Detecting fake images is becoming a major goal of computer vision. This need is becoming more and more pressing with the continuous improvement of synthesis methods based on Generative Adversarial Networks (GAN), and even more with the appearance of powerful methods based on Diffusion Models (DM). Towards this end, it is important to gain insight into which image features better discriminate fake images from real ones. In this paper we report on our systematic study of a large number of image generators of different families, aimed at discovering the most forensically relevant characteristics of real and generated images. Our experiments provide a number of interesting observations and shed light on some intriguing properties of synthetic images: (1) not only the GAN models but also the DM and VQ-GAN (Vector Quantized Generative Adversarial Networks) models give rise to visible artifacts in the Fourier domain and exhibit anomalous regular patterns in the autocorrelation; (2) when the dataset used to train the model lacks sufficient variety, its biases can be transferred to the generated images; (3) synthetic and real images exhibit significant differences in the mid-high frequency signal content, observable in their radial and angular spectral power distributions.

翻译：检测虚假图像正成为计算机视觉领域的主要目标。随着基于生成对抗网络（GAN）的合成方法不断改进，尤其是基于扩散模型（DM）的强大方法出现后，这一需求变得愈发紧迫。为此，深入了解哪些图像特征能更好地区分虚假图像与真实图像至关重要。本文报告了我们针对不同家族的大量图像生成器进行的系统性研究，旨在发现真实图像与生成图像最具取证相关性的特征。我们的实验提供了一系列有趣的观察结果，并揭示了合成图像的一些引人注目的特性：（1）不仅GAN模型，DM和VQ-GAN（矢量量化生成对抗网络）模型也会在傅里叶域中产生可见的伪影，并在自相关中表现出异常规律的模式；（2）当用于训练模型的数据集缺乏足够多样性时，其偏差可被迁移至生成的图像；（3）合成图像与真实图像在中高频信号内容上存在显著差异，这些差异可从其径向和角向频谱功率分布中观察到。