Facial Emotion Detection Using CNN

As Humans, we consider our Sentiments and emotions are significant and profound features of individual conduct. Recognition of human facial Expressions is the process of recognizing a person's current mood. It is essential for advertising, mental health treatment, and medical treatment. With the emergence of deep learning models many new Models like Convolutional Neural Networks (CNN), and Recurrent Neural Networks (RNN) were developed. Though the Deep Learning Models helped in better accuracy than the Machine Learning Models the accuracies were not much acceptable and the models are only suitable for certain controlled Scenarios. The Emergence of Transfer Learning helped develop a model with good accuracy and less computational speed. For this project, we used EfficientNet Model and taken the FER2013 dataset, produced by Kaggle and Google(2013). The dataset consists of almost 7 different classes mainly fear, happiness, anger, surprise, sad, disgust, and neutral.