A simplified representation of 3D human faces adapted from 2D images

Abstract

An approach towards simplified models of human faces, given 2D images, is proposed in this paper. The tool involves the utilization of the 2D views -upon which certain protuberant points are automatically detected- and the adaptation of a generic 3D head model (polygon mesh) according to the information gained by the available views. This mesh provides shape information which -combined with texture information- efficiently describes specific 3D human faces. Issues related to luminance differences and rotation variations between the available views, are successfully dealt with within the texture map creation process. A set of localized transformations is also applied, in order to preserve the continuity of the human head surface. Besides this, the problem of using a minimum organic model representation is addressed. The aspect under which this issue is verged upon, is that of the solution of a trade-off problem between low computational complexity and high approximation quality.