Deep Learning Framework for Compound Facial Emotion Recognition

Abstract

AbstractHuman facial expressions are an indication of true emotions. Accurately recognizing facial expressions is useful in artificial intelligence, computing, medicine, e-learning, and many more. Although facial emotion recognition (FER) can be accomplished primarily through the use of multiple sensors. However, research shows that using facial images/videos to recognize facial expressions are better because emotions can be conveyed through visual expressions that carry important information. In the past, much research was conducted in the field of FER using different approaches such as analysis through different sensor data, using machine learning and deep learning framework with static images and dynamic sequence. Previous FER research focused on studying seven basic emotions: anger, resentment, fear, excitement, sadness, surprise, and neutrality. However, humans that exhibit many more facial expressions are considered compound emotions. Recently, use of deep learning algorithms in FER has been considerable. State-of-the-art results show deep learning-based approaches are powerful over conventional FER approaches. This paper focuses on implementing deep learning frameworks for compound facial emotion recognition systems for detecting compound emotion using the facial expression image dataset compound facial expressions of emotion (CFEE).KeywordsFacial emotion recognitionCompound emotionDeep learningConvolution neural network