Capturing reality: Validation of omnidirectional video-based immersive virtual reality as a streetscape quality auditing method

Although many studies have developed indoor streetscape audit instruments, given the limitation of conventional image information, these have merely been optimized for auditing objective elements of a street. Against this backdrop, this study ascertains whether omnidirectional videos and fully immersive virtual experiences using head-mount displays can be applied to audit the subjective streetscape qualities, thereby replacing the field audit. The study conducts a series of experiments (test audits) using street view images, 360 VR video, and field experience, comparing them using quantitative and qualitative approaches. The results show that VR audit generates results more consistent with the field audit than the street view based audit. In particular, 360 VR video is appropriate for auditing items related to the sense of space, behavioral qualities, and overall street atmosphere. This result unveils the possibility of a recorded VR-based visual assessment survey as a new indoor audit method for evaluating the quality of urban design, which is rarely addressed in existing streetscape audit tools due to its high subjectivity. However, intra-rater agreement estimates of a VR audit demonstrate that its reproducibility remains unreliable. Moreover, a VR audit is less cost-effective than other methods and is not adequate for auditing items that require sensory experience other than the audiovisual. Thus, even if a VR audit can replace much of the streetscape audit, the need for on-site experience remains unlikely to be replaced.

Automated summary

This study investigates the feasibility of using omnidirectional videos and fully immersive virtual experiences for auditing subjective streetscape qualities. The results show that VR audit is more consistent with field audit, particularly for auditing items related to sense of space, behavioral qualities, and overall street atmosphere. However, reproducibility of VR audit is unreliable and it is less cost-effective than other methods and inadequate for auditing items requiring sensory experience other than audiovisual.