3D Stereo Reconstruction of Human Faces driven by Di erential Constraints

Conventional stereo algorithms often fail in accurately reconstructing a 3D object because the image data do not provide enough information about the geometry of the object. We propose a way to incorporate a priori information in a reconstruction process from a sequence of calibrated face images. A 3D mesh modeling the face is iteratively deformed in order to minimize an energy function. Di erential information extracted from the object shape is used to generate an adaptive mesh. We also propose to explicitly incorporate a priori constraints related to the di erential properties of the surface where the image information cannot yield an accurate shape recovery.