Vision systems having knowledge of position and attitude can reliably determine what scenes are being addressed by the system. Real images of the scene can then be augmented with computer generated information which relates to the known scene. A determination of position and attitude of the device identifies which scene is being addressed. A computer recalls information regarding the scene and generates imagery according to that recalled information. The computer generated imagery is then combined with an image of the scene to form an augmented image which is presented to the user aligned to the viewing direction of the device.