Facial performance synthesis using deformation driven polynomial displacement maps

Abstract

Acquisition, modeling, compression, and synthesis of realistic facial deformations using polynomial displacement maps are described. An analysis phase can be included where the relationship between motion capture markers and detailed facial geometry is inferred. A synthesis phase can be included where detailed animated facial geometry is driven by a sparse set of motion capture markers. For analysis, an actor can be recorded wearing facial markers while performing a set of training expression clips. Real-time high-resolution facial deformations are captured, including dynamic wrinkle and pore detail, using interleaved structured light 3D scanning and photometric stereo. Next, displacements are calculated between a neutral mesh driven by the motion capture markers and the high-resolution captured expressions. These geometric displacements are stored in one or more polynomial displacement maps parameterized according to the local deformations of the motion capture dots. For synthesis, the polynomial displacement maps can be driven with new motion capture data.