Quote:
Originally Posted by Kijou
Thanks for the hint, but is it possible to make entirely new images/scenes with that? Because what you show here is basically the same picture with maybe 5% variation. What I am looking for should be able to for example make up a new specialist if it was given the right dataset. I don't mind paying if it gives me the desired result
|
It is pretty easy to do it on your own PC without paying for some shit AI.There are tons of youtube videos about Stable Diffusion or Lora etc.
I worked on that few weeks ago. Just to show you the results I took 10 photos (completely wrong because all of them with a group of characters instead of one etc) and made a model.
Took it 5 minutes and completely 0 work (as I said. if I would make it really great I would spend time to make a real good model instead of 5 minutes shit) and got results like that:
of course the faces and in total entire image is awful. The reason is that I didn't make a real good model. Later today I will have some time so I will maybe create one proper model to show you how big potential it has.
About your question
with creating new SPs etc
Not really. Probably it would be possible but I am not sure if it's worth your time.
To make a new SP (or even just a skin) you would have to create hundreds of the same texture in a different positions etc. I am not an expert in this topic but I don't think it's currently possible to do.
But still you can make a great graphic.
Maybe (just maybe. I never checked that) there are some photoshop extensions with stable diffusion which allows to load the textures and AI will recolor them/modify etc. Idk. Never checked that.