A method includes detecting, by a camera of a mobile device, a face of a user of the mobile device. The method includes determining a sequence of video images of the face and applying three dimensional (3D) modeling to the sequence of video images of the face to create a 3D model of the face. The method includes identifying rotational movement of the mobile device. The method includes tracking the face to identify rotational movement of the face based on the 3D model of the face, and adjusting the tracking of the face based on the 3D model of the face and rotational movement of the mobile device.