Particle filtering for bearing-only audio-visual speaker detection and tracking

  • Andrew Rae*
  • , Alaa Khamis
  • , Otman Basir
  • , Mohamed Kamel
  • *Corresponding author for this work

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

5 Scopus citations

Abstract

We present a method for audio-visual speaker detection and tracking in a smart meeting room environment based on bearing measurements and particle filtering. Bearing measurements are determined using the Time Difference of Arrival (TDOA) of the acoustic signal reaching a pair of microphones, and by tracking facial regions in images from monocular cameras. A particle filter is used to sample the space of possible speaker locations within the meeting room, and to fuse the bearing measurements from auditory and visual sources. The proposed system was tested in a video messaging scenario, using a single participant seated in front of a screen to which a camera and microphone pair are attached. The experimental results show that the accuracy of speaker tracking using bearing measurements is related to the location of the speaker relative to the locations of the camera and microphones, which can be quantified using a parameter known as Dilution of Precision.

Original languageEnglish
Title of host publication3rd International Conference on Signals, Circuits and Systems, SCS 2009
DOIs
StatePublished - 2009
Externally publishedYes
Event3rd International Conference on Signals, Circuits and Systems, SCS 2009 - Medenine, Tunisia
Duration: 6 Nov 20098 Nov 2009

Publication series

Name3rd International Conference on Signals, Circuits and Systems, SCS 2009

Conference

Conference3rd International Conference on Signals, Circuits and Systems, SCS 2009
Country/TerritoryTunisia
CityMedenine
Period6/11/098/11/09

Keywords

  • Acoustic arrays
  • Direction of arrival estimation
  • Machine vision
  • Monte carlo methods
  • Position measurement

ASJC Scopus subject areas

  • Computer Networks and Communications
  • Control and Systems Engineering
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Particle filtering for bearing-only audio-visual speaker detection and tracking'. Together they form a unique fingerprint.

Cite this