Labeling images with facial emotion and the potential for pediatric healthcare
作者：
Highlights：
• Autism spectrum disorder (ASD) affects 750,000 American Children under the age of 10.
• Emotion classifiers integrated into mobile solutions can be used for screening and therapy.
• Emotion classifiers do not generalize well to children due to a lack of labeled training data.
• We propose a method of aggregating emotive video through a mobile game.
• We demonstrate that several algorithms can automatically label frames from video derived from the game.
摘要
•Autism spectrum disorder (ASD) affects 750,000 American Children under the age of 10.•Emotion classifiers integrated into mobile solutions can be used for screening and therapy.•Emotion classifiers do not generalize well to children due to a lack of labeled training data.•We propose a method of aggregating emotive video through a mobile game.•We demonstrate that several algorithms can automatically label frames from video derived from the game.
论文关键词：Mobile games,Computer vision,Autism,Emotion,Emotion classification
论文评审过程：Received 26 April 2018, Revised 14 May 2019, Accepted 24 June 2019, Available online 6 July 2019, Version of Record 1 August 2019.
论文官网地址：https://doi.org/10.1016/j.artmed.2019.06.004
