A Comparison of Distance Estimation in HMD-Based Virtual Environments with Different HMD-Based Conditions

Abstract

Underestimation of egocentric distances in immersive virtual environments using various head-mounted displays (HMDs) has been a puzzling topic of research interest for several years. As more commodity-level systems become available to developers, it is important to test the variation of underestimation in each system since reasons for underestimation remain elusive. In this article, we examine several different systems in two experiments and comparatively evaluate how much users underestimate distances in each one. To observe distance estimation behavior, a standard indirect blind walking task was used. An Oculus Rift DK1, weighted Oculus Rift DK1, Oculus Rift DK1 with an artificially restricted field of view, Nvis SX60, Nvis SX111, Oculus Rift DK2, Oculus Rift consumer version (CV1), and HTC Vive were tested. The weighted and restricted field of view HMDs were evaluated to determine the effect of these factors on distance underestimation; the other systems were evaluated because they are popular systems that are widely available. We found that weight and field of view restrictions heightened underestimation in the Rift DK1. Results from these conditions were comparable to the Nvis SX60 and SX111. The Oculus Rift DK1 and CV1 possessed the least amount of distance underestimation, but in general, commodity-level HMDs provided more accurate estimates of distance than the prior generation of HMDs.