文章摘要

文章信息/Info

作者:: 殷梦馨1; 2; 倪娜1; 尉怀怀2; 安飞2; 许欣悦1; 2; 武圣君2; 刘旭峰2; 王秀超2△; （1.陕西中医药大学公共卫生学院，咸阳 712046；2.空军军医大学军事医学心理学系，西安 710032）

Author(s):: YIN Mengxin1; 2; NI Na1; YU Huaihuai2; AN Fei 2; XU Xinyue1; 2; WU Shengjun 2; LIU Xufeng2; WANG Xiuchao2; （1.School of Public Health, Shaanxi University of Chinese Medicine , Xianyang 712046,China;2.Department of Military Medical Psychology,Air Force Medical University, Xi′an 710032,China）

Keywords:: Emotion recognition; Multiple physiological signals; Behavioral expression; Single mode; Multimodal fusion; Neural network; Feature extraction

摘要:: 情绪是对一系列主观认知经验的高度概括，对人类的发展意义重大。而在人机交互领域，情绪识别是实现机器智能化的关键一步。随着信息技术的发展，目前针对该领域的研究主要集中在基于神经生理模态、基于外在行为表现模态以及基于多模态的情绪识别。本文从以上角度出发，分别梳理了近年来情绪识别领域的发展进程和研究现状，并对未来的研究方向进行了展望。

Abstract:: Emotion is a high-level generalization of a series of subjective cognitive experiences, which is of great significance to human development. In the field of human-computer interaction, emotion recognition is a key step in realizing machine intelligence. With the development of information technology, for this emerging field, researchs are mainly focus on emotion recognition based on neurophysiological modalities, modalities based on external behavioral performance and multimodality. Based on these three perspectives, this paper combs the development process and research status of emotion recognition in recent years, and looks forward to the future research directions.