Quality/Style LoRA based on some of Telecom Anime Film/TMS Entertainment's overseas work, sourced on some screencaps of their work on Batman The Animated Series, Animaniacs, Tiny Toons, Little Nemo and other related work (not all)
The intent is to make mock-up TMS style screencaptures.
Passive LoRA, no designated trigger word needed
Description
Pony version
FAQ
Comments (2)
This project might be a bit of a white whale for me -- I might have to try again -- and again and again and again....then again.
Perhaps I should think of this on a case-by-case basis. Like separated by series/project (i.e. Little Nemo style instead of lumping with BTAS and Animaniacs)
The training part of the loras does not combine images if you have a large number of images with one style and add another style it will still separate them. In my case I had full body of characters and cowboyshots of other characters in another style, the only way to combine loras is either merging them or make a 0.45 0.45 render of those two loras combined and train another one with the results.
If they followed different art directions it's wise to find other cartoons in a similar style to have enough training data.