Differences from existing technologies?

univ-esuty / noisecollage

This is an official repository for the paper, NoiseCollage, which is a revolutionary extension of text-to-image diffusion models for layout-aware image generation.

42 stars 1 forks source link

Differences from existing technologies? #1

Open chenbaiyujason opened 6 months ago

chenbaiyujason commented 6 months ago

https://github.com/hako-mikan/sd-webui-regional-prompter What's the difference between the implementation here and the previous ones for latent and cross-attention area diffusion?

Mowenyii commented 5 months ago

same question