Communication and energy efficiency in visual sensor networks for people localization

This paper addresses the communication and energy efficiency in collaborative visual sensor networks (VSNs) for people localization, a challenging computer vision problem of its own. We focus on the design of a light-weight and energy-efficient solution where people are localized based on distributed camera nodes integrating the so-called certainty map generated at each node, that records the target non-existence information within the camera's field of view. We first present a dynamic itinerary for certainty map integration where not only each sensor node transmits a very limited amount of data but that a limited number of camera nodes is involved. Then, we perform a comprehensive analytical study to evaluate communication and energy efficiency between different integration schemes, i.e., centralized and distributed integration. Based on results obtained from analytical study and real experiments, the distributed method shows effectiveness in detection accuracy as well as energy and bandwidth efficiency.