3D Facial Emotion Recognition Using Deep Learning Technique

Abstract

In this paper, a 3D facial emotion recognition model using deep learning technique is proposed. In the deep learning architecture, two convolution layers and a pooling layer is used. Pooling is performed after convolution operation. The sigmoid activation function is used to obtain the probabilities for different classes of human faces. In order to validate the performance of deep learning based face recognition model, Kaggle dataset is used. The accuracy of the model is approximately 65% which is less than the other techniques used for facial emotion recognition. Despite dramatic improvements in representation precision attributable to the non-linearity of profound image representations.