ALA: Naturalness-aware Adversarial Lightness Attack

TL;DR

This paper introduces ALA, a naturalness-aware adversarial attack modifying image lightness to fool models while maintaining perceptual naturalness, improving robustness and practicality over traditional methods.

Contribution

The paper proposes a novel lightness-based adversarial attack with naturalness-aware regularization, enhancing attack success and realism compared to existing methods.

Findings

Effective on ImageNet and Places-365 datasets

Maintains natural appearance of adversarial images

Achieves high attack success rate

Abstract

Most researchers have tried to enhance the robustness of DNNs by revealing and repairing the vulnerability of DNNs with specialized adversarial examples. Parts of the attack examples have imperceptible perturbations restricted by Lp norm. However, due to their high-frequency property, the adversarial examples can be defended by denoising methods and are hard to realize in the physical world. To avoid the defects, some works have proposed unrestricted attacks to gain better robustness and practicality. It is disappointing that these examples usually look unnatural and can alert the guards. In this paper, we propose Adversarial Lightness Attack (ALA), a white-box unrestricted adversarial attack that focuses on modifying the lightness of the images. The shape and color of the samples, which are crucial to human perception, are barely influenced. To obtain adversarial examples with a high attack success rate, we propose unconstrained enhancement in terms of the light and shade relationship in images. To enhance the naturalness of images, we craft the naturalness-aware regularization according to the range and distribution of light. The effectiveness of ALA is verified on two popular datasets for different tasks (i.e., ImageNet for image classification and Places-365 for scene recognition).