English

RAPHAEL: Text-to-Image Generation via Large Mixture of Diffusion Paths

Computer Vision and Pattern Recognition 2024-03-12 v5

Abstract

Text-to-image generation has recently witnessed remarkable achievements. We introduce a text-conditional image diffusion model, termed RAPHAEL, to generate highly artistic images, which accurately portray the text prompts, encompassing multiple nouns, adjectives, and verbs. This is achieved by stacking tens of mixture-of-experts (MoEs) layers, i.e., space-MoE and time-MoE layers, enabling billions of diffusion paths (routes) from the network input to the output. Each path intuitively functions as a "painter" for depicting a particular textual concept onto a specified image region at a diffusion timestep. Comprehensive experiments reveal that RAPHAEL outperforms recent cutting-edge models, such as Stable Diffusion, ERNIE-ViLG 2.0, DeepFloyd, and DALL-E 2, in terms of both image quality and aesthetic appeal. Firstly, RAPHAEL exhibits superior performance in switching images across diverse styles, such as Japanese comics, realism, cyberpunk, and ink illustration. Secondly, a single model with three billion parameters, trained on 1,000 A100 GPUs for two months, achieves a state-of-the-art zero-shot FID score of 6.61 on the COCO dataset. Furthermore, RAPHAEL significantly surpasses its counterparts in human evaluation on the ViLG-300 benchmark. We believe that RAPHAEL holds the potential to propel the frontiers of image generation research in both academia and industry, paving the way for future breakthroughs in this rapidly evolving field. More details can be found on a webpage: https://raphael-painter.github.io/.

Keywords

Cite

@article{arxiv.2305.18295,
  title  = {RAPHAEL: Text-to-Image Generation via Large Mixture of Diffusion Paths},
  author = {Zeyue Xue and Guanglu Song and Qiushan Guo and Boxiao Liu and Zhuofan Zong and Yu Liu and Ping Luo},
  journal= {arXiv preprint arXiv:2305.18295},
  year   = {2024}
}

Comments

NeurIPS 2023

R2 v1 2026-06-28T10:49:32.673Z