Facial Expression Recognition Using Ensemble Learning Approach

Facial expression is a mixture of emotional and neutral elements of a person. Deep neural networks are the most advanced field of research in the artificial intelligence domain. They resolve very complex problems of recognizing facial expressions in computer vision. In this paper, a new network, called EmotionNet, for the recognition of facial expressions is proposed by resembling three deep learning models, custom CNN, custom residual, and custom separability convolution. The performance of the proposed network is demonstrated on three datasets, FER2013, CK+48, and JAFFE. The superior performance of the EmotionNet over the base paper and the state-of-the-art methods is shown.