Facial Trait Code and Its Application to Face Recognition

Authors:
Ping-Han Lee;Gee-Sern Hsu;Tsuhan Chen;Yi-Ping Hung
Affiliations:
National Taiwan University, ROC;National Taiwan University of Science and Technology, ROC;Carnegie Mellon University, USA;National Taiwan University, ROC
Venue:
ISVC '08 Proceedings of the 4th International Symposium on Advances in Visual Computing, Part II
Year:
2008

Citing 10
Cited 1

Contrast limited adaptive histogram equalization

Graphics gems IV
Eigenfaces vs. Fisherfaces: Recognition Using Class Specific Linear Projection

IEEE Transactions on Pattern Analysis and Machine Intelligence
Multiclass learning, boosting, and error-correcting codes

COLT '99 Proceedings of the twelfth annual conference on Computational learning theory
Unsupervised Learning of Finite Mixture Models

IEEE Transactions on Pattern Analysis and Machine Intelligence
Face Recognition by Support Vector Machines

FG '00 Proceedings of the Fourth IEEE International Conference on Automatic Face and Gesture Recognition 2000
Face recognition: component-based versus global approaches

Computer Vision and Image Understanding - Special issue on Face recognition
Face Class Code Based Feature Extraction for Face Recognition

AUTOID '05 Proceedings of the Fourth IEEE Workshop on Automatic Identification Advanced Technologies
Eigenfaces for recognition

Journal of Cognitive Neuroscience
Solving multiclass learning problems via error-correcting output codes

Journal of Artificial Intelligence Research
Face recognition vendor test 2002 performance metrics

AVBPA'03 Proceedings of the 4th international conference on Audio- and video-based biometric person authentication

Polymorphous facial trait code

ACCV'09 Proceedings of the 9th Asian conference on Computer Vision - Volume Part III

Quantified Score

Hi-index	0.00

Visualization

Abstract

We propose the Facial Trait Code (FTC) to encode human facial images. The proposed FTC is motivated by the discovery of the basic types of local facial features, called facial trait bases , which can be extracted from a large number of faces. In addition, the fusion of these facial trait bases can accurately capture the appearance of a face. Extraction of the facial trait bases involves clustering and boosting approaches, leading to the best discrimination of the human faces. The extracted facial trait bases are symbolized and make up the n-ary facial trait codes. A given face can be then encoded at the patches specified by the traits to render an n-ary facial trait code with each symbol in its codeword corresponding to the closest trait base. We applied FTC to a typical face identification problem, and it yielded satisfactory results under different illumination conditions.