Towards dynamically generating immersive video scenes for studying human-environment interactions

Abstract. Studies where participants interact with their environment are necessary, for example to investigate pedestrian behaviour, in urban planning or for environmental studies. Existing methods such as field studies or 3D virtual reality environments frequently require a lot of effort to create realistic 3D worlds or are subject to real-world interferences. In this paper we propose an alternative approach that uses an immersive video environment (IVE) and a dynamic scene generator (DSG).We use overlays to dynamically generate video scenes to facilitate studying specific combinations of environmental factors with participants. We report on an initial case study in the context of studying pedestrian behaviour, where we show how the approach can facilitate studying how crowds and pedestrian signage affect route choices. Our approach can help researchers studying human-environment interactions in the lab.