ALA：自然度感知的对抗性亮度攻击
ALA: Naturalness-aware Adversarial Lightness Attack
AI总结：
提出一种白盒无限制对抗攻击ALA，通过修改图像亮度并加入自然度感知正则化，在保持形状和颜色的同时实现高攻击成功率，并在ImageNet和Places-365上验证有效性。
摘要：
大多数研究者试图通过揭示和修复DNN的脆弱性来增强其鲁棒性，他们使用专门的对抗样本。部分攻击样本具有受Lp范数限制的不可感知扰动。然而，由于这些扰动具有高频特性，对抗样本可以通过去噪方法进行防御，并且难以在物理世界中实现。为了避免这些缺陷，一些工作提出了无限制攻击以获得更好的鲁棒性和实用性。令人失望的是，这些样本通常看起来不自然，并且可能引起守卫的警觉。在本文中，我们提出了对抗性亮度攻击（ALA），这是一种白盒无限制对抗攻击，专注于修改图像的亮度。样本的形状和颜色（对人类感知至关重要）几乎不受影响。为了获得高攻击成功率的对抗样本，我们提出了关于图像中明暗关系的无约束增强。为了增强图像的自然度，我们根据光的范围和分布设计了自然度感知的正则化。ALA的有效性在两个流行的数据集上针对不同任务（即用于图像分类的ImageNet和用于场景识别的Places-365）得到了验证。
英文摘要：
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).
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
