Video-Based Affect Detection in Noninteractive Learning Environments
Video-Based Affect Detection in Noninteractive Learning Environments
复制标题
非交互式学习环境中基于视频的情绪检测
DOI：
--
复制
发表时间：
2015
期刊：
影响因子：
--
通讯作者：
S. D’Mello
中科院分区：
文献类型：
--
作者：
Yuxuan Chen;Nigel Bosch;S. D’Mello
本文探讨了可能的解决方案，从面部表情在文本/图表的理解，上下文没有交互式事件，可用于推断影响检测情感状态的问题。这些数据为基于面部的情感检测提出了一个有趣的挑战，因为学生面部视频中情感面部表情的可能位置完全未知。在目前的研究中，学生从事的文本/图表的理解活动后，他们自我报告他们的困惑，挫折和参与的水平。从视频中的各个位置选择数据，并提取基于纹理的面部特征来构建影响检测器。还使用了不同量的数据来确定用于分析每个影响检测器的适当数据窗口。使用ROC曲线下面积（AUC）测量检测器性能，其中机会水平为0.5，完美分类为1。困惑（AUC = 0.637），参与（AUC = 0.554）和挫折（AUC = 0.609）在高于机会的水平上被检测到。展望改进的方法，找到可能的位置的情感状态进行了讨论。
The current paper explores possible solutions to the problem of detecting affective states from facial expressions during text/diagram comprehension, a context devoid of interactive events that can be used to infer affect. These data present an interesting challenge for face-based affect detection because likely locations of affective facial expressions within videos of students’ faces are entirely unknown. In the current study, students engaged in a text/diagram comprehension activity after which they selfreported their levels of confusion, frustration, and engagement. Data were chosen from various locations within the videos, and texture-based facial features were extracted to build affect detectors. Varying amounts of data were used as well to determine an appropriate window of data to analyze for each affect detector. Detector performance was measured using Area Under the ROC Curve (AUC), where chance level is .5 and perfect classification is 1. Confusion (AUC = .637), engagement (AUC = .554), and frustration (AUC = .609) were detected at above-chance levels. Prospects for improving the method of finding likely positions of affective states are also discussed.