Simple, Robust and Accurate Head-Pose Tracking Using a Single Camera

Abstract

Tracking the position and orientation of the head in real time is finding increasing application in avionics, virtual reality, augmented reality, cinematography, computer games, driver monitoring and user interfaces for the disabled. While developing a computer interface for blind computer users, we encountered the need for a robust head-pose tracking system for accurately monitoring the gaze position of the user on a virtual screen. Although many head-pose tracking systems and techniques have been developed, we found most existing systems either added considerable complexity and cost to our application or were not accurate enough for our requirements. For example, systems described in (Horprasert et al. 1996), (Kaminski et al. 2006) and (Newman et al. 2000) use feature detection and tracking to monitor the position of the eyes, nose and/or other facial features in order to determine the orientation of the head. Unfortunately these systems require considerable processing power, additional hardware or multiple cameras to detect and track the facial features in 3D space. Although monocular systems like (Horprasert et al. 1996), (Kaminski et al. 2006) and (Zhu et al. 2004) can reduce the cost of the system, they generally performed poorly in terms of accuracy when compared with stereo or multi-camera tracking systems (Newman et al. 2000). Furthermore, facial feature tracking methods introduce inaccuracies and the need for calibration or training into the system due to the inherent image processing error margins and diverse range of possible facial characteristics of different users.