Paper · arXiv 2305.10855
TextDiffuser: Diffusion Models as Text Painters
Jingye Chen, Yupan Huang, Tengchao Lv, Lei Cui, Qifeng Chen, Furu Wei
TextDiffuser, a two-stage model combining Transformer and diffusion models, generates visually appealing and coherent text images using text prompts and layouts, leveraging the MARIO-10M dataset for training and MARIO-Eval for evaluation.
Abstract
Diffusion models have gained increasing attention for their impressive generation abilities but currently struggle with rendering accurate and coherent text. To address this issue, we introduce TextDiffuser, focusing on generating images with visually appealing text that is coherent with backgrounds. TextDiffuser consists of two stages: first, a Transformer model generates the layout of keywords extracted from text prompts, and then diffusion models generate images conditioned on the text prompt and the generated layout. Additionally, we contribute the first large-scale text images dataset with OCR annotations, MARIO-10M, containing 10 million image-text pairs with text recognition, detection, and character-level segmentation annotations. We further collect the MARIO-Eval benchmark to serve as a comprehensive tool for evaluating text rendering quality. Through experiments and user studies, we show that TextDiffuser is flexible and controllable to create high-quality text images using text prompts alone or together with text template images, and conduct text inpainting to reconstruct incomplete images with text. The code, model, and dataset will be available at https://aka.ms/textdiffuser.