Hugging Rain Man (HRM)
收藏资源简介:
Hugging Rain Man (HRM) 数据集是由华中师范大学人工智能教育学院创建的,专门用于分析自闭症谱系障碍(ASD)儿童的非典型面部表情。该数据集包含约130,000帧由FACS专家手动标注的面部动作单元(AUs)和动作描述符(ADs),涵盖了自闭症儿童和典型发展(TD)儿童的摆拍和自发面部表情。数据集的创建过程包括多维度的标注框架,涵盖22个AUs、10个ADs和非典型性评分。HRM数据集的应用领域主要集中在ASD的早期筛查和面部表情分析,旨在通过客观的面部特征分析,提高对ASD儿童面部表情特征的理解,并为ASD的早期诊断提供潜在工具。
The Hugging Rain Man (HRM) dataset was created by the School of Artificial Intelligence Education at Central China Normal University, specifically for the analysis of atypical facial expressions in children with Autism Spectrum Disorder (ASD). This dataset comprises approximately 130,000 frames of facial Action Units (AUs) and Action Descriptors (ADs) manually annotated by FACS experts, covering posed and spontaneous facial expressions from both autistic children and typically developing (TD) children. The construction of the HRM dataset employs a multi-dimensional annotation framework that encompasses 22 AUs, 10 ADs, and atypicality scoring. The main application scenarios of the HRM dataset center on early ASD screening and facial expression analysis, aiming to advance the understanding of facial expression characteristics of children with ASD through objective facial feature analysis, and provide potential tools for the early diagnosis of ASD.




