Using X/Y Plot in a local #stablediffusion WebUI to create contact sheets exploring the latent space for my previous post. As a photo/video person I'm trying to bring the #aiartprocess closer to the workflows I use with clients.1/7 #aiphotography #aifashion #aiart #aiartcommunity 
I imagine if you're coming to AI art from a different set of creative fields and client expectations, you'd approach the latent space from a different collection of frameworks and processes. Very curious how different a UI/UX catering to painters could look. 2/7 #aiartprocess
After an exhaustive process of prompting, I use different combos of X/Y Plot to make variations. The image I posted looked at sampler method+step count. Still trying to figure out what flow makes sense for me. I'm sure I'll throw it out as soon as new stuff gets released lol 3/7
Having an RTX3090 really helps. Right now it's not that simple to get convincing photoreal images of people that also have the right aesthetics. Prompts get me into the vicinity, looking at 4-8 images per change, 4/7 #aiartprocess
then I'll crank out 100-300 images and take a walk. With 24GB of VRAM, it takes me about 15-30 minutes at 20 steps in batches of 4 to finish. 5/7 #aiartprocess
If I'm dealing with something that's not photoreal, I can get by on looking at like 6-12 options or so, but with AI photos of people around fashion magazine quality, there's extra fingers, mangled bodies, messed up clothes, paradoxes in space-time, etc. 6/7 #aiartprocess
Still a lot to figure out! I've not even attempted in/out painting yet or Deforum. I'm looking into img2img from photos/3D renders. I want to add AI layers to my photos. I'm learning so much from the incredibly generous #aiartcommunity Let's do all the #deeplearning! 7/7 #aiart
• • •
Missing some Tweet in this thread? You can try to
force a refresh
