Member Upsell
Click to cycle through images.

ControlNet, a new extension for Stable Diffusion, enables the use of a reference image to control the output of text-to-image generation. This means instead of coloring an image and then using image-to-image transformation, it's possible to go directly from a sketch to a fully painted image and maintain the position, pose and proportions of a character.

This example shows raw outputs cherry picked from ~100 generated images.
sketch slave girl 1Parent

BB-Kiwi 2023-02-27 18:34:211 ♡
Sorry for the critique, but with her hair that short, wouldn't she still have the wrist-links?
Admin 2023-02-27 20:19:001 ♡
This is based off the earliest sketch of Alison, before the full lore was established. In the inspiration page it's implied she complies immediately rather than the slow burn from the full story.

Also, AI has a lot of difficulty with her chunky collar because it's an uncommon accessory compared to a choker. In the Disney-esque one it turned into a weird ribbon thing and I just gave up.
Your Quiet Neighbor 2023-02-28 00:07:341 ♡
Pix #2 and 6 are tied for my top-favs of this set of AI-remix variants of this rendering of Allison!!
SF 2023-02-28 05:00:071 ♡
oooh, is Alison going to develop divine powers in opposition to the demonic powers of her masters? as predicted by the AI in #8 ;)
BB-Kiwi 2023-02-28 10:21:361 ♡
I was just curious, Admin. Hope I didn't annoy you.
Banner 2023-02-28 11:04:381 ♡
In all honesty, I prefer your faces to the AI faces. The AI faces just look 'normal' your faces have more of an expressiveness to them.
AI tries to establish a 'norm'
Art is not about a 'norm' it's about emotion. It's about what our minds connect to. This is why the AI will never replace people - that it's only a tool at best - because it can't connect with us humans.