"however in the ai community... scant few know about this"
Image settings: i used lineart_anime at 1 weight, depth at .5 weight and openpose at 1 weight.
enable multi controlnet
select none preprocessor and drag these into the blank space.
select lineart_anime when using my lineart_anime picture, etc.
use the new no eyes openpose image if you want to create generations of characters with less than 2 eyes (jogo for example).
v2.0 is extremely consistent compared to v1.0 thanks to the new depth image.
I manually created the openpose using https://github.com/huchenlei/sd-webui-openpose-editor. I made the lineart from upscaling an anime screenshot then running it through controlnet. I made the depth one by generating a person using a realism checkpoint then running through depth_midas for the improved depth detection.
The flare effect is not in this (obviously) so try out adding that in your photo editor of choice if you need that.
Please share any generations here so I can see what everyone's up to!
Have fun!
special thanks to not4talent's video.