arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2212.06009cs.CV
收藏

一种改进的嘴部情感自动识别方法

An Approach for Improving Automatic Mouth Emotion Recognition

University of Florence(佛罗伦萨大学);University of Perugia(佩鲁贾大学);Sapienza University of Rome(罗马第一大学)

Giulio Biondi, Valentina Franzoni, Osvaldo Gervasi, Damiano Perri

摘要:

该研究提出并测试了一种通过卷积神经网络(CNN)进行嘴部检测的自动情绪识别技术,旨在应用于支持具有沟通技能问题(例如肌肉萎缩,中风,自闭症,或者更简单地说,疼痛)的健康障碍患者,以识别情绪并生成实时反馈或数据馈送支持系统。软件系统开始计算,识别所获取的图像上是否存在面部,然后查找嘴部位置并提取相应的特征。这两个任务都是使用基于Haar分类器的分类器执行的,这保证了快速执行和有前途的性能。如果我们以前的工作集中在视觉微表情对单个用户的个性化训练,这种策略的目的是训练系统也广义的人脸数据集。

英文摘要:

The study proposes and tests a technique for automated emotion recognition through mouth detection via Convolutional Neural Networks (CNN), meant to be applied for supporting people with health disorders with communication skills issues (e.g. muscle wasting, stroke, autism, or, more simply, pain) in order to recognize emotions and generate real-time feedback, or data feeding supporting systems. The software system starts the computation identifying if a face is present on the acquired image, then it looks for the mouth location and extracts the corresponding features. Both tasks are carried out using Haar Feature-based Classifiers, which guarantee fast execution and promising performance. If our previous works focused on visual micro-expressions for personalized training on a single user, this strategy aims to train the system also on generalized faces data sets.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑