Book chapter
Autonomous Multi-camera Tracking Using Distributed Quadratic Optimization
Energy Minimization Methods in Computer Vision and Pattern Recognition
22 Mar 2018
Abstract
Multi-camera object tracking is an efficient approach commonly used in security and surveillance systems. In a conventional multi-camera setup, a central computational unit processes large amounts of data in real time that is provided by distributed cameras. High network traffic, cost of storage on the central unit, scalability of the system, and vulnerability of the central unit to attacks are among the disadvantages of such systems. In this paper, we present an autonomous multi-camera tracking system to overcome these challenges. We assume cameras that are capable of limited computation for locally tracking a subset of objects in the scene, as well as peer-to-peer network connectivity among the cameras with a decent bandwidth that is sufficient for message passing to achieve coordination. We propose an efficient distributed algorithm for coordination and load-balancing among the cameras. We also provide experimental results to validate the utility of the proposed algorithm in comparison to a centralized algorithm.
Metrics
21 Record Views
2 citations in Scopus
Details
- Title
- Autonomous Multi-camera Tracking Using Distributed Quadratic Optimization
- Creators
- Yusuf Osmanlıoğlu - Section of Biomedical Image Analysis, University of Pennsylvania, Philadelphia, USABahareh Shakibajahromi - Drexel UniversityAli Shokoufandeh - Drexel University
- Publication Details
- Energy Minimization Methods in Computer Vision and Pattern Recognition
- Series
- Lecture Notes in Computer Science
- Publisher
- Springer International Publishing; Cham
- Resource Type
- Book chapter
- Language
- English
- Academic Unit
- Computer Science (Computing)
- Scopus ID
- 2-s2.0-85044783724
- Other Identifier
- 991019173769904721