Facial expression recognition by deep learning

Abstract

Facial expressions have been proven to be a key element in social interaction. With the increasing popularity of artificial intelligence, facial expression systems using various methods have been designed and studied. Traditional machine learning methods such as support vector machines are widely used in this field. However, most of the traditional machine learning methods require a lot of domain expertise as features need to be identified manually. In contrast, deep learning makes use of network layers to learn features hierarchically by itself. Therefore, this project is to study and develop a facial expression recognition system using convolutional neural network. In this report, the way convolutional neural network works, datasets used (the Extended Cohn-Kanade Dataset), neural network used (VGG net), implementation of the system, design of graphical user interface and future works are discussed.