Jittering Reduction in Marker-Based Augmented Reality Systems

Augmented Reality systems have recently become widely used. This is due to the new open source libraries that have emerged for fast application development. In this paper we address one of the most relevant problems in this type of systems, oscillation in the camera pose estimates. We study the oscillation of a system developed using the ARToolkit library. We apply both average and Kalman filters to stabilize the estimates. Using filter substantially reduces oscillation, thus improving the system's usability.