Note
Preprocessing of Face Images: Detection of Features and Pose Normalization,☆☆

https://doi.org/10.1006/cviu.1997.0640Get rights and content

Abstract

The reliability, speed, and complexity of virtually any face recognition system are substantially improved if the location and the scale of the faces are known. We propose a method for automatic and robust detection of the eyes and mouth using the context freegeneralized symmetry transformand knowledge of faces. The features are extracted from the image of the intensities gradients and are then used to normalize the face images. We show that a normalization procedure based on affine transformations whose anchor points are the locations of the eyes and mouth substantially increases the effectiveness of general purpose classification techniques in face recognition.

Other normalization procedures for avoiding the effect of background and varying light conditions are proved to be instrumental as well.

References (45)

  • R.O. Duda et al.

    Pattern Classification and Scene Analysis

    (1973)
  • S. Edelman, D. Reisfeld, Y. Yeshurun, Learning to recognize faces from examples, Proceedings of the 2nd European...
  • I. A. Essa, A. P. Pentland, Facial expression recognition using a dynamic model and motion energy, Proceedings of the...
  • I. Fraser, D. Parker, Reaction time measures of feature saliency in perceptual integration task, Aspects of Face...
  • F. Galton

    Composite portraits, made by combining those of many different persons, into a single, resultant figure

    J. Anthropol. Inst.

    (1879)
  • V. Govindaraju, S. N. Srihari, D. B. Sher, A computational model for face location, Proceedings of the 3rd...
  • N. D. Haig, Investigating face recognition with an image processor computer, Aspects of Face Processing, H. EllisM....
  • N. Intrator, D. Reisfeld, Y. Yeshurun, Extraction of facial features for recognition using neural network, Int....
  • T. Kanade

    Picture Processing System by Computer Complex and Recognition of Human Faces

    (November 1973)
  • T. Kanade, J. P. Kender, Mapping image properties into shape constraints: skewed symmetry, affine-transformable...
  • M. Kirby et al.

    Application of the karhunen-loéve procedure for characterization of human faces

    IEEE Trans. Pattern Anal. Mach. Intell.

    (1990)
  • A. Lanitis, C. J. Taylor, T. F. Cootes, A unified approach to coding and interpreting face images, Proceedings of the...
  • Cited by (57)

    • DeepUAge: Improving Underage Age Estimation Accuracy to Aid CSEM Investigation

      2020, Forensic Science International: Digital Investigation
      Citation Excerpt :

      Such photos require image pre-processing and normalisation to align and remove unnecessary features. Reisfeld and Yeshurun (1998) suggest that the knowledge of the location and the scale of a face impacts positively on the speed and reliability of face recognition systems. In 2015, Han et al. (2015) designed a face pre-processing procedure to overcome image variations due to external factors.

    • Efficient generic face model fitting to images and videos

      2014, Image and Vision Computing
      Citation Excerpt :

      Finally, a geometrical face model is used to locate the actual position of a face. In [9], the authors showed that the eyes and mouth in facial images can be robustly detected. They used these points to normalize the images, assuming affine transformation, which can compensate for various viewing positions.

    • Classification of Heart Disease from MRI Images Using Convolutional Neural Network

      2021, Proceedings of IEEE International Conference on Signal Processing,Computing and Control
    • Deep Learning Approach to Classify Parkinson’s Disease from MRI Samples

      2021, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    View all citing articles on Scopus

    Supported by Grant 4478293 by the French-Israeli MOST R & D.

    ☆☆

    A. W. YoungH. D. Ellis

    E-mail: [email protected].

    View full text