A Fast and Accurate Multi-Model Facial Expression Recognition Method for Affective Intelligent Robots

Abstract

With the development of affective computing, emotion recognition based on facial expression recognition has become an attractive topic in human-robot interaction. We hope that affective intelligent robots can detect and identify human facial expressions through vision, and make different responses accordingly. In this paper, we compare different facial expression recognition methods that use traditional machine learning models or deep learning models, and propose a fast and accurate multi-model facial expression recognition method for affective intelligence robots to complete the real-time and high-precision facial expression recognition task.