TY - GEN
T1 - Image Captioning in Chinese and Its Application for Children with Autism Spectrum Disorder
AU - Zhang, Bin
AU - Zhou, Lixin
AU - Song, Sifan
AU - Chen, Lifu
AU - Jiang, Zijian
AU - Zhang, Jiaming
N1 - Funding Information:
This work is supported by the Shenzhen Science and Technology Innovation Commission, with a fundamental research grant JCYJ20170410172100520 and a grant 2019-INT020. We thank the Smart Children Education Center, Shenzhen, China and their healthcare professionals for providing facilitation and assistance to our facilitators in this preliminary test with our robot.
Publisher Copyright:
© 2020 ACM.
PY - 2020/2/15
Y1 - 2020/2/15
N2 - This research looks into the applications of image captioning in Chinese. In order to improve abilities of children with Autism Spectrum Disorder (ASD) in spontaneous language and turn-taking, during interacting with them, rehabilitation robots are used to track their attention and to describe attractive objects and scenes. This method has three advantages. First, the robots may attract the attention of children with ASD by describing the objects that of interest to them. Second, the robots may improve their ability of spontaneous language which is beneficial to enhancing superior language expressions and social communication with the robots or human beings. Third, describing objects that are interesting may develop the ability in picture description and may further improve their cognitive ability in rehabilitation training. The present study equips a socially interactive robot designed specifically for children with ASD with a dense image captioning approach. The contributions of this article are three-fold. First, an image captioning algorithm was developed to output Chinese description, whereas algorithms in previous studies were mostly catered for English description. Second, image captioning was adopted for the first time in the field of socially interactive robots. Third, our study demonstrates its potential in applying image captioning in robot-enhanced therapy, especially for Chinese children with ASD.
AB - This research looks into the applications of image captioning in Chinese. In order to improve abilities of children with Autism Spectrum Disorder (ASD) in spontaneous language and turn-taking, during interacting with them, rehabilitation robots are used to track their attention and to describe attractive objects and scenes. This method has three advantages. First, the robots may attract the attention of children with ASD by describing the objects that of interest to them. Second, the robots may improve their ability of spontaneous language which is beneficial to enhancing superior language expressions and social communication with the robots or human beings. Third, describing objects that are interesting may develop the ability in picture description and may further improve their cognitive ability in rehabilitation training. The present study equips a socially interactive robot designed specifically for children with ASD with a dense image captioning approach. The contributions of this article are three-fold. First, an image captioning algorithm was developed to output Chinese description, whereas algorithms in previous studies were mostly catered for English description. Second, image captioning was adopted for the first time in the field of socially interactive robots. Third, our study demonstrates its potential in applying image captioning in robot-enhanced therapy, especially for Chinese children with ASD.
KW - Autism Spectrum Disorder
KW - Deep Learning
KW - Image Captioning
KW - Rehabilitation Robots
KW - Rehabilitation Training
UR - http://www.scopus.com/inward/record.url?scp=85085948661&partnerID=8YFLogxK
U2 - 10.1145/3383972.3384072
DO - 10.1145/3383972.3384072
M3 - Conference Proceeding
AN - SCOPUS:85085948661
T3 - ACM International Conference Proceeding Series
SP - 426
EP - 432
BT - Proceedings of the 2020 12th International Conference on Machine Learning and Computing, ICMLC 2020
PB - Association for Computing Machinery
T2 - 12th International Conference on Machine Learning and Computing, ICMLC 2020
Y2 - 15 February 2020 through 17 February 2020
ER -