Analysis of emotion expression on frontal and profile facial images

A. Iskra
{"title":"Analysis of emotion expression on frontal and profile facial images","authors":"A. Iskra","doi":"10.24867/grid-2022-p22","DOIUrl":null,"url":null,"abstract":"Expressions of emotions are often found in facial images. In addition to the neutral facial expression, we know six basic expressions of emotion: joy, anger, sadness, fear, surprise, and disgust. The similarity of some emotion expressions sometimes leads to incorrect recognition or confusion of two emotions. In our study, we tried to find out how these substitutions manifest in the recognition of emotions on frontal and profile face images. The results of the substitutions in emotion recognition were presented with a substitution matrix. The second part of the study focused on confirming these results with the analysis of facial feature observation and fixation duration. In the analysis of facial features, the three main facial features (eyes, mouth, and forehead with nasal) that attract the most attention were considered. Fixation duration was also measured for these facial features. The basis of the research equipment was an eye tracker, which we used to define the areas of interest (AOI) for the analysis. The results of the observational proportions of facial features confirmed a relatively large scale of substitutions of the emotions fear and surprise, anger and disgust, and partial fear and disgust in frontal facial images. In profile facial images, the most frequent incorrect recognition were the emotions happiness and surprise, anger and disgust, fear and disgust, and anger and sadness. Since there is less information about the face in the profile facial image than in the frontal facial images, the results also confirmed a higher proportion of incorrect recognition in the profile face images and thus a more difficult recognition of emotions in the profile face images. The greater extent of incorrect recognition was also confirmed by the fixation duration results. Both results (observation proportions of facial features and fixation duration) were also presented in a graph.","PeriodicalId":107864,"journal":{"name":"Proceedings - The Eleventh International Symposium GRID 2022","volume":"45 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2022-11-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Proceedings - The Eleventh International Symposium GRID 2022","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.24867/grid-2022-p22","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0

Abstract

Expressions of emotions are often found in facial images. In addition to the neutral facial expression, we know six basic expressions of emotion: joy, anger, sadness, fear, surprise, and disgust. The similarity of some emotion expressions sometimes leads to incorrect recognition or confusion of two emotions. In our study, we tried to find out how these substitutions manifest in the recognition of emotions on frontal and profile face images. The results of the substitutions in emotion recognition were presented with a substitution matrix. The second part of the study focused on confirming these results with the analysis of facial feature observation and fixation duration. In the analysis of facial features, the three main facial features (eyes, mouth, and forehead with nasal) that attract the most attention were considered. Fixation duration was also measured for these facial features. The basis of the research equipment was an eye tracker, which we used to define the areas of interest (AOI) for the analysis. The results of the observational proportions of facial features confirmed a relatively large scale of substitutions of the emotions fear and surprise, anger and disgust, and partial fear and disgust in frontal facial images. In profile facial images, the most frequent incorrect recognition were the emotions happiness and surprise, anger and disgust, fear and disgust, and anger and sadness. Since there is less information about the face in the profile facial image than in the frontal facial images, the results also confirmed a higher proportion of incorrect recognition in the profile face images and thus a more difficult recognition of emotions in the profile face images. The greater extent of incorrect recognition was also confirmed by the fixation duration results. Both results (observation proportions of facial features and fixation duration) were also presented in a graph.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
正面和侧面面部图像的情感表达分析
情绪的表达经常出现在面部图像中。除了中性的面部表情,我们还知道六种基本的情绪表达:喜悦、愤怒、悲伤、恐惧、惊讶和厌恶。某些情绪表达的相似性有时会导致对两种情绪的错误识别或混淆。在我们的研究中,我们试图找出这些替换如何在正面和侧面面部图像的情绪识别中表现出来。用替换矩阵表示情感识别中的替换结果。研究的第二部分重点是通过对面部特征观察和注视时间的分析来证实这些结果。在面部特征的分析中,考虑了最引人注意的三个主要面部特征(眼睛、嘴巴和带鼻的前额)。对这些面部特征的注视时间也进行了测量。研究设备的基础是眼动仪,我们用它来定义分析的兴趣区域(AOI)。面部特征比例的观察结果证实,在正面面部图像中,恐惧和惊讶、愤怒和厌恶以及部分恐惧和厌恶情绪的替代程度相对较大。在侧面面部图像中,最常见的错误识别是快乐和惊讶、愤怒和厌恶、恐惧和厌恶以及愤怒和悲伤。由于侧脸图像中人脸的信息比正面面部图像少,结果也证实了侧脸图像中错误识别的比例更高,因此更难以识别侧面面部图像中的情绪。注视时间的结果也证实了更大程度的错误识别。两种结果(面部特征的观察比例和注视时间)也以图表形式呈现。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
自引率
0.00%
发文量
0
期刊最新文献
Defining cost-effective workflow for a photorealistic 3D character based on a real person using a metahuman framework Production of ink containing thermochromic dyestuff Influence of the difference electric charge energy transfer ink to paper in digital printing Development of antifungal packaging coating for bread Development of new UV LED curable inkjet varnishes
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1