CGI-DM: Digital Copyright Authentication for Diffusion Models via Contrasting Gradient Inversion

arXiv:2403.11162v1 Announce Type: cross
Abstract: Diffusion Models (DMs) have evolved into advanced image generation tools, especially for few-shot generation where a pretrained model is fine-tuned on a small set of images to capture a specific style or object. Despite their success, concerns exist about potential copyright violations stemming from the use of unauthorized data in this process. In response, we present Contrasting Gradient Inversion for Diffusion Models (CGI-DM), a novel method featuring vivid visual representations for digital copyright authentication. Our …

arxiv authentication cgi copyright cs.ai cs.cr cs.cv cs.cy cs.lg diffusion diffusion models digital gradient type via