Background initialisation by spatio-temporal motion estimation

Sriram Varadarajan, H. Wang, Bryan Scotney, O Nibouche

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Estimating the initial background of a scene is a key prerequisite for several applications in video analytics. In this paper, we present a simple approach that takes into account spatio-temporal motion intensities while estimating the true background. We tested the algorithm on real video sequences from the Scene Background Initialization (SBI) benchmark dataset, and the results show that the algorithm is competitive compared to the state of the art.
Original languageEnglish
Title of host publication14th IEEE International Conference on Advanced Video and Signal Based Surveillance
ISBN (Electronic)978-1-5386-2939-0
DOIs
Publication statusPublished (in print/issue) - 23 Oct 2017
Event14th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS) - Lecce, Italy
Duration: 29 Aug 20171 Sept 2017
https://ieeexplore.ieee.org/xpl/conhome/8055736/proceeding

Conference

Conference14th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS)
CityItaly
Period29/08/171/09/17
Internet address

Keywords

  • Video sequences
  • Measurement
  • Estimation
  • Image color analysis
  • Clustering algorithms
  • Neural networks
  • Adaptation models

Fingerprint

Dive into the research topics of 'Background initialisation by spatio-temporal motion estimation'. Together they form a unique fingerprint.

Cite this