Algorithms and System Aspects for Automatic Face Analysis

This book gives an insight in the structure and the§scientific algorithms of a system designed for the§automatic analysis of human faces. Thereby, Face§Analysis addresses the goal to extract as much§abstract information as possible from a face. The§applied methods base on a statistical modeling of the§appearance of a face mainly determined by its shape§and texture. During the analysis of a human face§within a video or a picture, the appearance model is§used to re-synthesize this face as optimal as§possible. Various optimizations and modifications on§several steps of the basic algorithm are described§herein. The task not only fights the challenge of§most accurate face analysis, but also the§difficulties of building up an integrated, fully§automatic software system which provides a high§computational efficiency plus techniques for the§extensive exploitation of modern standard hardware.§The evaluations compare the different developed§algorithms with respect to the quality of the§re-synthesized face, computational complexity, and§pattern recognition tasks, such as the determination§of e.g. the gender, age, head pose, and facial§expression of a person.