{"title":"Facial Expressions and Body Postures Emotion Recognition based on Convolutional Attention Network","authors":"T. Zhou, Shiru Gao, Yuanhao Mei, Ling Wang","doi":"10.1109/cits52676.2021.9618520","DOIUrl":null,"url":null,"abstract":"Emotion recognition plays an important role in the fields of medical care, education, services, and public safety. In the video, the emotion could be recognized through facial expressions and body postures. In this paper, we proposed the ER-FLS (Emotion Recognition based on Facial Landmark and Skeleton) Model, which could recognize emotions through the combination of the skeleton and facial landmarks. The model has a lightweight network structure and could focus on the key areas of face and skeleton landmarks with an attention mechanism. By calculating the similarity between global and local features, and update the weights, the recognition accuracy could be enhanced. The experimental analysis proved that the ER-FLS Model achieves 90.63% accuracy of emotional recognition.","PeriodicalId":211570,"journal":{"name":"2021 International Conference on Computer, Information and Telecommunication Systems (CITS)","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2021-11-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2021 International Conference on Computer, Information and Telecommunication Systems (CITS)","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/cits52676.2021.9618520","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
Emotion recognition plays an important role in the fields of medical care, education, services, and public safety. In the video, the emotion could be recognized through facial expressions and body postures. In this paper, we proposed the ER-FLS (Emotion Recognition based on Facial Landmark and Skeleton) Model, which could recognize emotions through the combination of the skeleton and facial landmarks. The model has a lightweight network structure and could focus on the key areas of face and skeleton landmarks with an attention mechanism. By calculating the similarity between global and local features, and update the weights, the recognition accuracy could be enhanced. The experimental analysis proved that the ER-FLS Model achieves 90.63% accuracy of emotional recognition.