Oasis: Procedurally Generated Social Virtual Spaces from 3D Scanned Real Spaces

Area of Science:

• Computer Science
• Human-Computer Interaction
• Virtual Reality

Background:

• Current virtual reality (VR) systems often require users to clear physical space, hindering accessibility.
• Existing VR environments lack seamless integration with real-world physical spaces and objects.

Purpose of the Study:

• To introduce Oasis, a novel system for automatic generation of immersive and interactive virtual reality environments.
• To enable natural, real-walking experiences in shared virtual environments for both single and multi-user scenarios.
• To allow users to create VR experiences in their own indoor spaces without furniture rearrangement.

Main Methods:

• 3D capture of indoor scenes and mapping of walkable areas.
• Utilizing depth information for real-world object recognition and pairing with virtual counterparts.
• Implementation of single and multi-user systems on a Tango tablet device.

Main Results:

• Demonstrated a novel approach to overlay virtual environments onto existing physical spaces.
• Successfully created immersive and interactive VR experiences for single and multi-user interactions.
• Enabled seamless, natural walking within shared virtual environments.

Conclusions:

• Oasis provides a new method for casual users to create VR environments without specialized equipment or training.
• The system enhances VR immersion by leveraging real-world physicality.
• Oasis facilitates natural interaction and movement in shared virtual spaces, overcoming limitations of current VR setups.