Design and evaluation of techniques to improve user comfort during prolonged use of virtual reality.

Abstract

Virtual Reality (VR) has become a consumer technology, and as such has spread to include many different applications at work, home, school and play, for multiple kinds of users. However, using conventional immersive head-mounted displays (HMD) can lead to degradation in health, such as physical and mental fatigue and motion sickness. It also creates difficulties in interacting with objects and people in the physical world: immersed users cannot easily locate objects around them (e.g., key- board, mouse, smartphone), or have meaningful face-to-face conversations with people.

This thesis addresses these issues from both the “comfort” side and the “real- world interaction” side. We propose (1) A novel immersive health-recovery technique called “Active Breaks”, (2) a means of dynamically controlling how much of the real world the immersed user can see, and (3) techniques for a non-immersed user to better socially connect with an immersed user. The systems described here are built around a unified framework (“Workspace VR”), that brings together existing and novel technologies for visual and audio cues to support real-world interactions (nearby objects, people) for the VR user, and face-to-face communication for the non-VR user. Using these solutions we present encouraging evidence in terms of implementation and improvements: Regarding Active Breaks, our users highly preferred both real-word-based and VR-based version, although the VR-based eye exercises used in this version had some drawbacks. Regarding systems, our combination of the visual channels resulted in a much larger field of view for the VR user to interact with the physical world. Initial investigations into our first HMD prototype showed that each channel provided accurate view areas (peripheral and central vision) for the user to interact with nearby objects in different task scenarios, received high user preference, potentially maintained a high level of immersion, and did not induce any significant VR sickness. In the latest version, we optimized the involved technologies, added more features that also support the non-VR user (audio channel, eye contact cues), designed a new HMD in a scalable fashion, and planned a face-to-face user study to evaluate the impact of the new system as a whole.