Recognizing Human Emotion Using Computer Vision

Detecting and recognizing human emotion is not an easy thing to do. This will be a big challenge in computer vision because many factors have to be taken while developing an emotion recognition system like facial hair, the presence of spectacles and diverse environments. This paper will use CNN (Convolution Neural Network) model to recognize human emotion from the livefeed using the FER2013 (Facial Expressions Recognition-2013)dataset to train the model. This model delivers 54% accuracy from recognizing seven basic human emotions such as neutral, angry, disgust, fear, happy, sad, and surprise.