Face reconstruction and recognition using a statistical model combining shape and spherical harmonics

We present a face reconstruction framework based on a statistical model that merges shape (2D and height maps), appearance (albedo), and spherical harmonics projection (SHP) information. The framework takes a 2D frontal face image under arbitrary illumination as input and outputs the estimated 3D shape and appearance. Face identification is performed using the shape and albedo coefficients of the fitting process, as well as the actual shape/texture reconstructions. Results show decent face reconstructions and a perfect identification rate for the frontal images of the Extended Yale Database B.