Abstract
While the theoretical foundation of optimal camera placement has been studied for decades, its practical implementation has recently attracted significant research interest due to the increasing popularity of visual sensor network. The discrete camera placement problem is NP-hard and many approximate solutions have been independently studied. The goal of this paper is to provide a comprehensive framework in comparing the merits of these techniques. We consider two general classes of camera placement problems and adapt some of the most commonly used approximation techniques in solving them. The accuracy, efficiency and scalability of each technique are analyzed and compared in depth. Extensive experimental results are provided to illustrate the strength and weakness of each method.
Highlights
Due to the significant progress in visual sensor technology, wireless communication, and pattern recognition algorithms, the deployment of wide-area visual sensor networks has become practical and cost-effective
We show the effectiveness of using semidefinite programming relaxations (SDP) on relaxation on a simple and a moderate complex environment
We have presented and compared strengths and weaknesses of various well-known optimization frameworks to solve the generic camera placement problem including a greedy approach, Markov-Chain Monte Carlo (MCMC) methods, and linear programming (LP) and SDP relaxations
Summary
Due to the significant progress in visual sensor technology, wireless communication, and pattern recognition algorithms, the deployment of wide-area visual sensor networks has become practical and cost-effective. Application-specific constraints and objective functions can be defined based on these variables to formulate a binary integer programming (BIP) optimization problem. A myriad of approximate algorithms have been applied in solving BIP-based camera placement problem. Many of the proposed approaches [1,2,3,4,5,6,7,8] are customized for specific applications and a fair evaluation of different approximate algorithms on solving the general camera placement problem remains elusive. We present the BIP formulation for a wide variety of camera placement problems and provide a comprehensive framework to study various approximation algorithms in solving them.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
