Multi-feature fusion in advanced robotics applications

  • Authors:
  • Zahid Riaz;Christoph Mayer;Michael Beetz;Bernd Radig;M. Saquib Sarfraz

  • Affiliations:
  • Technische Universität München, Garching, Germany;Technische Universität München, Garching, Germany;Technische Universität München, Garching, Germany;Technische Universität München, Garching, Germany;COMSATS Institute of Information Technology, Lahore, Pakistan

  • Venue:
  • Proceedings of the 7th International Conference on Frontiers of Information Technology
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper describes a feature extraction technique from human face image sequences using model based approach. We study two different models with our proposed approach towards multifeature extraction. These features are efficiently used for human face information extraction for different applications. The approach follows in fitting a model to face image using robust objective function and extracting textural and temporal features for three major applications naming 1) face recognition, 2) facial expressions recognition and 3) gender classification. For experimentation and comparative study of our multi-features over two models, we use same set of features with two different classifiers generating promising results to explain that extracted features are strong enough to be used for face image analysis. Features goodness has been investigated on Cohn Kanade Facial Expressions Database (CKFED). The proposed multi-features approach is automatic and real time.