Facial Emotion Recognition Based on Deep Learning

Abstract

In the field of human-computer interaction, facial expression recognition is a hot topic (HCI).Automatic facial expression analysis has been the topic of various studies due to its practical utility in friendly robotics, medical treatment, driver fatigue detection, and many other human-computer interaction systems. Due to the variability of human faces and differences in photos such as varied facial poses and illumination, accurate and robust FER by computer models remains a challenge. Deep learning models have showed tremendous promise among all FER techniques because to their powerful automatic feature extraction and computational efficiency. On the FER2013 dataset, we attain the greatest classification accuracy in this paper. Used a system that combines the VGG 16 model and Classic Neural Networks. Where the VGG16 model is used to extract features and Classic Neural Networks was used to categorization On the FER-2013 dataset, the model achieves state-of-the-art accuracy of 89.31 percent.