Task and Environment-Aware Virtual Scene Rearrangement for Enhanced Safety in Virtual Reality.

Area of Science:

• Computer Science
• Human-Computer Interaction
• Virtual Reality

Background:

• Virtual reality (VR) offers immersive experiences but poses physical safety risks due to user unawareness of surroundings.
• Current VR safety solutions like guardian spaces and locomotion systems involve trade-offs, impacting immersion or causing motion sickness.

Purpose of the Study:

• To propose a novel method for dynamically rearranging VR scenes to align with users' physical spaces.
• To enhance safety and user experience in VR by integrating physical constraints into virtual environments.

Main Methods:

• A computational model was designed to optimize VR scene rearrangement using a cost function.
• The model ensures collision-free interactions while preserving visual fidelity and task completion goals.

Main Results:

• Experimental results indicate significant improvements in user experience and safety within VR environments.
• The proposed method effectively harmonizes physical and virtual environments.

Conclusions:

• Dynamic VR scene rearrangement offers a promising solution for mitigating physical hazards in immersive VR.
• This approach enhances VR usability by seamlessly blending the physical and virtual.