I made a Test to see how the MetaHuman Animator handles coverage of the face during a recording when processing/solving the facial animation.
I’m working on a project which involves Sign Language and there will be moments where coverage of the face will be unavoidable. Obviously this will create errors in the animation.
My Question…
does anyone have a work-around to mitigate the errors this creates? (except completly recreating these parts via keyframe animating by hand)
… or if someone from the Unreal Team finds this Topic …
are there any plans for features to (for example) mark certain frameranges to be ignored or treated differently during processing?
I included this .gif of my Test to give an example of the problem I’m talking about.
I hope there are more accessibility options coming in the future.
Until then, since your hand and arm animations will have to be captured by some other method anyway, you might try recording them at a different time than your face and then marrying them up on the sequencer.
That’s actually an interessting thought, thanks for sharing it!
Might prove difficult because facial expressions are a key-part of sign language which I also didn’t know before learning about it, but I’ll keep it in mind going forward
Yes we are looking into this! We plan to deliver handling of this type of scenario, where an actor covers their face for some frames, with the next major MetaHuman plugin release. This will be roughly around the time of UE5.4.
