Robust 3D face recognition from expression categorisation

  • Authors:
  • Jamie Cook;Mark Cox;Vinod Chandran;Sridha Sridharan

  • Affiliations:
  • Speech, Audio, Image and Video Technology Laboratory, Queensland University of Technology, Brisbane, Queensland, Australia;Speech, Audio, Image and Video Technology Laboratory, Queensland University of Technology, Brisbane, Queensland, Australia;Speech, Audio, Image and Video Technology Laboratory, Queensland University of Technology, Brisbane, Queensland, Australia;Speech, Audio, Image and Video Technology Laboratory, Queensland University of Technology, Brisbane, Queensland, Australia

  • Venue:
  • ICB'07 Proceedings of the 2007 international conference on Advances in Biometrics
  • Year:
  • 2007

Quantified Score

Hi-index 0.00

Visualization

Abstract

The task of Face Recognition is often cited as being complicated by the presence of lighting and expression variation. In this article a novel combination of facial expression categorisation and 3D Face Recognition is used to provide enhanced recognition performance. The use of 3D face data alleviates performance issues related to pose and illumination. Part-face decomposition is combined with a novel adaptive weighting scheme to increase robustness to expression variation. By using local features instead of a monolithic approach, this system configuration allows for expression variability to be modelled and aid in the fusion process. The system is tested on the Face Recognition Grand Challenge (FRGC) database, currently the largest available dataset of 3D faces. The sensitivity of the proposed approach is also evaluated in the presence of systematic error in the expression classification stage.