Distributed Algorithms for Feature Extraction Off-loading in Multi-Camera Visual Sensor Networks

TL;DR

This paper presents distributed algorithms for off-loading feature extraction tasks in multi-camera visual sensor networks, aiming to minimize processing time through optimization and coordination strategies.

Contribution

It formulates the task off-loading as an optimization problem and proposes algorithms for distributed solutions, analyzing equilibrium and coordination effects.

Findings

Distributed optimization achieves low completion times with sufficient information.

Central coordination improves stability and predictability.

Network topology and video characteristics significantly impact performance.

Abstract

Real-time visual analysis tasks, like tracking and recognition, require swift execution of computationally intensive algorithms. Visual sensor networks can be enabled to perform such tasks by augmenting the sensor network with processing nodes and distributing the computational burden in a way that the cameras contend for the processing nodes while trying to minimize their task completion times. In this paper, we formulate the problem of minimizing the completion time of all camera sensors as an optimization problem. We propose algorithms for fully distributed optimization, analyze the existence of equilibrium allocations, evaluate the effect of the network topology and of the video characteristics, and the benefits of central coordination. Our results demonstrate that with sufficient information available, distributed optimization can provide low completion times, moreover predictable and stable performance can be achieved with additional, sparse central coordination.