The discrete surface kernel: Framework and applications

This paper presents a framework for the computation of a discrete surface kernel, defined as the set of points from which the whole surface is visible. The first part of the paper exposes the theoretical background related to the kernel computation. In this part we also demonstrate the invariance of the surface kernel to rigid geometric transformations. The second part describes two exploitations of the surface kernel concept in the computer vision area, namely spherical surface shape representation and surface registration. This framework has been tested and experimented on real 3D face surfaces.