The Effect of Camera Perspectives on Locomotion Accuracy in Virtual Reality

Abstract

This paper investigates the effect of commonly used camera perspectives (First-person perspective, over the shoulder, behind the back and from top to bottom) on locomotion efficiency in virtual reality (VR). To test this research objective in a user study, a virtual test environment was implemented in which a ball can be maneuvered using four different camera perspectives. We analyzed the time taken to complete four different paths and the offset between the path taken and the optimal one (path accuracy) displayed on the ground. The study results show that the "first person perspective" and "behind the back" perspectives allow more precise maneuvering in VR, while no difference in speed was found. We defined guidelines that can help game designers and VR developers to create VR applications with high accuracy needs. Our findings are relevant, especially to decide which perspectives are best for specific applications that require high accuracy, such as remote surgery.