Real Time Human Emotion Recognition Using Artificial Neural Networks

Now a day’s one of the unsolved problem in computer vision is recognizing or understanding other people's emotions and feelings. Although recent methods achieve close to human accuracy in controlled scenarios, the recognition of emotions in the wild remains a challenging problem. In this paper we proposed MAM Pooling (Mean of Average and Maximum) method with CNN to recognize human emotions. We focus on automatic identification of six emotions in real time: Happiness, Anger, Sadness, Surprise, Fear, and Disgust. Convolutional Neural Network (CNN) is a biologically inspired trainable architecture that can learn invariant features for a number of applications. In general, CNNs consist of alternating convolutional layers, non-linearity layers and feature pooling layers. In this work, a Novel feature pooling method, named as MAM pooling is proposed to regularize CNNs, which replaces the deterministic pooling operations with a stochastic procedure by taking the average of  max pooling and average pooling methods. The advantage of the proposed MAM pooling method lies in its wonderful ability to address the over fitting problem encountered by CNN generation.