Level Up the Deepfake Detection: a Method to Effectively Discriminate Images Generated by GAN Architectures and Diffusion Models
This work addresses the deepfake detection problem for security and media integrity, but it is incremental as it builds on existing methods with a new dataset and hierarchical approach.
The paper tackled the problem of detecting and recognizing AI-generated images by introducing a hierarchical multi-level approach to distinguish real images from fakes, differentiate between GANs and diffusion models, and identify specific architectures, achieving over 97% classification accuracy in all cases.
The image deepfake detection task has been greatly addressed by the scientific community to discriminate real images from those generated by Artificial Intelligence (AI) models: a binary classification task. In this work, the deepfake detection and recognition task was investigated by collecting a dedicated dataset of pristine images and fake ones generated by 9 different Generative Adversarial Network (GAN) architectures and by 4 additional Diffusion Models (DM). A hierarchical multi-level approach was then introduced to solve three different deepfake detection and recognition tasks: (i) Real Vs AI generated; (ii) GANs Vs DMs; (iii) AI specific architecture recognition. Experimental results demonstrated, in each case, more than 97% classification accuracy, outperforming state-of-the-art methods.