Camera Network Topology Discovery Literature Review

Abstract

We present a literature review on the problem of camera network topology discovery, focussing on its two main types: overlapping and non-overlapping camera fields of view. These two problems are fundamentally different and each requires a specifically tailored approach. We describe the most popular approaches for each problem and analyse their suitability for our project. The INVS (Intelligent Networked Video Surveillance) project has been ongoing in the ISRD IAE Group since 2007. The 2010-2011 year of the project focuses on multi-camera tracking and matching. The aim is to demonstrate a live camera handover system for a network of about 10 cameras with both overlapping and non-overlapping views. Hence an important decision in this project is how to handle camera network topology discovery. Here we present a literature review on the problem of camera network topology discovery, focussing on its two main types: overlapping and non-overlapping camera fields of view. These two problems are fundamentally di erent and each requires a specifically tailored approach. We describe the most popular approaches for each problem and analyse their suitability for this project. UNCLASSIFIED

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Nov 01, 2011
Accession Number
ADA558224

Entities

People

  • Dmitri Kamenetsky

Organizations

  • Defence Science and Technology Group

Tags

Communities of Interest

  • Autonomy

DTIC Thesaurus Topics

  • Abstracts
  • Algorithms
  • Artificial Intelligence
  • Australia
  • Computer Science
  • Computer Vision
  • Cross Correlation
  • Detection
  • Detectors
  • Instructions
  • Literature
  • Literature Surveys
  • Machine Learning
  • Network Topology
  • Reconnaissance
  • Surveillance
  • Topology

Fields of Study

  • Computer science

Readers

  • Computer Vision.
  • Distributed Systems and Data Platform Development
  • Systems Analysis and Design