Application-Oriented Design of Smart Camera Networks

Abstract

Smart camera networks are well suited for a wide range of applications. However, system design aspects must be taken into consideration to effectively map an application onto the constraints of a smart camera network. For this reason, we propose an application-oriented design methology that enables the determination of an output set of operation parameters given an input set of application requirements. This design methology for example can yield the minimum camera resolution and density to achieve accuracy and latency performance targets. We illustrate the application-oriented design flow on a distributed Bayesian filtering algorithm for single object tracking. We will also demonstrate how the approach can be readily extended to related applications like occupancy sensing and multi-object tracking. Observation models for single camera and stereo vision systems are introduced with a particular focus on low-resolution image sensors, which offer reduced bandwidth and inherent privacy. Simulation results are presented that answer some of the fundamental tradeoffs between operation parameters and accuracy performance.