This project involves building and training a convolutional neural network (CNN) in Keras to recognize facial expressions. The data consists of 48x48 pixel grayscale images of faces. The objective is to classify each face based on the emotion shown in the facial expression into one of seven categories (0=Angry, 1=Disgust, 2=Fear, 3=Happy, 4=Sad, 5=Surprise, 6=Neutral).OpenCV is used to automatically detect faces in images and draw bounding boxes around them. After training, saving, and exporting the CNN, The model is served to a web interface to perform real-time facial expression recognition on video and image data. The expression recognition can also be carried out on a live video stream captured from device camera.
Note - The training and validation set taken for this project is the FER2013 dataset from kaggle. However, it hasnt been uploaded here does to its large size.