Gaze-contingent focus cues to enhance depth perception in mixed reality environments

Abstract Mixed reality (MR) has proved to be a useful tool to enhance the physical environment with virtual content. However, significant challenges related to perceptual aspects in MR visualizations still remain. Inaccurate depth perception in MR can lead to errors in tasks such as reaching, object manipulation, and spatial alignment, ultimately reducing usability and immersion. To address this, we introduce a visualization method that provides additional focus cues that can potentially improve depth perception in mixed reality environments. Specifically, we introduce a technique to incorporate gaze-contingent deconvolution-based focus cues into commercial immersive media. We support the effectiveness of our technique in improving depth perception by experimental validation through a user study. Participants were asked to reach a physical position indicated by a virtual object on two types of mixed reality devices. The user study showed an error reduction of 33.3% for the 3D monitor and 53.1% for the Microsoft Hololens 2, compared to standard MR visualizations that lack auxiliary focus cues. Our work shows that auxiliary gaze-contingent focus cues can provide better depth and distance perception while providing an adequate immersive experience.