Using Spatialized Audio to Improve Human Spatial Knowledge Acquisition in Virtual Reality

Picture schematization, the construction of spatial object maps from images, has useful applications ranging from indoor exploration to augmented reality. Since using human spatial knowledge improves schematization, crowdsourcing workflows are introduced for extracting spatial information from pictures. As 360° pictures are now available in virtual reality (VR), crowdsourced 360° picture schematization also becomes essential. Yet, the vergence-accommodation conflict (VAC) in head-mounted displays (HMDs) causes inaccurate spatial perception in VR. We propose integration of spatial audio in VR as a cost-effective and intuitive feature to support spatial perception. This study indicates spatial audio cues in VR are naturally incorporated by humans to significantly improve human spatial knowledge accuracy and, subsequently, crowdsourcing schematization.