sd-webui-incantations
sd-webui-incantations copied to clipboard
v0xie
Enhance Stable Diffusion image quality, prompt following, and more through multiple implementations of novel algorithms for Automatic1111 WebUI.
Very very WIP implementation of "TextCenGen: Attention-Guided Text-Centric Background Adaptation for Text-to-Image Generation" https://arxiv.org/abs/2404.11824 There are some really interesting techniques that we can use to manipulate attention.
Refactors the Save Attention Maps script for plotting cross attention maps Adds options for different attention map filtering (Per-Token Attention Maps, One-Hot Argmax maps) Each plot shows the token index,...
This fixes a shape mismatch with controlnet that at least exists with the reference_only processor. Possibly others too, but I haven't tested and this shouldn't wrongly affect anything except similar...
WARNING:incantation_base.py:Module CFG Combiner does not implement get_xyz_axis_options WARNING:scripts.incantation_base:Module CFG Combiner does not implement get_xyz_axis_options Any idea why? Thanks!
Adds a new method of combining the guidance from PAG and CFG. Derives from "High-fidelity Person-centric Subject-to-Image Synthesis": https://arxiv.org/abs/2311.10329 In the paper they are combining the guidance from two different...
Value of rho should vary based on SD1.5, SDXL, etc.
Got those errors for each step when running "Perturbed Attention Guidance". Just got this extension installed, so can't say it worked for me before... Breaks for both SD15 and SDXL...
Hello I just installed the extension and it throws an error on startup: --- *** Error loading script: incantation_base.py Traceback (most recent call last): File "F:\Projects\Stable Diffusion\stable-diffusion-webui\modules\scripts.py", line 319, in...
Automatic1111 has an option to save default UI options, but it looks like this extension doesn't support it. Would be great to add that feature. 
Hey, very nice work. I need to test this further. Any way to use this extension and the 3 modules with the API ? Thanks for your time ! Cheers...
