Head Pose Accuracy in Augmented Reality

I worked with a colleague, Tyrone Vincent, to develop a method to analyze pose accuracy in augmented reality. This is very important in demanding applications, such as for example, surgical navigation. We developed a method to derive the relative head-to-object pose accuracy, based on sensor uncertainties. The positional uncertainties can be visualized as 3D ellipsoids.

One useful benefit of having an explicit representation of uncertainty is that we can mathematically fuse sensor data from a combination of sensors, in order to improve the overall registration accuracy.

For full details, see our paper: Hoff, W. A. and T. Vincent, “Analysis of Head Pose Accuracy in Augmented Reality,” IEEE Trans. Visualization and Computer Graphics, Vol 6., No. 4, pp. 319-334, 2000. (pdf)