A Distributed Framework for Spatio-Temporal Analysis on Large-Scale Camera Networks

Cameras are becoming ubiquitous. Applications including video-based surveillance and emergency response exploit camera networks to detect anomalies in real time and reduce collateral damage. A well-known technique for detecting anomalies is spatio-temporal analysis -- an inferencing technique employed by domain experts (e.g., vision researchers) to answer spatio-temporal queries. In this paper, we propose a distributed framework that facilitates the development and deployment of spatio-temporal analysis applications on large-scale camera networks and backend computing resources. We make the following contributions: (a) an investigation of the computation/communication costs associated with spatio-temporal analysis, (b) a programming framework designed for large-scale spatio-temporal analysis, and (c) performance evaluations for each step of the spatio-temporal analysis with realistic algorithms.