14 March 2013 Face recognition with support vector machine
Author Affiliations +
Proceedings Volume 8768, International Conference on Graphic and Image Processing (ICGIP 2012); 876866 (2013) https://doi.org/10.1117/12.2012838
Event: 2012 International Conference on Graphic and Image Processing, 2012, Singapore, Singapore
Face detection and recognition have received much interest over pass ten years due to the many applications range from access control to driver’s licenses. In general, face recognition systems can be classified as: geometric featurebased approaches, template matching and neural approaches. One main drawback of geometric feature-based approaches is not easy to extract and measure the feature. Compared to geometric feature-based approaches, a templatebased approach recognizes faces as a whole. The main idea of these methods is to transform the face image into a low dimensional space. Although the approach of template matching and neural are very efficient, the computation is more complex compared to other algorithms. Support Vector Machines recently have been regarded as an effective statistical learning method for pattern recognition. In this paper, we introduce a support vector machine for face recognition. Next, we have shown the experiment result using the polynomial kernel.
© (2013) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Jian Zhang, Jian Zhang, "Face recognition with support vector machine", Proc. SPIE 8768, International Conference on Graphic and Image Processing (ICGIP 2012), 876866 (14 March 2013); doi: 10.1117/12.2012838; https://doi.org/10.1117/12.2012838


2D DOST based local phase pattern for face recognition
Proceedings of SPIE (April 30 2017)
Face recognition security entrance
Proceedings of SPIE (June 27 2000)
Face recognition using SURF features
Proceedings of SPIE (October 29 2009)

Back to Top