OMG: Occlusion-friendly Personalized Multi-concept Generation in Diffusion Models

arXiv:2403.10983v1 Announce Type: new
Abstract: Personalization is an important topic in text-to-image generation, especially the challenging multi-concept personalization. Current multi-concept methods are struggling with identity preservation, occlusion, and the harmony between foreground and background. In this work, we propose OMG, an occlusion-friendly personalized generation framework designed to seamlessly integrate multiple concepts within a single image. We propose a novel two-stage sampling solution. The first stage takes charge of layout generation and visual comprehension information collection for handling occlusions. The second …

abstract arxiv concept concepts cs.cv current diffusion diffusion models framework identity image image generation multiple personalization personalized preservation text text-to-image type work