Facial Expression Recognition using Convolutional Neural Network

Facial Expressions pass on a lot of data outwardly instead of articulately. From the past few years, Facial Expression Recognition has been a challenging task in computer vision for Human-Machine Interaction as the way of expressing the emotions varies significantly. The main objective of Facial Expression Recognition (FER) systems is to detect an expressed emotion and recognize the same based on geometry and appearance features. Facial Expression Recognition is performed in four-stages namely pre-processing, face detection, feature extraction, and expression recognition to identify the seven key human emotions such as anger, disgust, fear, happiness, sadness, surprise and neutrality. The FER systems can be used in applications containing behavioural analysis on humans. This paper presents the comparison of different existing systems of Facial Expression Recognition.