Group Interaction Analysis in Dynamic Context

Computer understanding of human actions and interactions is one of the key research issues in human computing. In this regard, context plays an essential role in semantic understanding of human behavioral and social signals from sensor data. This paper put forward an event-based dynamic context model to address the problems of context awareness in the analysis of group interaction scenarios. Event-driven multilevel dynamic Bayesian network is correspondingly proposed to detect multilevel events, which underlies the context awareness mechanism. Online analysis can be achieved, which is superior over previous works. Experiments in our smart meeting room demonstrate the effectiveness of our approach.

I. Introduction

Implicit human–computer interaction [1] is a typical human–computer interaction mode for pervasive computing. Computer systems analyze users' actions and intentions based on multimodal sensor data, and further provide attentive services to users without drawing their attention away from their current tasks. In intelligent environments such as smart meeting rooms, attentive services may include online archiving of meeting data, dynamic control of active sensors, and intelligent broadcasting. Such services can only be provided if computers understand current and previous group interactions that have been taking place inside the meeting room.