10000% creative, also made me laugh so very usefull.
this post was submitted on 01 Oct 2023
1367 points (97.8% liked)
Memes
45586 readers
2162 users here now
Rules:
- Be civil and nice.
- Try not to excessively repost, as a rule of thumb, wait at least 2 months to do it if you have to.
founded 5 years ago
MODERATORS
Is there a community for these type of AI memes? I'm absolutely here for them.
Conversely I'd love it if they were restricted to a specific community so I can block it. I hate it honestly.
No hate to people who like it or anything just not my thing
What kind of memes do you want?
Generally I'm not picky about memes I'm okay with a lot. I just don't like how ai has become so widespread.
And it's not even just memes. Ai voices used in videos is just as off-putting to me.
youtube keeps pushing AI Plankton from Spongebob covering songs on me, despite never having watched an AI song cover.
I heard Dale from King of the Hill singing Rich Men North of Richmond the other day.
I’m all in now.
What a creative way to refresh the meme. Great job!
any link to the site you using
Would like to know that as well. I just stole the meme from non-fediverse meme site
It’s probably Stable diffusion. I use comfyui since you can watch the sausage get made but there’s also other UIs like automatic1111. Originally for a qr pattern beautifier, there is a controlnet that takes a two tone black and white “guide” image. but you can guide it to follow any image you feed it. Such as a meme edited to be black and white, or text like “GAY SEX.”
Looks like somebody created an outline mask and then used that in img2img with a prompt for the particular scenery.
I remember seeing somebody use that technique to generate the same pose and model but different colored outfits with that technique
Maybe Canny nap like used here
I've used InvokeAI (stable diffusion models) with ControlNets for it too, it's a bit easier to use than comfy/a1111 but not as powerful IMHO.
There's also SD.Next with some enhancements to automatic1111.
This is done by combining a Diffusion model with ControlNet interface. As long as you have a decently modern Nvidia GPU and familiarity with Python and Pytorch it's relatively simple to create your own model.
The ControlNet paper is here: https://arxiv.org/pdf/2302.05543.pdf
I implemented this paper back in March. It's as simple as it is brilliant. By using methods originally intended to adapt large pre-trained language models to a specific application, the author's created a new model architecture that can better control the output of a diffusion model.
Check this out:
https://github.com/camenduru/controlnet-colab
I started looking after you responded. Havnt gotten it to work yet.
This is amazing.
Hello there fellow scholar.
Papers: held
AI memes. Yes, please!
Hmmm, how about this:
Juat doesnt work if you enhance the size. Amazing.
OP should have used sand/desert landscape for this meme.
I don't like sand. It's coarse, rough and irritating
Understandable. Greens it is.
It gets everywhere.
I really really really enjoyed this, thank you whoever created it (OP?).
Just wow
This is the top quality content that I came here for
We're reaching a whole new level of deep fried
Incredible!
So, using something like a hybrid illusion generator to make a pair of masks and providing different prompts for the highs/lows of each mask? https://charliecm.github.io/hybrid-image/
Not quite, this was made with a ControlNet. A hybrid image wouldn't work as well as this does. But the underlying visual phenomena is the same.
Whoah! Thats very cool!
This shit blows my mind every time
And that was the last anyone heard of OP
It kind of hurts to look at lol.