HybridSphere: Enhancing Hybrid Meetings with Avatar-Based VR Environments.

Area of Science:

• Human-Computer Interaction
• Virtual Reality
• Communication Technology

Background:

• Hybrid meetings are increasingly common due to technological advances.
• Remote participants in hybrid settings often face reduced contextual awareness and social isolation.
• Existing systems struggle to provide remote users with a sense of presence and shared space.

Purpose of the Study:

• To propose and evaluate HybridSphere, a novel system designed to improve remote participant experience in hybrid meetings.
• To reconstruct a shared virtual reality environment for remote attendees using real-time avatars of local participants.
• To investigate the impact of immersive VR representations on remote engagement and presence.

Main Methods:

• Development of HybridSphere system utilizing a 360-degree camera and pose estimation.
• Generation of real-time avatar representations for local attendees.
• User study comparing HybridSphere with a baseline (unmodified 360-degree video) condition.
• Evaluation of remote participant experience, including presence, engagement, and avatar perception.

Main Results:

• Participants using HybridSphere reported appreciation for the seated-6DoF (six degrees of freedom) function, enhancing spatial flexibility.
• Avatars were rated lower in perceived trustworthiness and likability, attributed to limited visual fidelity.
• Despite avatar limitations, the immersive VR environment showed potential for enhancing remote engagement.

Conclusions:

• Immersive and spatially flexible VR representations, like those in HybridSphere, can significantly enhance remote engagement in hybrid meetings.
• Addressing avatar visual fidelity is crucial for improving user perception and trust in such systems.
• HybridSphere demonstrates a promising approach to mitigating isolation and improving contextual awareness for remote participants.