By integrating visual media with positioning information obtained with our wearable computer, we create new opportunities for using visuals both in the field and at the workstation. The position information we store with each visual is direction, pitch, roll, location, focal length, and zoom. This information allows any system to reconstruct the frustum of the visual, and, if height data is available to reconstruct which parts of the earth are visible in the visual. This enables position based lookup and 3D mosaicing of visuals to reconstruct a 3D model.
Neill W. Campbell, Henk L. Muller, Cliff Randell