Facial Expression Recognition Using Deep Learning

Information Technology

Human beings are emotional and expressive. Their state of being i.e. their emotions, decide a lot of things. Whether they are understanding something or if they are confused, one can get an idea, by seeing their facial expression. This is the idea behind this paper, which proposes to build a system to recognize various emotions of a person by capturing his/her facial expressions using the deep learning technology. The application of such a system would be to use it in a school or any other educational system to keep a track of how many students actually understand what is being taught and how many don’t. As per the statistics, teaching methodologies can be improved to get better results. The six human emotions include: anger, fear, happiness, sadness, disgust and surprise. We will be using convolutional neural networks to analyze our datasets. Bearing in mind the challenge faced by teachers to understand the understanding of a student, we thereby propose a solution which is automatic and accurate using the revolutionary and most practical technology-‘ Deep Learning’. We employed deep convolutional neural networks to train our model. Also, we analyzed various classifiers and their performance which includes Distance Based Algorithms (DBA), Neural Networks and Normal Bayesian Classifier (NBC).

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.