A room of one’s own: A high-accuracy calibration procedure to align spatial dimensions between a virtual and a real room

Virtual Reality allows to place persons in an unlimited number of different rooms that may or may not resemble real rooms. Controlled experiments on the other side may require that the visual representation of the virtual room is aligned to the room dimensions of the real room in which the experiment is conducted. This is especially important when visual cues in virtual reality are combined with other sensory cues in the real room, e.g. sound played via a loudspeaker (auditory) or objects that have to be manipulated (haptic). Crucially, while the creation of virtual rooms is supported by modern game engines, the alignment of virtual and real rooms is typically not included as a build-in function. To address this issue, we have developed a room calibration procedure that closely aligns virtual and real rooms based on translational and rotational information derived from motion controllers that are placed on fixed mounts. An evaluation of the procedure revealed that the technique can achieve accuracy of 2.86 cm for translation and 0.96 degree for rotation. The data also point towards optimal conditions to improve calibration accuracy. Further use-cases like the positioning of freely moving objects are discussed. The calibration procedure is made freely accessible as an Unreal Engine plugin. Furthermore, construction drawings for the mount are provided. In sum, we present an open-access, easy-to-implement, and high-accuracy calibration procedure.