this post was submitted on 14 Dec 2023
76 points (80.6% liked)

AI Generated Images

7170 readers
190 users here now

Community for AI image generation. Any models are allowed. Creativity is valuable! It is recommended to post the model used for reference, but not a rule.

No explicit violence, gore, or nudity.

This is not a NSFW community although exceptions are sometimes made. Any NSFW posts must be marked as NSFW and may be removed at any moderator's discretion. Any suggestive imagery may be removed at any time.

Refer to https://lemmynsfw.com/ for any NSFW imagery.

No misconduct: Harassment, Abuse or assault, Bullying, Illegal activity, Discrimination, Racism, Trolling, Bigotry.

AI Generated Videos are allowed under the same rules. Photosensitivity warning required for any flashing videos.

To embed images type:

“![](put image url in here)”

Follow all sh.itjust.works rules.


Community Challenge Past Entries

Related communities:

founded 1 year ago
MODERATORS
 
you are viewing a single comment's thread
view the rest of the comments
[–] tal@lemmy.today 2 points 11 months ago* (last edited 11 months ago) (3 children)

The first entry in the previously-submitted set of images by @M0oP0o@mander.xyz that I linked to above does do that:

On Stable Diffusion, one can normally just cut the prompt weight a bit on the element that's portrayed in a different style (here, Sailor Moon). I know from a previous conversation that Midjourney's prompt weighting system apparently doesn't work like that -- it also requires one to split up elements of the image. I don't know if it can be done in DALL-E; there may be a different approach to get that effect.

[–] M0oP0o@mander.xyz 2 points 11 months ago (1 children)

This was done in DALL-E (bingilator) but there is no way to ask for it, just have to run it over and over. (hence why I have so many)

[–] tal@lemmy.today 1 points 11 months ago* (last edited 11 months ago) (1 children)

but there is no way to ask for it

considers

I bet that DALL-E users have worked up some kind of way to get that effect, because it has to have come up.

thinks

I haven't used DALL-E, but what about appending "in the style of the scream by edvard munch" or something, so that there are more references to the painting in the sentence? I'd assume that if that has any effect, it'd also affect Sailor Moon.

googles

This is a book about DALL-E 2 prompt techniques:

https://dallery.gallery/wp-content/uploads/2022/07/The-DALL%C2%B7E-2-prompt-book-v1.02.pdf

If a style is proving elusive, try 'doubling down' with related terms (artists, years, media, movement) years, e.g: rather than simply '…by Picasso', try '…Cubist painting by Pablo Picasso, 1934, colourful, geometric work of Cubism, in the style of "Two Girls Reading." Or try unbundling!)

[–] M0oP0o@mander.xyz 2 points 11 months ago

These are all using things like "in the style of edvard munch's scream". Dall-e 3 (what bing uses) does like more prompts like you say, but there is also a lot of "extras" added to your prompt you can not see, meaning you can get drastically different things from the same prompt. It also means sometimes you get neat alternate sailor moons you did not ask for:

[–] roawre@feddit.ch 1 points 11 months ago

To be honest i don't know a lot about ai genaration, this one's better but there is still a difference between the main and background

[–] roawre@feddit.ch 1 points 11 months ago

And i love the "artist" you linked (talking only about sailor moon related)