Year: 2025
Type: conference-paper
Abstract: Figure 1: (a) SpaceShare reconstructs the environment in real time, allowing users to independently explore the space, understand each other's location and viewpoint, and communicate attention using virtual pointers.(b) While in use, the system generates multimodal summaries of the space by combining conversation transcripts and contextual cues, enabling users to retrieve specific objects or locations using natural language queries.
Authors: Hyeongjin Kim, Erzhen Hu, Seongkook Heo
Institution: University of Virginia
Language: English
Subfield: Human-Computer Interaction
Field: Computer Science
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W7084069638