FACIAL EMOTION RECOGNITION USING MULTISENSOR SYSTEMS

Abstract

Research on how we, as human beings, react with the world around us and how we interact with it remains one of the greatest challenges in science. Human-computer interaction would be more natural if computers were able to register and respond to human emotions. This paper presents some of the systems for emotions recognition in human-computer interaction, that are based on facial expressions, voice, handwriting, as well as the use of mouse and keyboard. Also, the practical results of a system for recognizing four different emotional states (neutral emotional state, joy, surprise, anger) from face images taken with a thermal camera and a visible light camera based on the HOG features and the SVM classifier are presented.