CISE Research Infrastructure: SENSORIUM: Research Infrastructure for Managing Spatio-Temporal Objects in Video Sensor Networks

Trustees Of Boston University, Boston

Investigators

Abstract

EIA 02-02067 Bestavros, Azer Betke, Margrit; Crovella, Mark E.; Matta, Ibrahim; Sclaroff, Stan Boston University RI: SENSORIUM: Research Infrastructure for Managing Spatio-Temporal Objects in Video Sensor Networks This project, developing an open sensor network research infrastructure (SENSORIUM), aims to catalyze fundamental advances in image and video computing, network protocols, and resource management to deal with unique spatio-temporal constraints of sensor networks. SENSORIUM is composed of a sensor network of video cameras spanning several rooms, networked processing units, and a terabyte database, managed together to satisfy queries using those generated by mobile users within this environment. The infrastructure enables the following research projects: a. Modeling, interpretation, and prediction of human motion in video streams at multiple scales in space/time and at multiple layers of detail; b. Development of efficient location management, routing, transport, and content distribution protocols for multi-resolution/scale streaming sensory data networks; c. Characterization of traffic and access patterns in mobile sensory networks; d. Instrumentation of embedded real-time operating systems to enable coordinated resource management and the development of middleware services for the management of active sensor networks; e. Indexing and mining of large spatio-temporal non-textual sensory datasets, with a particular emphasis on mining of human motions and activities; f. Enhancing code safety for embedded systems through the use of type systems and run-time support, with emphasis on flow-oriented programming; g. Development of algorithms and protocols for supporting security and trust, and for protecting the confidentiality and integrity of data in video sensor networks and repositories. These collaborative projects target two vertical applications. The first aims to merge the physical and cyber worlds in an integrated, well defined, privacy-protecting manner. It involves the development of a system capable of gathering, interpreting, routing, and storing data from distributed video sensors, and answering queries about the physical world on the Web. The second aims to develop assistive environments for people with severe disabilities, to help them gain access to computers, and thereby obtain a tool to communicate with their environments. These applications will unify the various research projects leveraged by the SENSORIUM by acting as catalysts for the development of generic technologies that could be used (and reused) in other vertical applications.