A Cloud Architecture for Emotion Recognition Based on the Appraisal Theory

Designing robots with the ability to infer a person's emotional state represents one of the major challenges in social robotics. This work proposes a cloud system for online human emotion recognition in spontaneous human-robot verbal interaction, structured as a set of REST API endpoints. Based on the appraisal theory of emotion, the system acquires data about the person's expected appraisal of a given situation, depending on their needs and goals, and combines it with sensory data, such as facial expressions, angles of the head, and gaze of the person, and distance between the person and the robot. The whole set of data is used to infer the person's emotional state during the interaction through a Random Forest classifier, trained for binary classification (i.e., positive vs. negative emotions). Results confirmed that using both data sources improved performance in both the K-fold and the Leave One Person Out scenarios.