Waifu Diffusion is the name for this project of finetuning Stable Diffusion on images and captions downloaded through Danbooru
(Note: This project has no affiliation with Danbooru.)
Prompt: touhou 1girl komeiji_koishi portrait
All thanks goes to CompVis and Stability AI for releasing this codebase!
Model Link: https://huggingface.co/hakurei/waifu-diffusion
Stable Diffusion was made possible thanks to a collaboration with Stability AI and Runway and builds upon our previous work:
-
Our codebase for the diffusion models builds heavily on OpenAI's ADM codebase and https://github.com/lucidrains/denoising-diffusion-pytorch. Thanks for open-sourcing!
-
The implementation of the transformer encoder is from x-transformers by lucidrains.
@misc{rombach2021highresolution,
title={High-Resolution Image Synthesis with Latent Diffusion Models},
author={Robin Rombach and Andreas Blattmann and Dominik Lorenz and Patrick Esser and Björn Ommer},
year={2021},
eprint={2112.10752},
archivePrefix={arXiv},
primaryClass={cs.CV}
}