Robust Head Pose Estimation using ContourletSD Transform and GLCM

Head pose estimation is an important preprocessingstep in many computer vision and pattern recognition systemssuch as face recognition. Compared to face detection andrecognition which have been wildly used in computer visionsystems, head pose estimation has fewer proposed systems andgeneric solutions. In this paper we propose a novel approach forrobust human head pose estimation using contourletSDtransform. At first we apply contourletSD transform on images,then we create feature vector by computing gray-level cooccurrencematrix (GLCM) from each contourlet sub-band.Linear discriminant analysis (LDA) is used for dimensionalityreduction of feature vector. Finally, we classify obtained featurevectors using Support Vector Machine (SVM), K-nearestNeighbor (KNN) and hierarchical decision tree (HDT) classifiers,separately. Experimental results on FERET databasedemonstrate robustness of the proposed method than previousmethods in human head pose estimation

Keywords:

Head pose estimation , ContourletSD transform GLCM , LDA , SVM , K-NN , HDT