Using Motion Controllers in Virtual Conferencing

Abstract

At the end of 2010 Microsoft released a new controller for the Xbox 360 called Kinect. Unlike ordinary video game controllers, the Kinect works by detecting the positions and movements of a user’s entire body using the data from a sophisticated camera that is able to detect the distance between itself and each of the points on the image it is capturing. The Kinect device is essentially a low-cost, widely available motion capture system. Because of this, almost immediately many individuals put the device to use in a wide variety applications beyond video games.

This thesis investigates one such use; specifically the area of virtual meetings. Virtual meetings are a means of holding a meeting between multiple individuals in multiple locations using the internet, akin to teleconferencing or video conferencing. The defining factor of virtual meetings is that they take place in a virtual world rendered with 3D graphics; with each participant in a meeting controlling a virtual representation of them self called an avatar.

Previous research into virtual reality in general has shown that there is the potential for people to feel highly immersed in virtual reality, experiencing a feeling of really ‘being there’. However, previous work looking at virtual meetings has found that existing interfaces for users to interact with virtual meeting software can interfere with this experience of ‘being there’. The same research has also identified other short comings with existing virtual meeting solutions.

This thesis investigates how the Kinect device can be used to overcome the limitations of exiting virtual meeting software and interfaces. It includes a detailed description of the design and development of a piece of software that was created to demonstrate the possible uses of the Kinect in this area. It also includes discussion of the results of real world testing using that software, evaluating the usefulness of the Kinect when applied to virtual meetings.