Finding yourself [panoramic image sensors]

This work presents a robust camera pose algorithm and a working system to compute bounded-error estimates of the position and orientation of panoramic images captured within large, arbitrarily complex environments while moving the camera within a plant. A planning algorithm is used to place fiducials in an environment so as to satisfy a set of fiducial constraints, including the number of visible fiducials, the distance from the viewpoint to the fiducials, and the angle subtended by pairs of fiducials. The fiducial positions are subject to a set of constraints, which, in conjunction with an error model, enables pose estimation at any desired accuracy. Furthermore, results for several large indoor environments, consisting of multiple interconnected spaces are presented and analyzed.