Predict-and-Drive: Avatar Motion Adaption in Room-Scale Augmented Reality Telepresence with Heterogeneous Spaces

Area of Science:

• Computer Science
• Human-Computer Interaction
• Virtual Reality

Background:

• Augmented Reality (AR) telepresence enables 3D interaction via avatars in remote spaces.
• Heterogeneous environments pose challenges for synchronizing avatar and user motion.
• Existing methods for avatar placement have limitations, including restricted interaction areas and lack of smooth transitions.

Purpose of the Study:

• To investigate the impact of transition delay on avatar-mediated telepresence.
• To identify user-preferred transition styles in the presence of delay.
• To develop a system for seamless avatar motion synchronization in AR telepresence.

Main Methods:

• User studies were conducted to analyze the effects of transition delay and preferred styles.
• A Predict-and-Drive controller was proposed to minimize avatar transition delay.
• A grouping component was introduced for rapid target calculation, further reducing delay.
• An avatar control component utilizing artificial potential fields was developed for obstacle-aware navigation.

Main Results:

• Transition delay significantly negatively impacts the telepresence experience.
• User studies provided insights into preferred transition styles.
• The proposed Predict-and-Drive controller and grouping component effectively reduce avatar transition delay.
• Ablation studies validated the effectiveness of the developed components.

Conclusions:

• Addressing avatar transition delay is crucial for effective AR telepresence.
• The proposed Predict-and-Drive controller offers a solution for smoother avatar motion synchronization.
• The system enhances the semantic synchronization between user actions and avatar movements in heterogeneous AR environments.