Detecting and Simulating Artifacts in GAN Fake Images
This addresses the challenge of fake image detection in security and forensics without access to the attacker's model, though it is incremental as it builds on existing artifact analysis methods.
The paper tackles the problem of detecting GAN-generated fake images when the specific GAN model is unknown, by proposing AutoGAN to simulate artifacts from common GAN pipelines and a spectrum-based classifier, achieving state-of-the-art performance on models like CycleGAN.
To detect GAN generated images, conventional supervised machine learning algorithms require collection of a number of real and fake images from the targeted GAN model. However, the specific model used by the attacker is often unavailable. To address this, we propose a GAN simulator, AutoGAN, which can simulate the artifacts produced by the common pipeline shared by several popular GAN models. Additionally, we identify a unique artifact caused by the up-sampling component included in the common GAN pipeline. We show theoretically such artifacts are manifested as replications of spectra in the frequency domain and thus propose a classifier model based on the spectrum input, rather than the pixel input. By using the simulated images to train a spectrum based classifier, even without seeing the fake images produced by the targeted GAN model during training, our approach achieves state-of-the-art performances on detecting fake images generated by popular GAN models such as CycleGAN.