Layered animation of captured data

This paper proposes a novel technique for building layered animation models of real articulated objects from 3D surface measurement data. Objects are scanned using a hand-held 3D sensor to acquire 3D surface measurements. A n ovel geometric fusion algorithm is presented which enables reconstruction o f a single surface model from the captured data. This algorithm overcomes t he limitations of previous approaches which cannot be used for hand-held se nsor data as they assume that measurements are on a structured planar grid. The geometric fusion introduces the normal volume of a triangle to convert individual triangles to a volumetric representation. A layered model is co nstructed to animate the reconstructed high-resolution surface. The model c onsists of 3 layers: a skeleton for animation from key-frame or motion capt ure; a low-resolution control model for real-time mesh deformation; and a h igh-resolution model to represent the captured surface detail. Initially th e skeleton model is manually placed inside the low-resolution control model and high-resolution scanned data. Automatic techniques are introduced to m ap both the control model and captured data into a single layered model. Th e high-resolution captured data is mapped onto the low-resolution control m odel using the normal volume. The resulting model enables efficient, seamle ss animation by manipulation of the skeleton while maintaining the captured high-resolution surface detail. The animation of high-resolution captured data based on a low-resolution generic model of the object opens up the pos sibility of rapid capture and animation of new objects based on libraries o f generic models.