this post was submitted on 14 Nov 2023
126 points (92.6% liked)

AI Generated Images

7174 readers
97 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
noodle@sh.itjust.works
god@sh.itjust.works
theUnlikely@sopuli.xyz
thelsim@sh.itjust.works
M0oP0o@mander.xyz
Deceptichum@quokk.au
126
A giant, muscular penguin roaring in a comic book style (lemmy.ca)
submitted 1 year ago by danielquinn@lemmy.ca to c/imageai@sh.itjust.works
8 comments fedilink hide all child comments
 

I needed something for a presentation I'm doing on advanced Linux, so I thought something like this might be appropriate.

Annoyingly, I can't seem to get Bing to generate an image that isn't square.

you are viewing a single comment's thread
view the rest of the comments
[–] tal 7 points 1 year ago* (last edited 1 year ago)

Annoyingly, I can’t seem to get Bing to generate an image that isn’t square.

My experience with Stable Diffusion -- which is trained on square images -- has been that one is generally better-off generating square images to get an initial scene, and then cropping, upscaling (something that generative AIs do well and I'd guess that Bing probably can do, though I don't know for sure), and possibly outpainting as a way of getting more pixels and the aspect ratio desired.

Non-square images have been more-prone to things like weird mutant monsters with lots of legs, though I did try a run with my current model (based on SDXL) on a non-square image and it seemed to be working all right, so I don't know if things have improved here or if I just got lucky.