Depth judgment measures and occluding surfaces in near-field augmented reality

Abstract

In this paper we describe an apparatus and experiment that measured depth judgments in augmented reality at near-field distances of 34 to 50 centimeters. The experiment compared perceptual matching, a closed-loop task for measuring depth judgments, with blind reaching, a visually open-loop task for measuring depth judgments. The experiment also studied the effect of a highly salient occluding surface appearing behind, coincident with, and in front of a virtual object. The apparatus and closed-loop matching task were based on previous work by Ellis and Menges. The experiment found maximum average depth judgment errors of 5.5 cm, and found that the blind reaching judgments were less accurate than the perceptual matching judgments. The experiment found that the presence of a highly-salient occluding surface has a complicated effect on depth judgments, but does not lead to systematically larger or smaller errors.