The system needs support for layers for expressions like it has for body motion so that animation can be built incrementally using keyframes (Face Key) and Face Puppet (or Mo-Cap).


As it is, if you attempt to do Face Puppet "over the top of" a clip based on keyframes (from Face Key), the discrete Face Key-generated keyframe are obliterated (flattened). There should be separate tracks so they can be viewed and edited independently (while being blended/mixed when played back).