Researchers from the Beijing Academy of Artificial Intelligence have developed OmniGen, a diffusion model designed for unified image generation. Unlike other models which require auxiliary modules, OmniGen can handle multiple image generation tasks, such as text-to-image, picture editing, and visual-conditional generation, without additional components. Its streamlined, user-friendly architecture allows for efficient knowledge transfer between tasks. The model's performance is enhanced by the X2I dataset, a large-scale, unified image production dataset.
Sort: