METHODS AND SYSTEMS OF PERFORMING PERFORMANCE CAPTURE USING AN ANATOMICALLY-CONSTRAINED LOCAL MODEL

摘要

Techniques and systems are described for generating an anatomically-constrained local model and for performing performance capture using the model. The local model includes a local shape subspace and an anatomical subspace. In one example, the local shape subspace constrains local deformation of various patches that represent the geometry of a subject's face. In the same example, the anatomical subspace includes an anatomical bone structure, and can be used to constrain movement and deformation of the patches globally on the subject's face. The anatomically-constrained local face model and performance capture technique can be used to track three-dimensional faces or other parts of a subject from motion data in a high-quality manner. Local model parameters that best describe the observed motion of the subject's physical deformations (e.g., facial expressions) under the given constraints are estimated through optimization. The optimization can solve for rigid local patch motion, local patch deformation, and the rigid motion of the anatomical bones. The solution can be formulated as an energy minimization problem for each frame that is obtained for performance capture.

主权项

1. A computer-implemented method of performing facial performance tracking of a subject using an anatomically-constrained model of a face of the subject, the method comprising:
obtaining the anatomically-constrained model, the anatomically-constrained model including a combination of a local shape subspace and an anatomical subspace, the local shape subspace including deformation shapes for each patch of a plurality of patches representing a geometry of the face, wherein a deformation shape of a patch defines a deformation of the patch for an observed facial expression, and wherein the anatomical subspace includes an anatomical bone structure constraining each of the plurality of patches; obtaining motion data of the face of the subject as the subject conducts a performance; determining, for each patch using the motion data, parameters of the anatomically-constrained model that match a facial expression in the performance, the parameters including rigid local patch motion defining one or more positions of each patch on the face, local patch deformation of each patch defined by a combination of deformation components for each patch, and rigid motion of one or more underlying bones of the anatomical bone structure relative to each patch; modifying the plurality of patches using the determined parameters to match the plurality of patches to the facial expression in the performance; and combining the deformed plurality of patches into a global face mesh for the face.