Stable Diffusion, NovelAI, Machine Learning Art - AI art generation discussion and image dump

  • 🐕 I am attempting to get the site runnning as fast as possible. If you are experiencing slow page load times, please report it.
Hammy Halloween!!!
News Hams from Across the Network
0-1.jpg
0-2.jpg0-3.jpg
3-1.jpg
Alternate title "Suit and Tie Nimma"
3-2.jpg3-3.jpg3-4.jpg3-5.jpg3-6.jpg3-7.jpg3-8.jpg3-9.jpg
4-1.jpg
4-2.jpg4-3.jpg4-4.jpg4-5.jpg4-6.jpg4-7.jpg
5-2.jpg
5-1.jpg5-3.jpg
2-3.jpg
1-1.jpg1-2.jpg2-1.jpg2-2.jpg2-4.jpg2-5.jpg

Please give more news channel ideas or news events for the News Hams to cover
 
Artist appeals copyright denial for prize-winning AI-generated work
Jason Allen—a synthetic media artist whose Midjourney-generated work "Théâtre D'opéra Spatial" went viral and incited backlash after winning a state fair art competition—is not giving up his fight with the US Copyright Office.

Last fall, the Copyright Office refused to register Allen's work, claiming that almost the entire work was AI-generated and insisting that copyright registration requires more human authorship than simply plugging a prompt into Midjourney.

Allen is now appealing that decision, asking for judicial review and alleging that "the negative media attention surrounding the Work may have influenced the Copyright Office Examiner's perception and judgment." He claims that the Examiner was biased and considered "improper factors" such as the public backlash when concluding that he had "no control over how the artificial intelligence tool analyzed, interpreted, or responded to these prompts."

You probably remember seeing Théâtre D'opéra Spatial:
Théâtre_D’opéra_Spatial.jpg
They might have to change this one day:
 
I wonder if you could rig an AI to set up a V-tuber model..then you could send the news through Notebook LLM..and set up an actual Hamster News Network through pure AI.
It is possible to use additional embedding types in addition to text prompt embeds, such as game control embeddings used in this project: https://arxiv.org/abs/2408.14837

But instead of controlling a game character, there could be motion capture sensor embeddings that you could use to read the rigging of all your movements, then generate output based on that. You would need quite a powerful machine to do it in real-time though. Perhaps DLSS could help interpolate between generated images faster.
 
looking back on it it's amazing how fast A.I has advanced. the details of this photo are a mess, you can make stuff that looks 10x better now.
Yeah but if you were a judge who didn't know Midjourney was a thing, you'd think those fucked-up details were deliberate and meaningful, so it may have worked in its favour.

Actually kinda sad to think there's probably a whole category of surreal touches an artist could deliberately do which are basically off-limits in art now because it could trigger people's AI buttholes to clench.
 
Back