Orientation Template Matching for Face Localization in Complex Visual Scenes

An important topic in face recognition as well as in video coding or multi-modal human machine interfaces is the automatic localization of faces or headand-shoulder regions in visual scenes. The algorithms therefore should be computationally eficient and robust against distortions like varying lighting conditions. This paper describes a novel method for segmenting frontal head and shoulder views of persons from grey level images. The segmentation is done by oriented template correlation. This matching method only depends o n edge information, especially the Orientation of the edges. W e describe the generation of a statistical model of edge orientation within a human face from a training sample. I n the matching stage we calculate the probability f o r a face at the current image position using this model. The detection capabilities of the presented algorithm are evaluated o n a large database of 1114 images each containing one or more faces.