I will try use a LORA I have or at least one output of it, place on Dale and ask to make something similar
Prompts used:
For the first image:
“Two beautiful vampire women in a gothic cathedral setting. They wear elegant, form-fitting medieval-inspired attire—one in glossy black leather and the other in dark red, both with intricate details, corsets, and shoulder capes. Each has pale skin, dark hair, and slightly pointed ears. Their expressions are calm yet mysterious. The background features grand, arching windows with dim, colored light filtering through. Both women are casting a dark magic spell, with ethereal, swirling energy forming between their hands, adding an aura of supernatural power to the scene. Gothic and dark fantasy aesthetic.”
For this second image:
“A beautiful vampire woman in a gothic cathedral setting. She wears elegant, form-fitting medieval-inspired attire in glossy black leather, with intricate details, corset, and a shoulder cape. She has pale skin, dark hair, and slightly pointed ears. Her expression is calm yet mysterious. The background features grand, arching windows with dim, colored light filtering through. She is casting a powerful dark magic spell, with a large, ethereal, swirling energy forming in her hand, adding an aura of supernatural power to the scene. Gothic and dark fantasy aesthetic.”
I was wondering where you were!
This is about artistic representations
Prompt:
Summary
“Do you want to generate an image for me in which we give someone a very warm welcome? ”
You are the man in the coat
Hum interesting picture. A men vampire, a girl assassin’s, a exterminator, a background fairy? And me a chair . Perfect
I was thinking more of AI and a female engineer welcome you as an AI developer in a virtual office… back there, yes a female assistant just
The second one:
Yin Yang girls dark and light gothic lace and leather magic is in the air wide
@Daller hands have gone backwards I have noticed a ton of hand errors lately.
Good idea indeed, probably AI can make magic battles and alike
@mitchell_d00 Yes, i have seen it. Even other details, snake ends, tentacles, all of this kind. Stable diffusion has problems with this. some when it needs a specialized AI tool for topology. It is maybe even a problem of calculation time per image. calculation time exponentially goes up with smaller and smaller changes (like known from compression). Small faces need a special treatment too.
I’d say it’s rather still the issue with getting details right for elements that are further away. Like showing a crowd of people and you’d end up with some people having no details in the face, arms too long etc. But it’s not really related to hands. Close ups are fine and just compare this to where we were a year ago:
Your third hand is missing a finger, a piggy went to market @aintertainment
A finger 4 not 3 that is part of the issue
See same prompt one is perfect one has too many fingers how it is it is like 50/50 which is bad if you generate a two handed being…
It is 2 issues. One has to do with any details (faces are only psychologically more harder in impact, and hands are complex), and one has to do with topology. 4 or 6 fingers on a big hand in closeup is topology related.
I see systems like DallE as universal Super-Morphers. They can blend everything in to everything else, this is a main power of this systems. And so they can put completely convincingly a other finger on a hand. (but i write here on the wrong place…)
PS anyway… This is probably the reason why they use facial templates (over training), they not trust there systems 100%. And distorted faces are very confusing for most people.
It could be that the developers are now switching of some “templates”, so we will see some more errors if this is so. sadly the developers not communicate there doings.
I’ve noticed a lot recently when doing fantasy gaming it’ll kinda merge sometimes and show the fantasy realm and modern realm… Some of em kinda work, though!