A comparison of three methods of face recognition for home photos.
Journal
International Conference on Computer Graphics and Interactive Techniques, SIGGRAPH 2009, New Orleans, Louisiana, USA, August 3-7, 2009, Poster Proceedings
Date Issued
2009
Author(s)
Abstract
This poster presents experimental results of three face recognition methods -- Support Vector Machine (SVM), Local Binary Pattern (LBP)-based, and Sparse Represented-based Classification (SRC). We will show the experimental results based on AR face database and on home photos. The experiments show that the three algorithms can achieve over 85% recognition rate in AR database. However, the recognition rate is extremely reduced in home photos. SVM and SRC-based method encounter challenges of selecting training model while LBP-based method encounters the challenge of merging over scattered clusters. Our goal is to improve the accuracy and efficiency especially in home photos based on the three methods.
Type
conference paper