Crazy fun with AI generated images - give me your requests and I'll crank them out!

David Bowie by Vincent Van Gogh

00065.png
00066.png
00067.png
00068.png
00069.png
00070.png
00071.png
00072.png
00073.png
00074.png
 
1st verse from Dead Dogs Two by clouddead?

"From the height of the highway onramp we saw,
Two dogs, dead in a field,
Glowing on the oakland coliseum green seats wasteland,
Dogs, dogs we thought were dead,
They rose up, rose up when whistled at, their rib cage inflating like men on the beach being photographed,
A guard dog, guard dog, for what? for what?
Against overzealous pennyless athletics fanatics,
Getting into games through a hole in the fence,
For the owner of the blue tarp tent,
Pitched by a creek beneath an onramp,
In the privacy of the last three,
Skin and bony tree, devoid of leaves,
And us undeceased, and our new cds,
Dippin' on goodies, oakland"
 
Not really, just a suggestion lol.

Maybe..

Sing a song of sixpence, a pocket full of rye; Four and twenty blackbirds baked in a pie. When the pie was opened, the birds began to sing, Wasn't that a dainty dish to set before the king?
People are actually already using this to create storybook illustrations, no joke. I fear for all the art degrees that already struggled to make money--this can't be good for that future.

00099.png
00098.png
00097.png
00096.png
00095.png
 
1st verse from Dead Dogs Two by clouddead?

"From the height of the highway onramp we saw,
Two dogs, dead in a field,
Glowing on the oakland coliseum green seats wasteland,
Dogs, dogs we thought were dead,
They rose up, rose up when whistled at, their rib cage inflating like men on the beach being photographed,
A guard dog, guard dog, for what? for what?
Against overzealous pennyless athletics fanatics,
Getting into games through a hole in the fence,
For the owner of the blue tarp tent,
Pitched by a creek beneath an onramp,
In the privacy of the last three,
Skin and bony tree, devoid of leaves,
And us undeceased, and our new cds,
Dippin' on goodies, oakland"
Lol, this one got pretty weird:

00104.png
00103.png
00102.png
00101.png
00100.png
 
This is awesome!

I have 2. How about a Mesa Boogie but in a Marshall headshell?

Also, how about a tsunami, earthquake, and raging fire all at once?
 
This is what it gave us for "Line 6 Helix floor, Behringer brand" :

View attachment 514

It also gave us this... must be some kind of new product. The, uhh, Becky?

View attachment 515
It's really interesting to see how it responds to language elements (writing), which it clearly "sees" as shapes and color, and not fully-formed symbols. Similarly, how it doesn't always know to leave well enough alone and treat human faces as complete/contained objects.
 
It's really interesting to see how it responds to language elements (writing), which it clearly "sees" as shapes and color, and not fully-formed symbols. Similarly, how it doesn't always know to leave well enough alone and treat human faces as complete/contained objects.
Exactly! There is an AI engine that google is working on though (of course), called Parti, which is freakishly smart and accurate. It can deal with text, and its interpretation of prompts and the resulting images is crazy good. Once that gets unleashed on the world, it's going to become very difficult to know what is a real photo and what is AI-generated. Honestly scares me a bit, due to the implications and how bad people already are at discerning reality from deception.

The cool and unique thing about the AI engine I used in this thread (Stable Diffusion) is that it can be run entirely locally on your own computer if you have a strong enough GPU, and is free/open-source. The others like Dalle-2 are proprietary and server-based.
 
Back
Top