Facial Expression Recognition Using Multiple Feature Sets

Over the years, human facial expression recognition has always been a challenging problem in computer vision systems. In this paper, we have worked towards recognizing facial expressions from the images given in JAFFE database. From literature, a set of features have been identified to be potentially useful for recognizing facial expressions. Therefore, we propose to use a combination of 3-different types of features i.e. Scale Invariant Features Transform (SIFT), Gabor wavelets and Discrete Cosine Transform (DCT). Some pre-processing steps have been applied before extracting these features. Support Vector Machine (SVM) with radial basis kernel function is used for classifying facial expressions. We evaluate our results on the JAFFE database under the same experimental setup followed in literature. Experimental results show that our proposed methodology gives better results in comparison with existing literature work so far.