Approach of facial surfaces by contour

In this work we have made a face recognition system based on the analysis of facial surfaces by analyzing the contours extracted from the 3D representation of human faces, called facial curves. Our recognition system uses the face representation of the image to the gray level as 2D surface in 3D space, with the third coordinate proportional to the intensity values of pixels. To extract the feature vector of face we used the method of analysis of facial curves. This approach is based on the mathematical definition of the space of facial surfaces. In this space the deformations are modeled by the geodesic path between two facial surfaces. The proposed method has been used for face recognition and tested on the database of face images known ORL database, developed by AT & T Laboratories Cambridge University in England, was used to evaluate the performance of our method under conditions where the pose and sample size are varied.

[1]  Haibin Ling,et al.  Deformation invariant image matching , 2005, Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1.

[2]  Alexander M. Bronstein,et al.  Three-Dimensional Face Recognition , 2005, International Journal of Computer Vision.

[3]  J A Sethian,et al.  A fast marching level set method for monotonically advancing fronts. , 1996, Proceedings of the National Academy of Sciences of the United States of America.

[4]  Anuj Srivastava,et al.  3D Face Recognition Using Shapes of Facial Curves , 2006, 2006 IEEE International Conference on Acoustics Speech and Signal Processing Proceedings.

[5]  Hassen Drira,et al.  A Riemannian analysis of 3D nose shapes for partial human biometrics , 2009, 2009 IEEE 12th International Conference on Computer Vision.

[6]  Kuo-Chin Fan,et al.  Human face detection using geometric triangle relationship , 2000, Proceedings 15th International Conference on Pattern Recognition. ICPR-2000.