Generating Adversarial Events: A Motion-Aware Point Cloud Framework

TL;DR

This paper introduces MA-ADV, a novel motion-aware framework that generates adversarial events for event cameras by leveraging point cloud representations, highlighting security vulnerabilities in event-based perception systems.

Contribution

It is the first to generate adversarial events using point cloud representations and employs a diffusion-based approach for smooth, minimal perturbations considering spatial and temporal event relationships.

Findings

Achieves 100% attack success rate with minimal perturbations

Enhances robustness against existing defenses

Validates effectiveness through extensive experiments

Abstract

Event cameras have been widely adopted in safety-critical domains such as autonomous driving, robotics, and human-computer interaction. A pressing challenge arises from the vulnerability of deep neural networks to adversarial examples, which poses a significant threat to the reliability of event-based systems. Nevertheless, research into adversarial attacks on events is scarce. This is primarily due to the non-differentiable nature of mainstream event representations, which hinders the extension of gradient-based attack methods. In this paper, we propose MA-ADV, a novel \textbf{M}otion-\textbf{A}ware \textbf{Adv}ersarial framework. To the best of our knowledge, this is the first work to generate adversarial events by leveraging point cloud representations. MA-ADV accounts for high-frequency noise in events and employs a diffusion-based approach to smooth perturbations, while fully leveraging the spatial and temporal relationships among events. Finally, MA-ADV identifies the minimal-cost perturbation through a combination of sample-wise Adam optimization, iterative refinement, and binary search. Extensive experimental results validate that MA-ADV ensures a 100\% attack success rate with minimal perturbation cost, and also demonstrate enhanced robustness against defenses, underscoring the critical security challenges facing future event-based perception systems.