@thinkyhead@marlinfirmware colors and layout are pretty cool, but the font kerning could use a little work. Love the accidental Debian logo easteregg in the eye 😁
@layers@marlinfirmware Well now I'll have to fix those rough character spacings. I'm making fun of the excesses of AI art generators, of course. But AI can do some pretty cool things with our existing art…. #stablediffusion
Journey through the stars with Yoshimi, the guardian of galaxies, as she stands watch over realms untold. This artwork captures the essence of adventure, the solitude of space, and the bravery required to explore the unknown.
I gave it "Black Circle" by Malevitch as an input.
Prompt: a black circle with a white background, by Kazimir Malevich, traditional japanese concept art, light-brown skin, moor, inspired by Shōzō Shimamoto, intrincate, by Ram Chandra Shukla, heavy pigment, nothingness, subject in the centre, gazeta, 000
My workflow now is, instead of using Ultimate SD Upscale, is to upscale and then manually inpaint details with a well-crafted prompt for that part of the image and even some manual corrections and additions. It gets much better results, flexibility and avoids the usual vertical an horizontal artefacts.
This is the GIMP file with only the inpainted bits
@BeAware ah that makes sense. I only casually use chatgpt free myself. The ai image stuff I got working locally and open source so a better fit for my budget lol
@hans Oh yeah. They have no idea what a drawing is. They're just outputting pixels. Until recently, I had not looked at what stable diffusion does. But it's quite interesting. Indeed, it can only "think" in pixels.
As you guessed, today, I was trying to see if it was possible to extract some basic geometry and basic elements of a graphical language: squares, lines, etc.
#StableDiffusion + #ChatGPT = endless fun coming up with new concept characters for #Overwatch. I generated the image with SD-XL, then asked ChatGPT to look at the image and come up with a description. Both turned out super cool, imo.