Deep neural networks for head pose classification
Yang Lü, Shujuan Yi, Nan Hou, Jingfu Zhu, Tiemin Ma · 2016
Head pose classification and estimation are essential for many face detection and recognition tasks and tracking applications. This paper proposes the robust and fast algorithms for head pose classification from labeled head pose database by employing deep neural networks (DNNs). The DNNs have capabilities to learn from raw images and process large-scale training image datasets. The proposed DNNs method has fast convergence and higher learning ability. The experimental results show that the proposed method outperforms existing methods in accuracy and robustness.