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.
The first idea I have is producing a usable shot of Smaug breaking out of Erebor, just without the gold, from footage of the second Hobbit movie. I don't know if that's even feasible, but any guidance is appreciate
People did that already without the need of AI, you can see it in the fan edit of the Hobbit called The Book Cut.


As for doing these yourself, last time I've seen a locally usable video generator it required absurd amounts of RAM/VRAM going well past the value of 64. You have to use online generators instead, like Hailuo AI.
 
  • Informative
Reactions: Blade of Grass
People did that already without the need of AI, you can see it in the fan edit of the Hobbit called The Book Cut.


I'm very aware of what the fanedits have done. M4 (which aside from a few issues is the best Hobbit fanedit) awkwardly cuts away right as he bursts through the wall, others try to repurpose footage from another wall break from earlier, some skip it entirely and just make the dwarves react to the sound of him breaking through, and one recolors the gold to make it look not gold, to a kinda ok but not fully convincing result.

Which is why I'm toying with the idea of whether AI can replace the gold more convincingly, or generate a usable new wall break. Even if it's just a bit longer than what M4 does.

As for doing these yourself, last time I've seen a locally usable video generator it required absurd amounts of RAM/VRAM going well past the value of 64. You have to use online generators instead, like Hailuo AI.

Thanks, I'll look into that. Do you know if it can be used to modify footage, or is it just to generate from prompts from scratch?
 
  • Like
Reactions: Blade of Grass
I come back every few weeks/months, am in awe what is possible and then proceed to have absolutely no usecase whatsoever for it. I wish I had this stuff twenty years ago, when I needed graphics assets for vidya I was writing. It would have been a major motivator to actually continue.

Very handy, provided no one else has it and can make use it. If the field changed to everyone having it then you would know your work will get buried under a pile so big only a machine could locate it. That would become a significant demotivator.

"You can create ANYTHING!", oh okay uh. Leon from RE4 in a mario costume! Haha.. Uh, Leon fucking that mouse girl.. heh, hot. Okay, what now? Back to my games. There truly is no use for AI art outside porn and OCs. My local barber did some funny xmas picture of a barbershop in the woods, cause he quite literally lives in the woods and has his shop there. Obviously nobody seethed, they went "oh that's AI teehee funny" and moved on with their lives.
There is no real other use for AI than these one-off gimmicks, exactly like 3D printing. "Dooood I can make anything!", okay, what'd you make? "Well I replaced this tiny hinge I broke". Cool?

It teeters on the most slimmest of edge cases from this pov. Being the farms the first thing that comes to mind is coomer art but children's authors that make thoughtless retard level shit for the children's book genre are living in a guilded age of not paying illustrators, publishing houses, agents etc. while cranking out a new mountain of crap every 24/7 365 days a year. All they have to do is pump it up on youtube or other social media and you could be set for life.

The true uses and capabilitiy for AI has to not be public at this point. I don't think they are spending trillions upon trillions to invent anti-human robots to load AI into while scheming to divert nuclear energy into AI farms for the purpose of drawing the perfect penis onto Emma Watson's body.
 
Results so far are not promising.

A user in the Piracy Thread linked this, it has a list of AI resources as well.

I really recommend going through this and bookmarking it. It's one of the best directories for free shit I've ever seen.

https://fmhy.net/

It lists VideoFX and PixVerse as the top two video generators in a long collection of websites. You can give these a shot too, I don't have much to say as it's not in my field of interest.
 
Meet "pokemon" Sisa Iras?

Sisa Iras.jpeg

(prompt: "pokemon cards")
 
Corn

I don't have a screenshot, but I remember when Stable Diffusion first came out, someone on 4chan was incredibly pleased despite how rudimentary the original Stable Diffusion was sans LORAs, and they said something like "I have spent the last hour upscaling pictures of a pile of crabs, the future is looking bright" and then there was an obscenely high-resolution picture of a roomful of crabs they had prompted; and this is how this LORA makes me feel.
 
I've got access to the Grok 3 beta now, so let's generate more sexy cars.
Reeeaaaaally struggling a lot - to the point that I asked Grok to help me generate prompts - and then got ChatGPT to generate prompts for Grok.
But two of the pictures, while very far away from what I want, are really sexy, so I'm sharing them.
image (12).jpgimage (11).jpg
 
I've got access to the Grok 3 beta now, so let's generate more sexy cars.
Reeeaaaaally struggling a lot - to the point that I asked Grok to help me generate prompts - and then got ChatGPT to generate prompts for Grok.
But two of the pictures, while very far away from what I want, are really sexy, so I'm sharing them.
View attachment 7004545View attachment 7004546

The blue one is very Lotus-esque concept very nice! Would absolutely steal it I was making assets for a driving game.
 
Back