Abstract
The perception of distance is known to be compressed in virtual reality in comparison with the real world (Creem-Regehr, Stefanucci and Thompson, 2015). This compression has been attributed to errors in the calibration of the view frustrum and it has been shown that distance perception is more accurate when these errors are removed (Ponto, Gleicher, Radwin and Shin, 2013). We investigated the accuracy of distance perception in consumer virtual reality using a distance bisection task. Stimuli were created using Unreal Engine and presented using an HTC Vive, following a standard room scale calibration. Participants were presented with two spheres at eye height. On each trial a reference sphere was presented at a distance of between 1 and 10m, at intervals of 1m. The participants adjusted the distance of a target sphere so that it appeared to be at half the distance as the reference. Across this range distance bisection settings were accurate and there was no evidence of distance compression for distances of between 1 and 10m. This demonstrates that consumer virtual reality can be used in applications for which accurate distance perception is important.
Meeting abstract presented at VSS 2017