demonstrate the effectiveness of the proposed attack in deceiving YOLOv3 and Mask R-CNN in physical settings. Evaluations conducted in an indoor environment show an attack success rate of up to 100% under low ambient light conditions, highlighting the potential damage of our attack in real-world driving scenarios.