Head Pose Estimation from Low-Resolution Image with Hough Forest

An approach for head pose estimation has been proposed in this paper using Hough forest. The estimation of pose are generated by voting from image patches as in a Hough transform. The basic idea is that image patches which contain eyes, hair or neck can give rich information about the head position and orientation. The voting process is implemented by randomized forest which is an efficient and robust tool for classification and regression. The method is quantitatively evaluated by comparing the estimated pose to the ground truth.