Towards 360 VR Sickness Mitigation: From Virtual Reality Eye-Tracking to Visual Communication.

Area of Science:

• Virtual Reality (VR) and Human-Computer Interaction
• Visual Perception and Neuroscience

Background:

• Virtual reality (VR) content development often overlooks VR sickness, a critical issue impacting user safety and adoption.
• Existing VR systems lack effective methods to mitigate VR sickness, hindering immersive experiences.

Purpose of the Study:

• To investigate a novel VR sickness mitigation framework leveraging human visual characteristics.
• To develop and validate a foveated filtering approach for reducing VR sickness in 360 VR content.

Main Methods:

• Creation of a large-scale 360 VR content database (VRSP360) for VR sickness analysis.
• Conducting eye-tracking experiments to understand user gaze patterns during VR sickness.
• Designing a foveated filtering framework that limits peripheral high-frequency textures.

Main Results:

• Observed a center-biased gaze distribution in users experiencing high VR sickness.
• Demonstrated that foveated filtering effectively mitigates VR sickness by reducing peripheral visual information.
• Developed metrics (visual texture index, perceptual information index) to assess information preservation.

Conclusions:

• The proposed foveated filtering framework successfully mitigates VR sickness by approximately 19%.
• The method balances sickness reduction with the conservation of perceptual quality, crucial for user presence.
• This framework offers a promising solution for safer and more comfortable VR experiences.