Mouth's Action Units Recognition Base on Non-Frontal View 3D Images


Automatic facial expression recognition because of its applications in many areas such as human and robot interactions, clinical psychology, psychiatry, neurology, pain assessment and lie detection, become one of the principal investigation subjects in image processing. In several last decades, considerable studies were done. Most of these studies were based on 2D static images or 2D video sequences and only in recent years, researches were attended to use of 3D images. We based our research subject on 3D methods Because of these method abilities to overcome facial expressions recognition problems. In this paper we propose a method for discriminate smile and open mouth from neutral face. We extract features by using Maximum Curvatures and Linear Discriminate Analysis and then use RBF neural network for classifying. At the end, we test our proposal method with some non-frontal view 3D images from FRAV3D database and achieve a 91.3% average recognition rate.

  • Abstract
  • Key Words
  • 1. Introduction
  • 2. Project Description
  • 3. Principal Curvatures
  • 4. Linear Discriminant Analysis
  • 5. Radial Basis Function Neural Network
  • 6. FRAV-3D Database
  • 7. Experimental Results
  • 8. Summaries
  • 9. Acknowledgments
  • Refrences

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In