Want to learn how to create images like this one?
Check out our crash course in prompt engineering & AI art generation!

Luv2Grill posted about 1 month ago
160 views 0 comments

Here is another one of the spiral cut gingers I made recently after a request to make more of her. Along with using ControlNet to set the pose I also use an IP Adapter with an image cropped to her face from the original. This way you create multiple images based on the same face and features. Unfortunately it's not guaranteed to be a perfect match, but it's good enough in my opinion.

When it comes to ComfyUI I only use, except in a couple of rare occasions, custom workflows. I have built several of my own reusable workflows to satisfy different types of images. I like to experiment with various techniques, sometimes using nodes in ways that may not have been the original intent to see what the outcome would be.