Event Trojan: Asynchronous Event-based Backdoor Attacks

TL;DR

This paper introduces Event Trojan, a novel backdoor attack framework for asynchronous event data in vision tasks, demonstrating effective triggers that threaten current event-based systems and highlighting the need for improved defenses.

Contribution

The paper proposes a new backdoor attack method for event data, including immutable and mutable triggers with an adaptive learning mechanism and a stealthy loss function, advancing security research in event-based vision.

Findings

Effective backdoor triggers demonstrated on public datasets

Mutable triggers with adaptive learning maximize attack success

Stealthy triggers minimize detection while maintaining effectiveness

Abstract

As asynchronous event data is more frequently engaged in various vision tasks, the risk of backdoor attacks becomes more evident. However, research into the potential risk associated with backdoor attacks in asynchronous event data has been scarce, leaving related tasks vulnerable to potential threats. This paper has uncovered the possibility of directly poisoning event data streams by proposing Event Trojan framework, including two kinds of triggers, i.e., immutable and mutable triggers. Specifically, our two types of event triggers are based on a sequence of simulated event spikes, which can be easily incorporated into any event stream to initiate backdoor attacks. Additionally, for the mutable trigger, we design an adaptive learning mechanism to maximize its aggressiveness. To improve the stealthiness, we introduce a novel loss function that constrains the generated contents of mutable triggers, minimizing the difference between triggers and original events while maintaining effectiveness. Extensive experiments on public event datasets show the effectiveness of the proposed backdoor triggers. We hope that this paper can draw greater attention to the potential threats posed by backdoor attacks on event-based tasks. Our code is available at https://github.com/rfww/EventTrojan.