Feature-based Face Detection Against Skin-color Like Backgrounds with Varying Illumination

Abstract

A three-stage scheme for real-time reliable face detection is presented. The proposed three-stage scheme is a feature-based method that is mainly based on skin color and facial features. Skin regions are obtained using a YCbCr skin-color model in the first stage. In the second stage, a face template measure is used to obtain face candidates and then a suitable face box is used to effectively remove non-face regions from the face can- didates. Finally, facial features are measured to detect faces from face candidates in the third stage. Experimental results show that the proposed method has good performance in the face detection of faces in various poses, faces in skin color-like backgrounds, faces under varying illumination, and faces of various races.