Detecting Image Attribution for Text-to-Image Diffusion Models in RGB and Beyond

arXiv:2403.19653v1 Announce Type: new
Abstract: Modern text-to-image (T2I) diffusion models can generate images with remarkable realism and creativity. These advancements have sparked research in fake image detection and attribution, yet prior studies have not fully explored the practical and scientific dimensions of this task. In addition to attributing images to 12 state-of-the-art T2I generators, we provide extensive analyses on what inference stage hyperparameters and image modifications are discernible. Our experiments reveal that initialization seeds are highly detectable, along with other …

abstract arxiv attribution beyond creativity cs.cv detection diffusion diffusion models dimensions fake generate image image detection image diffusion images modern practical prior research scientific studies text text-to-image type