Real-Time Facial Emotion Classification using Deep Convolution Neural Network

In the age of digitalization, machine learning researchers are fascinated by the identification of human emotions via their facial expressions or body movements. The emotions which one can see on a person's face reveal a lot about their personal thought processes and shows what's going on inside the mind. Facial expression recognition (FER) is an important and crucial task for the machines to understand the emotional changes in human beings. The main objective of FER is to train a particular model to recognise and analyse human emotions in the similar way that humans do. By constructing a multi-class classifier model, we aim to categorise a human facial image into one of the seven emotions under consideration in this study. To attain the optimum accuracy, we developed our model with varying depths, maximum pooling and dense layers, eventually achieving 86 percent accuracy on the test set. To address overfitting, we applied tactics such as dropout and maxpooling layers. We also demonstrate a real-time emotion recognition method utilising a webcam for users live video feed that gives correct results for user's faces in real time.