4 3D face recognition
4.1 3D face recognition methods

3D face recognition in comparison to 2D face recognition profits from bigger information flow about the face characteristics. Similarly, both approaches need basic preprocessing like face-size normalization, rotation to a neutral position etc. Added information not only about the 2D shape but about the depth analysis offers rich source of information not captured in 2D images. Main advantages to 2D face analysis are:

  • Not affected by illumination variations or use of cosmetics
  • Less sensitive to appearance variations
  • Easier to handle pose variations
  • Projective nature of 2D images
  • Simplifies face & facial feature detection, pose estimation & pose compensation

The selected basic 3D face recognition methods:

image
Fig. 3.1 – Surface-based 3D recognition
image
Fig. 3.2 – Appearance-based 3D recognition
image
Fig. 3.3 – Model-based 3D recognition