search

ZYM-PKU / UDiffText

UDiffText: A Unified Framework for High-quality Text Synthesis in Arbitrary Images via Character-aware Diffusion Models
MIT License
192 stars 16 forks source link

How to run text to image demo? #9

Open Sundragon1993 opened 7 months ago

Sundragon1993 commented 7 months ago

Dear authors,

Thank you so much for your great contribution to the community, could you elaborate how to run the "text to image" function as denoted in figure (c) in your paper?

ZYM-PKU commented 7 months ago

Sorry, but we do not provide the text-to-image code because it utilizes other dependencies. One can leverage the inpainting function and run text-to-image in a two-stage manner.