this post was submitted on 15 Sep 2023
79 points (100.0% liked)
196
16488 readers
1522 users here now
Be sure to follow the rule before you head out.
Rule: You must post before you leave.
founded 1 year ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
What's the prompt/model?
I can share the text prompt (which is fairly simple) but there are four leading images that also compromise the prompt. One is a crop of my original art, one is a still from an anime I found on duck duck go, one is a photo of myself, and lastly still from Arcane. The Arcane still seems to add the beautiful shadowing and ensures that the results are non-binary since I am combining a photo of myself with a female character from the show. Without it I get similar results but far more anime looking.
EDIT: My style of Midjourney prompts are always image source heavy as it lets me steer the results significantly rather than leave it up to the AI's interpretation of my text prompt. The text portion of this prompt is something short like "cyberpunk rocker boy with cybernetic implants holding a gun". As you can see, the image leads overwrote most of the text prompt.