Skip to Main Content
In order to accomplish subject-independent facial expression recognition, a facial expression recognition approach based on wavelets decomposition and adaboost feature weighting is presented in this paper. At first, wavelet is adopted to decompose images into several bands of frequency images from which the LBP features are extracted. Then adaboost is introduced to learn the dichotomy-dependent weights for SVM classification because different image region has different contribution when dichotomizing different expression pairs. Finally, we compare the recognition accuracy with several other popular expression recognition paradigms. The results show that the proposed improvements in this paper have promoted the performance of facial expression recognition prominently.