The Influence of the Stereo Base on Blind and Sighted Reaches in a Virtual Environment

Abstract

In virtual environments, perceived distances are frequently reported to be shorter than intended. One important parameter for spatial perception in a stereoscopic virtual environment is the stereo base-that is, the distance between the two viewing cameras. We systematically varied the stereo base relative to the interpupillary distance (IPD) and examined influences on distance and size perception. Furthermore, we tested whether an individual adjustment of the stereo base through an alignment task would reduce the errors in distance estimation. Participants performed reaching movements toward a virtual tennis ball either with closed eyes (blind reaches) or open eyes (sighted reaches). Using the participants' individual IPD, the stereo base was set to (a) the IPD, (b) proportionally smaller, (c) proportionally larger, or (d) adjusted according to the individual performance in an alignment task that was conducted beforehand. Overall, consistent with previous research, distances were underestimated. As expected, with a smaller stereo base, the virtual object was perceived as being farther away and bigger, in contrast to a larger stereo base, where the virtual object was perceived to be nearer and smaller. However, the manipulation of the stereo base influenced blind reaching estimates to a smaller extent than expected, which might be due to a combination of binocular disparity and pictorial depth cues. In sighted reaching, when visual feedback was available, presumably the use of disparity matching led to a larger effect of the stereo base. The use of an individually adjusted stereo base diminished the average underestimation but did not reduce interindividual variance. Interindividual differences were task specific and could not be explained through differences in stereo acuity or fixation disparity.

Size matters: How reaching and vergence movements are influenced by the familiar size of stereoscopically presented objects

“…Schematic representation of the experimental setup (not drawn to scale, forehead rest not shown); adapted from [19, 43, 44]. …”

A Review of Requirements and Approaches for Realistic Visual Perception in Virtual Reality

“…Few contributions from literature address the influence of the interpupillary distance (IPD) on the disparity and therefore on the distance estimation. According to Bruder et al (2012) and Renner et al (2015) distances are underestimated, the greater the IPD is set. Nevertheless they disregard the IPD decrease induced by convergence.…”

“…Also Utsumi et al (1994) show this dependency in an early publication. Renner et al (2015) investigate the influence of IPD deviations (70%, 100% and 130% of the measured IPD) on the observation of objects at a distance between 25 and 35 cm in a CAVE. In accordance with previous sources, the measured distance values are below the expected values.…”

“…In accordance with previous sources, the measured distance values are below the expected values. What is not taken into account when calculating the expected values of Renner et al (2015), however, is the reduction in IPD due to eye convergence. Position and orientation of the virtual cameras are estimated only by the position of the head.…”

An approach to reducing distance compression in audiovisual virtual environments

“…In entertainment applications veridical perception may not be critical, although it has been shown to be an important factor in preference and enjoyment [17]. However, applications such as virtual museum tours, actions involving reaching to nearby objects, and augmented remote medical surgery (telerobotic surgery) can benefit from a perception of space as close as possible to the real world [13,19,22].…”