Fast Face Detection using Support Vector Machine

Y.-H. Tasi and Y.-S. Huang (Taiwan), and T. Poggio (USA)


face component, face detection,support vector machine, image processing.


This paper describes a framework for detecting frontal and near-frontal views of faces in still gray images. This approach mainly consists of four processing steps: eye candidate finding, eyes verification, nose and mouth finding, and whole face verification. The verification of face is performed separately on different step in the framework to speed up the face detection procedure. Support Vector Machine (SVM) classifiers, including eye, nose, mouth, and face are trained to do the verification work. The proposed face detection system can be used to locate multiple faces embedded in complicated background. In addition, this approach can be applied successfully in different kinds of image variations such as face sizes, lighting conditions, and image qualities. Experimental results are included to show the effectiveness of this method.

Important Links:

Go Back