3d Task Performance Using Head-coupled Stereo Displays

\Head-coupled stereo display" refers to the use of a standard graphics workstation to display stereo images of three-dimensional scenes using perspective projections de ned dynamically by the positions of the observer's eyes. The user is presented with a virtual scene located within or in front of the workstation monitor and can move his or her head around to obtain di erent views. We discuss the characteristics of head-coupled stereo display, the issues involved in implementing it correctly, and three experiments that were conducted to investigate the value of this type of display. The rst two experiments tested user performance under di erent viewing conditions. The two variables were (a) whether or not stereoscopic display was used and (b) whether or not head-coupled perspective was used. In the rst experiment, subjects were asked to subjectively rank the quality of the viewing conditions through pairwise comparisons. The results showed a strong acceptance of head-coupled stereo and a preference for headcoupling alone over stereo alone. Subjects also showed a positive response to head-coupled stereo viewing in answers to questions administered after the experiment. In the second experiment, subjects performed a task that required them to trace a path through a complex 3D tree structure. Error rates for this task showed an order of magnitude improvement with head-coupled stereo viewing compared to a static display, and the error rates achieved under head-coupling alone were signi cantly better than those obtained under stereo alone. The nal experiment examined the e ects of temporal artifacts on 3D task performance under full head-coupled stereo viewing. In particular, the e ects of reduced frame