DALL•E Mini/craiyon - Description-based AI Image generator (that won't let you enter naughty things)

  • 🏰 The Fediverse is up. If you know, you know.
  • Want to keep track of this thread?
    Accounts can bookmark posts, watch threads for updates, and jump back to where you stopped reading.
    Create account
h4A6C3B9B.jpeg
 
DALL•E 2 seems to have a built-in copyright filter, learned this while watching Vinny's first stream on the program.


So if you want to have a scenario with a copyrighted character keep in mind that what will be generated will be a frame from the MAD! Cartoon show.
bandicam 2022-07-18 17-34-50-412.jpg


The funny thing about this is supposably just a week ago or so, people put in characters like Vader and Homer Simpson and they looked a lot more accurate than the current generated photos.

Maybe there's just favoritism but I'm pretty sure this isn't bought up by Disney or at least not yet.
Disney did not like this.png
 
So for whatever reason, YouTube recommended a soundtrack utilizing some music composition plugin called Delay Lama?

https://youtube.com/watch?v=uh1cb-m9JwQ
After humoring YouTube and listening to the video, I tried getting Craiyon to generate some variant of this guy in a green Santa hat. It didn't work. I did get some pretty cute results though:

View attachment 3495238 View attachment 3495239


View attachment 3495246 View attachment 3495251
Hah YT also reccomended me this, great minds think alike.
 
I posted my Astronaut Cats in general already but they belong here they are special and i love them.
astrocat.png

catnaut.png

spacecat.png
 
You could fuck with the image or use an image that isn't a person especially the face.
once you play with it for a while you can really recognize the rendering style. it's the same with dall-e 2, midjourney, waifulabs, artbreeder, etc etc
or maybe it's just me because i have fuckall to do
 
Back
Top Bottom