Abstract :
Vision-based location systems that use fiducials, specially-designed targets placed at known positions within an environment, are an attractive way for determining the location of a participant in an augmented reality environment. There are several freely-available toolkits that calculate positioning relative to fiducials, and this paper presents a series of experiments that compare two of them, ARToolkit and VPS. The results show the level of accuracy that can be expected from any such toolkit
