Pose　estimation　is　one　of　key　issues　in　face　recognition　in　complex　background　and　human-computer　interaction.　In　this　paper,　we　propose　a　novel　algorithm　to　estimating　facial　pose　using　deep　learning.　We　design　a　convolutional　neural　network　with　four　convolutional　layers,　and　a　fully-connected　layer.　The　experimental　results　on　CMU-PIE　database　show　that　the　proposed　method　outperforms　previous　traditional　methods　facial　pose.