Acquisition of directional knowledge in virtual environments created by panoramic videos

This thesis documents the creation and analysis of virtual environments generated using panoramic video. These virtual environments offer greater visual realism, but are expensive and time consuming to produce. Experiments were needed to assess how efficiently they support directional tasks or sense of presence. In this study, participants’ ability to locate specific places in the environment and their subjective sense of presence were compared across three conditions: panoramic video, regular video and slide show. Participants reported a stronger sense of presence in the panoramic video condition, although none of the techniques demonstrated a greater efficiency in providing directional knowledge. Thus, it does not appear that the costs of creating panoramic video are warranted, especially for those applications involving only the sequential learning of specific landmark locations. However, the current experimental design was found not revealing differences between the three different locomotion techniques, as the tasks were too difficult for participants.