DeepAI AI Chat
Log In Sign Up

Intriguing properties of synthetic images: from generative adversarial networks to diffusion models

by   Riccardo Corvi, et al.

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.


page 1

page 2

page 5

page 6


On the detection of synthetic images generated by diffusion models

Over the past decade, there has been tremendous progress in creating syn...

Stabilization of generative adversarial networks via noisy scale-space

Generative adversarial networks (GAN) is a framework for generating fake...

Leveraging Frequency Analysis for Deep Fake Image Recognition

Deep neural networks can generate images that are astonishingly realisti...

Explicit Use of Fourier Spectrum in Generative Adversarial Networks

Generative Adversarial Networks have got the researchers' attention due ...

On the use of Benford's law to detect GAN-generated images

The advent of Generative Adversarial Network (GAN) architectures has giv...

Vector quantization loss analysis in VQGANs: a single-GPU ablation study for image-to-image synthesis

This study performs an ablation analysis of Vector Quantized Generative ...

Exploring the Asynchronous of the Frequency Spectra of GAN-generated Facial Images

The rapid progression of Generative Adversarial Networks (GANs) has rais...