[3P5-OS-16b] OS-16 (2)

[3P5-OS-16b-04] Estimation of Participation Rate in Group Activities Based on posture Distribution during Childcare Activities

〇Ryo Aoyagi¹, Masahiro Miyata¹, Tetsuji Yamada¹, Tomoaki Nakamura², Takashi Omori¹ (1. Tamagawa University , 2. The University of Electro-Communications)

Keywords:Machine learning, posture, Probability distribution

There is a big difference in the growth of children from individual to individual. Some of the children need special support from experts. However, there is no objective and qualitative method available currently to find the ones who need help, and we still depend on human experience and knowledge. A challenge of behavioral science is the development of an alternative method that is based on the behavioral analysis data to understand children’s developmental state not depending on the human experience. However, most of the current engineering methods use specialized equipment that are attached to human body for observation. But they cause a problem when applied to children. So, in this study, we aim to develop a method to record children's daily activities with a monocular camera and detect children who need special care using a posture probability distribution.