Emadさんのイラストまとめ

“A (very creepy) story in four parts”
After guided diffusion AI to create images that resonate, which I believe we have, the next step is to be able to create story boards with coherence to provide the foundation for video.
Here is an early experiment that turned out quite eerie.

Experimenting with faster AI portrait creation of different types. Think we can take some shortcuts but occasional bit of weirdness pops up

“Supersymmetry”.

I feel this is similar to some of the “AlphaGo” stuff.
Sure it isn’t Lee Sedol as a player but can play in unexpected ways.
Someone trained to paint Picasso style for decades couldn’t paint a Picasso either
Like this piece is art, but whose art? Idk
Is it creative? I’d say yes https://t.co/RTHd7PO9bd

"Exile to Wonderland" https://t.co/y8IQdWP3Ek

@Artistken1 @AliceFromNFTL Yeah, you can avoid them by using a different upscaler but sometimes like in this and the “in exile” images it’s super cool as it gives that other world effect

“Tears in exile”
#AIArtwork

“In exile”.

Totoro let himself go, needs a shave. https://t.co/MNrcSedawX

Think this may be my favourite piece of AI art by @DavidSHolz
“Orbital paths of a galactic collision”
Crazy good composition and balance, will become even better as we continue to push the boundaries of these models.