Comparing reality and cinematic virtual reality for visual walkability evaluation. An exploratory study

Abstract

This exploratory study examines how the evaluation context may influence the experience of assessing visual walkability in Cinematic Virtual Reality (CVR) and in the real world. While previous research has demonstrated the potential of CVR for visual walkability evaluation, its accessible and repeatable 360° representations also come with limited interactivity, which may shape how users experience visual walkability evaluation across its core dimensions – Accessibility, Safety, Comfort, and Pleasurability. This was a point of departure for our user study. Five participants evaluated four urban streets through both in-person walks and 360° CVR experiences, using eye-tracking, questionnaires, and interviews. Semantic segmentation of participants’ viewports revealed distinct visual content patterns: CVR emphasized facades, greenery, and sky, while on-site viewports included more sidewalks and road surfaces. Participants identified similar urban features in both conditions but experienced them differently. CVR demonstrated particular strengths in assessing higher-level visual qualities of walkability, such as comfort and pleasurability, while its use for assessing more embodied dimensions, such as safety and accessibility, remained more limited compared with on-site evaluation.