sure, aiさんのイラストまとめ

me when i hook up my brain to the ai and we steal art together https://t.co/D7BYmwTDfK

"anger"

for day 5 of #septembAIr
@septembair

look like anyone you know? 🤔

#septembAIr2022

#stablediffusion #imagesynth

@_yogik You'd think, but the models do not link his name to his style.

1: generic character outputs from SD
2: negligible changes from CLIP-guided models in Midjourney - don't mind the black and white lady - she shows up a lot in other "unrecognizeds"

"echo"

for day 4 of #septembAIr
@septembair

#septembAIr2022

#stablediffusion #imagesynth

"treasure"

for day 3 of #septembAIr
@septembair
#septembAIr2022

#stablediffusion #imagesynth

"ethereal"

for day 2 of #septembAIr @septembair #septembAIr2022

#stablediffusion #imagesynth

@haze_long Back when diffusion models were new, orbs would frequently show up in generations unprompted.

Some examples from my work circa Dec 2021\Jan 2022

more isometrics #septembAIr #septembAIr2022 #stablediffusion

I don't expect a single one of these rendered images to have all the desirable local data, so I made a composite of all of the details pulled from different images.

My first composite draft didn't include any of the previous images. In true surea.i fashion, it was gloomy ASF. 6/

I then used this image as an "init", an image from which another image is synthesized, resulting in new images with similar compositions, but different details.

Can't remember how many times I did this, but it was a lot. Probably in the ~50x range 5/