[KT] FACS, Procedural Expressions and Lipsync.
Added 2023-05-01 21:01:41 +0000 UTC
Facial Action Coding System. This is something I wanted to work on for quite some times. But, this requires lots of modeling work to create all the required facial blend shapes for visemes and expressions.
Switching to a DAZ based model greatly alleviates this burden and I could focus on coding. Expressions are simply collections of weighted face poses and shapes.
The FACS can generate basic procedural facial animations like eyes blinking, eyes micro saccades and head random movements. This makes the creation of idle animations easier and add more variations to the characters posture.
As a result, characters look more lively. Each expressions can be easily tuned. All blend shapes weights are controlled by curves that can be scaled up or down at will.
I've also added a basic lipsync system. Characters speeches are preprocessed using Microsoft Speech API to generate timings to play the appropriate visemes (viseme are actually expressions) using the FACS.
Here is a short video showing the current result. It's still unpolished, but characters will definitely look more lively with FACS.
So, when the next build will be ready...
Honestly, I don't really know. Switching to the new engine and to DAZ models is far more time consuming than expected and there's always some unexpected events (Murphy's law) that don't help. But, I'll do my best to release a new build this month. I still have lots of animations, outfits and materials to convert, but I'll do only the necessary ones at first so that I can release a build soon.
Thanks for your support.
