Contribution of skin color cue in face detection applications

Abstract

Face detection has been considered as one of the most active areas of research due to its wide range of applications in computer vision and digital image processing technology. In order to build a robust face detection system, several cues, such as motion, shape, color, and texture have been considered. Among available cues, color is one of the most effective ones due to its computational efficiency, high discriminative power, as well as robustness against geometrical transform. This chapter investigates the role of skin color cue in automatic face detection systems. General overview of existing face detection techniques and skin pixel classification solutions are provided. Further, illumination adaptation strategies for skin color detection are discussed to overcome the sensitivity of skin color analysis against illumination variation. Finally, two case studies are presented to provide more realistic view of contribution of skin color cue in face detection frameworks.