Multi-Modal 2D and 3D Biometrics for Face Recognition

  • Authors:
  • Kyong I. Chang;Kevin W. Bowyer;Patrick J. Flynn

  • Affiliations:
  • -;-;-

  • Venue:
  • AMFG '03 Proceedings of the IEEE International Workshop on Analysis and Modeling of Faces and Gestures
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

Results are presented for the largest experimental study to datethat investigates the comparison and combination of 2D and 3D facedata for biometric recognition. To our knowledge, this is also theonly such study to incorporate significant time lapse betweengallery and probe image acquisition. Recognition results arepresented for gallery and probe datasets of 166 subjects imaged inboth 2D and 3D,with six to thirteen weeks time lapse betweengallery and probe images of a given subject. Using a PCA-basedapproach tuned separately for 2D and for 3D, we find nostatistically significant difference between the rank-onerecognition rates of 83.1% for 2D and 83.7% for 3D. Using acertainty-weighted sum-of-distance approach to combining 2D and 3D,we find a multi-modal rank-one recognition rate of 92.8%, which isstatistically significantly greater than either 2D or 3D alone.