Group Facial Emotion Analysis System Using Convolutional Neural Network

Facial emotion analysis is one of the most important tasks in affective computing, human-computer interaction, computer vision, video game testing and market research. Facial expressions are a kind of non-verbal communication because it shows the internal feelings and emotional states of a person. Emotions are a kind of feedback through which what customers want and need can be identified. In this paper, the human group facial sentiment recognition system is proposed using a deep learning approach. The proposed framework uses the Haar filter to detect and extract face features. Then the convolutional neural network (CNN) is developed to recognize facial expressions and classify them into five basic emotion states namely happy, sad, anger, surprise and neutral. Finally, the predicted group emotions are fed into an audio synthesizer to get audio output. The proposed model achieves a final accuracy of 65% for Facial Expression Recognition (FER)-2013 and 60% for custom datasets.