Shadows - Silhouette | Concept LoRA IllustriousXL
Weight can go from 0.8 up to 2. I recommend 1. Works with decent amount of Styles and Character. Might not work on every checkpoint. You might need to adjust weight according to prompts. Baked in Char tend to be resistent, so use higher weight.
It's experimental, so dont expect perfect results.
Trigger Word:
colored skin, black skin, (black hair, silhouette:1.5), no eyes, faceless, no highlights
Helper Negative Prompt:
colored inner hair, hair highlight, highlight hair, glow, shiny hair, gradient hair, reflective hair, lighting on hair,
How to use LoRA's in auto1111:
Download LoRA
Copy the file to
stable-diffusion-webui/models/loraSelect your LoRA and use the triggerword
Make sure to change the weight if needed (by default it's
:1)Have Fun!
Description
FAQ
Comments (5)
Does this filter out the eyes and mouth of the character if they are included in the prompt? Particularly, I mean eye color and mouth movement prompts which are the most common for image generation.
Also out of curiosity, how well this LORA could replace the outline of character bodies with a different color? Black usually works but I'm curious. I see some keywords which if changed around may make this LORA useful for other things, say a transformation frame in a comic/story.
this does not filter out eyes and mouth, if they are prompted. Thats one of the reasons, why some loras might not work, due to different tagging and quality.
I didnt test out replacing the outline, because most of the Data was only black and that was alrdy pretty difficult.
It looks pretty impressive, though blacking out entire bodies seems overkill - when the most important context of such a thing is a black 'void face' hood interior.
Thnx just what I needed!
Or/And also Imu-sama?
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.