MVIEWS: Multimodal Tools for the Video Analyst

Abstract

Full-motion video has inherent advantages over still imagery for characterizing events and movement. Military and intelligence analysts currently view live video imagery from airborne and ground-based video platforms, but few tools exist for efficient exploitation of the video and its accompanying metadata. In pursuit of this goal, SRI has developed MVIEWS, a system for annotating, indexing, extracting, and disseminating information from video streams for surveillance and intelligence applications. MVIEWS is implemented within the Open Agent Architecture, a distributed multiagent framework that enables rapid integration of component technologies; for MVIEWS, these technologies include pen and voice recognition and interpretation, image processing and object tracking, geo-referenced interactive maps, multimedia databases, and human collaborative tools.