Attack Anything: Blind DNNs via Universal Background Adversarial Attack

TL;DR

This paper introduces a novel background adversarial attack framework that effectively fools deep neural networks across diverse objects, models, and tasks in both digital and physical domains, revealing the critical impact of background variations on DNN robustness.

Contribution

The paper proposes a universal background adversarial attack method with a new ensemble strategy and smooth constraint, along with theoretical convergence analysis, advancing the understanding of DNN vulnerabilities.

Findings

Effective attacks across various objects and models

High transferability of adversarial perturbations

Background variations significantly affect DNN robustness

Abstract

It has been widely substantiated that deep neural networks (DNNs) are susceptible and vulnerable to adversarial perturbations. Existing studies mainly focus on performing attacks by corrupting targeted objects (physical attack) or images (digital attack), which is intuitively acceptable and understandable in terms of the attack's effectiveness. In contrast, our focus lies in conducting background adversarial attacks in both digital and physical domains, without causing any disruptions to the targeted objects themselves. Specifically, an effective background adversarial attack framework is proposed to attack anything, by which the attack efficacy generalizes well between diverse objects, models, and tasks. Technically, we approach the background adversarial attack as an iterative optimization problem, analogous to the process of DNN learning. Besides, we offer a theoretical demonstration of its convergence under a set of mild but sufficient conditions. To strengthen the attack efficacy and transferability, we propose a new ensemble strategy tailored for adversarial perturbations and introduce an improved smooth constraint for the seamless connection of integrated perturbations. We conduct comprehensive and rigorous experiments in both digital and physical domains across various objects, models, and tasks, demonstrating the effectiveness of attacking anything of the proposed method. The findings of this research substantiate the significant discrepancy between human and machine vision on the value of background variations, which play a far more critical role than previously recognized, necessitating a reevaluation of the robustness and reliability of DNNs. The code will be publicly available at https://github.com/JiaweiLian/Attack_Anything