Obama above has been thoroughly picasso’ed. (and is that Putin in the back?)
And the name of the star is called Wormwood: and the third part of the waters became wormwood; and many men died of the waters, because they were made bitter.
These are great.
This is sofa king cool.
I continue to blow through credits through my obsessive experimentation. I’ve developed series and series of evolutions through Artistic, Coherent, different modifiers, different weights, different orders of runs, random modifiers, specific modifiers, song lyric titles, no titles, etc. etc.
I’m not really sure what I did to achieve this one but I’m going to call this one a success.
Aside from the lady having three lips and two lenses on the left side of her glasses, that’s pretty great.
Right. I planned it that way. It’s a commentary on the duality of mind-sets in the modern paradigm of enlightened antidisestablishmentarianism as it relates to the paternal hierarchy that exists withing the Mormon church.
Duh.
…Wow, “antidisestablishmentarianism” actually almost makes sense in the context of that sentence.
Ran across this system today. From a little experimenting, it seems to totally suck.
“bill gates riding a harley davidson with an explosion in the background”
Well, it could be Bill Gates, but his face got pretty messed up in the explosion. Could be worse.
I don’t know how you guys get all these mostly coherent looking pictures. I can’t do it even using “Coherent”. I still enjoy bringing song lyrics to bulbous, weird life. This one is as close to NSFW as I’ve gotten and I didn’t even try.
Milkshake Brings All The Boys to the Yard
A couple of recent ones that I found interesting.
This one has a pretty elaborate block of fake text even though nothing in the prompt hinted for it (no use of “wanted poster” or “postcard”, for example). lt kind of reminds me of a customs label for an international package.
I like that this one spontaneously created a series of concentric circle-oids. (And the overall flattish style of the image.)
An actual art gallery served as the starting image.
Nice pictures but the benches got weird.
This is interesting. (Well, at least I think so.) I generated an image that had half the face of a cat/hyena type thing on the edge. I mirrored it to get a whole face then ran it through one of the Superresolution AIs. Both halves of the face input were identical except for being mirrored. (Possibly extremely small differences crept in due to the jpeg compression.) But if you look closely at the upscaled output, you will see small differences in how it “guessed” at the image details from one side to the other.
A few nice landscape/flora type images.
DALL-E to become more available, be ridiculously expensive.
This makes me think of an art gallery, with paintings in the back and statues in the front. (Cropped from a larger image)
Despite being on a pedestal, I kind of get the impression that the robot little girl is a museum visitor, not an exhibit.
I’m on to a new algorithm, consisting of fewer iterations, but longer more accurate runs, and I’ve been plucking artists out of Wikipedia searches as opposed to the ones listed at the site.
I came up with this:
Which, several hours later, was Retweeted by the artist himself.
Cool.