Improving Relative Depth Judgments in Augmented Reality with Auxiliary Augmentations Mikko Kytö, Aleksi Mäkinen, Jukka Häkkinen and Pirkko Oittinen School of Science, Aalto University Abstract Signiﬁcant depth judgment errors are common in augmented reality. This study presents a visualization approach for improving relative depth judgments in aug- mented reality. The approach uses auxiliary augmented objects in addition to the main augmentation to support ordinal and interval depth judgment tasks. The auxiliary augmentations are positioned spatially near real-world objects, and the location of the main augmentation can be deduced based on the rela- tive depth cues between the augmented objects. In the experimental part, the visualization approach was tested in the ”X-ray” visualization case with a video see-through system. Two relative depth cues, in addition to motion parallax, were used between graphical objects: relative size and binocular disparity. The results show that the presence of auxiliary objects signiﬁcantly reduced errors of depth judgment. Errors in judging the ordinal location with respect to a wall (front, at or behind) and judging depth intervals were reduced. In addition to reduced errors, the presence of auxiliary augmentation increased the conﬁ- dence of depth judgments, and it was subjectively preferred. The visualization approach did not have an eﬀect on the viewing time. 1 Introduction In augmented reality (AR) applications, the user must be able to simultaneously perceive the surrounding reality and augmented stimuli. In visual augmented reality, the challenge is the mixing of the perception of overlaid graphical objects with objects in the real world. This is problematic because humans are not accustomed to seeing graphical non-photorealistic objects mixed into the ﬁeld of view. Kruijﬀ et al. (2010) state that incorrect depth interpretation of graphical * Authors’ addresses: Department of Media Technology, PL 15500, FI-00076 AALTO, mikko.kyto@aalto.ﬁ, aleksi.makinen@aalto.ﬁ, jukka.hakkinen@aalto.ﬁ, pirkko.oittinen@aalto.ﬁ. c ACM, 2013. This is the author’s version of the work. It is posted here by permission of ACM for your personal use. Not for redistribution. The deﬁnitive version was published in ACM Transactions on Applied Perception, 10, 1, February 2013 http://doi.acm.org/10.1145/2422105.2422111 2