Bericht (virtuelles) MLUGS Treffen im September 2022
09. October 2022
Protokoll
Andreas: Bilder aus Sätzen generieren mit "Stable Diffusion" auf der eigenen GPU.
- pre-trained stable diffusion model: https://huggingface.co/CompVis/stable-diffusion-v1-4
- verwendetes Codebeispiel: https://gist.github.com/mfa/821926475beffe3881899247081b9c45
- Blogposts:
- https://www.paepper.com/blog/posts/how-and-why-stable-diffusion-works-for-text-to-image-generation/
- https://simonwillison.net/2022/Aug/29/stable-diffusion/
- https://www.unite.ai/how-stable-diffusion-could-develop-as-a-mainstream-consumer-product/
- cool projects using stable diffusion:
- https://github.com/jquesnelle/txt2imghd
- https://github.com/lkwq007/stablediffusion-infinity
- data stable diffusion is trained on
- https://laion.ai/blog/laion-5b/
next
- 2022-10-18