Facial Emotion Recognition

Humans may make thousands of facial expressions throughout a discussion, varying in intricacy, passion, and significance. This paper discusses way of recognizing different emotions produced by humans using a software application that make use of Haar-Cascade Algorithm and a pre-trained dataset DeepFace. We have used DeepFace with the help of Which we have achieved roughly about 97 percent accuracy approximately. In this paper we also have tried to analyze the problem associated with previous methods. We have also made comparisons of different other technologies with deep face and compared their Accuracies. In paper we have carried out real-time emotion detection in a webcam which is able to detect the emotion of one person. It can be further upgraded to detect the emotion of multiple faces at a time.