论文标题
人类视觉系统和对抗性AI
The Human Visual System and Adversarial AI
论文作者
论文摘要
本文应用了有关人类视觉系统的理论，以使对抗性AI更有效。迄今为止，对抗性AI已使用LP规范对图像的清洁和对抗性示例进行了对感知距离的建模。这些规范具有简单的数学描述和近似感知距离的合理有效性的好处。但是，在过去的几十年中，图像处理的其他领域已经超越了更简单的模型（例如均方误差（MSE））朝着更好地近似人类视觉系统（HVS）的更复杂模型。我们演示了将HVS模型纳入对抗性AI的概念证明。
This paper applies theories about the Human Visual System to make Adversarial AI more effective. To date, Adversarial AI has modeled perceptual distances between clean and adversarial examples of images using Lp norms. These norms have the benefit of simple mathematical description and reasonable effectiveness in approximating perceptual distance. However, in prior decades, other areas of image processing have moved beyond simpler models like Mean Squared Error (MSE) towards more complex models that better approximate the Human Visual System (HVS). We demonstrate a proof of concept of incorporating HVS models into Adversarial AI.
