Recognizing Facial Expression with Non-negative Matrix Factorization
Pan Qi-a · Microcomputer Information · 2009
In this paper two image representation approaches called Non-negative Matrix Factorization (NMF) and Principal Component Analysis (PCA) have been applied to facial databases for recognizing three basic facial expressions: joy, surprise and neutral pose. The facial expression features extracted by the NMF and the PCA are send to the Maximum correlation classifier separately to compare the different result. Proved in the experiment based on the CMU face database, we got the conclusion that the NMF had a better performance.