Measuring facial expressions by computer image analysis
Measuring facial expressions by computer image analysis
复制标题
DOI：
10.1017/s0048577299971664
复制
发表时间：
1999-03-01
期刊：
影响因子：
3.7
通讯作者：
Sejnowski, TJ
中科院分区：
文献类型：
--
作者：
Bartlett, MS;Hager, JC;Sejnowski, TJ
面部表情为研究情绪、认知过程和社会互动提供了重要的行为测量手段。面部动作编码系统（Ekman &弗里森，1978）是一种客观的方法，用于量化面部运动的组成部分的行动。我们将计算机图像分析应用于自动检测图像序列中的面部动作的问题。三种方法进行了比较：整体空间分析，明确的测量功能，如皱纹，估计运动流场。这三种方法结合在一个混合系统中，该系统以91%的准确率对六种上面部动作进行分类。混合系统在这项任务上的表现优于人类非专家，并且表现得与训练有素的专家一样好。一个自动化的系统将使面部表情测量更广泛地作为行为科学和情感神经基础研究的一种研究工具。
Facial expressions provide an important behavioral measure for the study of emotion, cognitive processes, and social interaction. The Facial Action Coding System (Ekman & Friesen, 1978) is an objective method for quantifying facial movement in terms of component actions. We applied computer image analysis to the problem of automatically detecting facial actions in sequences of images. Three approaches were compared: holistic spatial analysis, explicit measurement of features such as wrinkles, and estimation of motion flow fields. The three methods were combined in a hybrid system that classified six upper facial actions with 91% accuracy. The hybrid system outperformed human nonexperts on this task and performed as well as highly trained experts. An automated system would make facial expression measurement more widely accessible as a research tool in behavioral science and investigations of the neural substrates of emotion.