A novel stereoscopic display technique that minimizes perceptual artifacts
56.431, Tuesday, May 14, 2:45 - 6:45 pm, Orchid Ballroom
Paul Johnson1, Joohwan Kim2, Martin Banks3; 1UC Berkeley/UCSF Graduate Program in Bioengineering, University of California, Berkeley, 2Vision Science Program, University of California, Berkeley, 3School of Optometry, University of California, Berkeley
Common stereoscopic displays use either temporal interlacing or spatial interlacing to present different images to the two eyes. The ones using temporal interlacing are prone to visible flicker, unsmooth motion appearance, and depth distortions; the depth distortions are changed in perceived depth as a function of the horizontal speed of the stimulus. Those using spatial interlacingalternating pixel rows to the two eyeshave reduced resolution. We propose a novel "hybrid" display protocol that combines the better properties of temporal and spatial interlacing. In the hybrid protocol, the left- and right-eye views are interlaced spatially, but the rows corresponding to each eye alternate every frame. We performed psychophysical experiments that measured perceptual artifacts and resolution with temporal and spatial interlacing, and with the hybrid protocol. We measured motion artifacts by presenting a moving stimulus on a dark background: we found the stimulus speed above which motion artifacts were perceived. We measured depth distortion by presenting moving objects and having observers adjust their disparity until they appeared to move in the plane of the display. We determined effective spatial resolution by measuring letter acuity with stereoscopic presentation. Motion artifacts and depth distortions were reduced with the hybrid protocol compared to temporal interlacing, and effective spatial resolution was improved compared to spatial interlacing. The results suggest that the hybrid display protocol combines the best qualities of both spatial and temporal interlacing and thereby provides a more realistic perceptual experience.