Browsing by Author "Cao, Zekun"

As a visual cue, rest-frames are isotropically vertical and relatively stationary to subjects; they are used to alleviate visually induced motion sickness in the virtual reality experience. However, the use of rest-frames as a visually induced motion sickness–alleviation technique poses several challenges for the user. This is primarily because the existing design is content-based, not graphics-driver-level, so it cannot be seamlessly applied to various platforms. The design also blocks some peripheral vision, sacrificing the user’s spatial-recognition ability.

This dissertation aims to understand and improve rest-frames’ design, making their use a viable technique to reduce visually induced motion sickness. This technique is graphics-driver-level, so it can run seamlessly in any environment. With this goal, the dissertation proposes and validates a novel design of rest-frames inspired by amodal completion: granulated rest-frames. A generic method to process stereoscopic videos to study user-experience is also presented. The new design implements rest-frames as visual noise with changeable settings to maintain the user’s spatial-recognition ability and provide more flexibility at the graphics-driver level.

This study finds the optimal size and sparsity of granulated rest-frames by investigating the user’s visual-search performance at different settings. It also validates the effect of granulated rest-frames in alleviating visually induced motion sickness in interactive virtual environments. The main finding is that granulated rest-frames can help first-time users adapt to a virtual environment more quickly without affecting the spatial-recognition ability of their peripheral vision. This dissertation also demonstrates the applicability of granulated rest-frames in stereoscopic videos in a preliminary study, as part of which a generic method to investigate users’ reactions to 360° videos is proposed. The dissertation also offers an experimental design to explore the utilization of granulated rest-frames’ effects in general-purpose interaction techniques in a future study.

With increasing prevalence and capabilities of Virtual Reality (VR) as a part of education, entertainment, data analysis and industrial applications, the consideration of constructing a more user-friendly virtual environment becomes more important, which requires the reduction or elimination of simulator sickness (SS) -- a discomfort caused in simulated environment. As a form of motion sickness (MS) -- caused by movement in travel, some signs and symptoms of SS are the same as those of MS's, such as pallor, cold sweating, nausea, and vomiting. Several convincing theories or hypothesis to explain the cause of SS have been advanced, such as Sensory Conflict Theory, Postural Instability Theory and Rest Frames Hypothesis. Based on the first two theories aforementioned, high-quality tracking systems and modification of field of view (FOV) have been proposed to reduce the mismatch between visual perception and vestibular system in virtual environment (VE). Two main drawbacks exist in these techniques: (1) They will make uses weary in VE needs long distance physical navigation; (2) Modifying the field of view will decrease immersion.

To this end, inspired by the rest frame hypothesis and previous study on field of view, the goal of this research is to evaluate the effect of rest frames -- portions of the virtual environment that remain fixed in relation to the real world and do not move as the user moves around -- on SS reduction. A study was performed in this research, in which all participants experienced two different VR sessions: one with rest frames and the other without rest frames. The rest frames were developed by using a transparent cockpit attached with a black metal net in our experimental environment. A questionnaire asking users' discomfort was shown periodically in the VE during the study, which allowed us to record the level of participants' comfort in real time. Participants were required to finish a Simulator Sickness Questionnaire (SSQ) and a presence questionnaire after completing either session, to analyze the influence of rest frames on presence and their overall sickness induced by VR. Through comparing the time spent in VE, the real-time subjects' discomfort levels and scores of SSQ and presence questionnaire from both sessions, we found that rest frames have advantages as follows:

(1) Users could experience VEs without physical navigation;

(2) Rest frames significantly helped users acclimate to VR;

(3) The level of discomfort in VR was reduced;

(4) The level of presence was not impacted.