Vicarious: Context-aware Viewpoints Selection for Mixed Reality Collaboration

Abstract

Mixed-perspective, combining egocentric (first-person) and exocentric (third-person) viewpoints, have been shown to improve the collaborative experience in remote settings. Such experiences allow remote users to switch between different viewpoints to gain alternative perspectives of the remote space. However, existing systems lack seamless selection and transition between multiple perspectives that better fit the task at hand. To address this, we present a new approach called Vicarious, which simplifies and automates the selection between egocentric and exocentric viewpoints. Vicarious employs a context-aware method for dynamically switching or highlighting the optimal viewpoint based on user actions and the current context. To evaluate the effectiveness of the viewpoint selection method, we conducted a user study (n = 27) using an asymmetric AR-VR setup where users performed remote collaboration tasks under four distinct conditions: No-view, Manual, Guided, and Automatic selection. The results showed that Guided and Automatic viewpoint selection improved users' understanding of the task space and task performance, and reduced cognitive load compared to Manual or No-view selection. The results also suggest that the asymmetric setup had minimal impact on spatial and social presence, except for differences in task load and preference. Based on these findings, we provide design implications for future research in mixed reality collaboration.