Multi-agent Multimodal Human Emotion Recognition Architecture

Abstract

Expressing and perceiving emotions is a vital part of communication between people. The ability to automatically recognize human emotions allows human-machine interactions to resemble the natural way people communicate as an important improvement on the Human Machine Interface. Human emotion recognition can be based on input data from different modalities that contribute for robust and efficient recognition. The paper presents a general framework of a multi-agent system for emotion recognition based on a multilayered multimodal architecture with intelligent agents.