Forum Discussion
Hyper optimization technique for Quest for facial/head/eye performance animation.. ?
I have just one question really, If I have a character with perfect facial rigging (over 70 face bones) and I animate the face (in 3ds Max) so that:
Frame 0 = Base Pose
Frame 1= Viseme AA
Frame 2= Viseme EE
etc etc
Frame 25=Smile
Frame 26=Blink
etc.
How do I turn that into anything in Unreal? Is there any way to 'capture' the relative positions of the face bones at frame 1 or 2 etc, and call that a 'preset' and put it on a slider? In 3ds max I invented a technique where I animate faces in real-time with my Wacom Pen using the Motion Capture utility with 'pose-o-matic' (which does allow me to 'capture' the 'relative positions' of a collection of bones on a slider and (it looks just like Morpher Etc). I'm able to turn my keyframed visemes/expression into usable tools, sliders, *then animate the sliders in real-time against an audio playback, using motion capture and my pen, in order to build up layers of complex facial animation (and head and eye movement ), BUT I can't find a way to emulate this powerful process in Unreal,.
I know I could go to each keyframe and output a morph target, but my whole idea was to be able to do all animation with only bones (for Oculus Quest) and in that way save the 20-50+ Morph Targets usually required,. Maybe just a couple for forehead wrinkles here and there,.. BUT with my Visemes in successive Keyframes on the Max timeline, I am uncertain how to use that in Unreal!
I don't even know what to google!! 'converting visemes in keyframes on a timeline, to sliders for realtime motion capture using Motion Controllers' ???
THANK YOU!
Robert
Frame 0 = Base Pose
Frame 1= Viseme AA
Frame 2= Viseme EE
etc etc
Frame 25=Smile
Frame 26=Blink
etc.
How do I turn that into anything in Unreal? Is there any way to 'capture' the relative positions of the face bones at frame 1 or 2 etc, and call that a 'preset' and put it on a slider? In 3ds max I invented a technique where I animate faces in real-time with my Wacom Pen using the Motion Capture utility with 'pose-o-matic' (which does allow me to 'capture' the 'relative positions' of a collection of bones on a slider and (it looks just like Morpher Etc). I'm able to turn my keyframed visemes/expression into usable tools, sliders, *then animate the sliders in real-time against an audio playback, using motion capture and my pen, in order to build up layers of complex facial animation (and head and eye movement ), BUT I can't find a way to emulate this powerful process in Unreal,.
I know I could go to each keyframe and output a morph target, but my whole idea was to be able to do all animation with only bones (for Oculus Quest) and in that way save the 20-50+ Morph Targets usually required,. Maybe just a couple for forehead wrinkles here and there,.. BUT with my Visemes in successive Keyframes on the Max timeline, I am uncertain how to use that in Unreal!
I don't even know what to google!! 'converting visemes in keyframes on a timeline, to sliders for realtime motion capture using Motion Controllers' ???
THANK YOU!
Robert
1 Reply
Replies have been turned off for this discussion
- I think you have to look for the Oculus Hand template which is included
in the Oculus distribution. Hands are animated with bones and translated
in gestures poses.
