Increasing the Tracking Efficiency of Mobile Augmented Reality Using a Hybrid Tracking Technique

Traditional vision tracking approaches have failed to support the fast tracking required by Mobile Augmented Reality (MAR) for a real-time interactive effect. The main problem addressed by this research is the tracking problem in MAR where the computational load is a major issue. In this paper we proposed a more efficient visual tracking technique based on fast detectors and small binary descriptors which are capable of providing both; scale and rotation invariance. We also propose a new hybrid tracking technique which can speed up the overall performance of MAR by reducing the detection rate of the tracking process. The hybrid technique is based on the usage of inertial sensors such as accelerometer, gyroscopes and gravitational vectors which can readily improve the efficiency of any vision based feature tracking system that uses computer vision. The preliminary tests carried out during the course of the study produced very promising results.