Techniques are described formodeling layered facial reflectance consistingof specular reflectance, single scattering,and shallow and deep subsurface scattering.Parameters of appropriate reflectance modelscan be estimated for each of these layers,e.g., from just 20 photographs recorded ina few seconds from a single view-point.Spatially-varying specular reflectance andsingle-scattering parameters can be extractedfrom polarization-difference images underspherical and point source illumination.Direct-indirect separation can be employed todecompose the remaining multiple scatteringobserved under cross-polarization into shallowand deep scattering components to modelthe light transport through multiple layers ofskin. Appropriate diffusion models can bematched to the extracted shallow and deepscattering components for different regionson the face. The techniques were validated bycomparing renderings of subjects to referencephotographs recorded from novel viewpointsand under novel illumination conditions.Related geometry acquisition systems andsoftware products are also described.
