Variant Pose Face Detection Based on Multi-classifier Fusion
Jimin Liang · Jisuanji fangzhen · 2009
A face detection algorithm for variant pose faces detection in color image based on multi-classifiers fusion was proposed.A frontal face classifier and a profile face classifier were trained by the AdaBoost algorithm.The two classifiers worked in parallel and their detection results were combined to form candidate face regions.These candidate regions were further verified by a skin color model in YCbCr chrominance space.The proposed algorithm integrated information from different pose classifiers, also information of gray scale and color distribution of human face.It is robust to face pose and background variation and runs fast.Testing with a large set of color images containing one or more faces with variant pose, the result shows the efficiency and feasibility of the presented algorithm.It improves the correct face detection rate and decreases the false detection significantly.