情感建模与情感识别

情感计算是关于、产生于和影响于情感方面的计算,其目的是赋予计算机识别、理解、表达和适应人情感的能力。情感计算通过各种传感器获取由人的情感所引起的表情及生理变化信号,利用“情感模型”对这些信号进行识别,从而理解人的情感并做出适当的响应。该文主要讨论了Picard教授在情感计算中情感识别部分的研究成果,着重分析了面部表情、语音、生理信号的情感模型与情感识别,这是情感计算研究的一个关键问题之一,也是建立和谐人机环境的基础之一。

Recognizing and Expressing Affect

Affective computing is computing th at relates to,arise from or deliberately influences emotions.These facts combine d with abilities computers are acquiring in recognizing?understanding and expre ssing affect.Affective computing achieves the expression or physiological signa ls that attribute to human's emotions through all kinds of sensors,and uses"af fective models"to recognize these signals,understands human's affect and respo nds correspondingly.This paper introduces Pro.Picard' research of affective c omputing,focused on the affective models of facial expressions,vocal and physi ological signals,and affective recognition,which are the key technologies of a ffective computing,and the foundations of harmonious human-computer interactio ns.