[3G5-OS-24b] 日常生活知識とAI

[3G5-OS-24b-01] Indoor injury risk visualization with posture generation of a child digital human mode

Fumiya Endo², 〇Natsuki Miyata¹, Yusuke Maeda² (1. National Institute of Advanced Industrial Science and Technology, 2. Yokohama National University)

Keywords:Digital Human, Injury prevention, Visualization

This study develops a system to visualize areas where accidents can occur due to children’s reach in indoor
environments, using the posture estimation function of digital human models. As statistical behavioral dimensions
such as children’s reach and climbable height are insufficient for evaluating actual complicated environment such as
home, we have introduced a digital human model that reflects the physical characteristics and behavioral strategies
of children and enables evaluation of three-dimensional interaction with the product, which is expected to help
parents to understand and do improve their home environment. Posture of the child digital human model that
represents each age was estimated by solving inverse kinematics that satisfies given standing position and reaching
target. Potential risk areas due to the reach of children were visualized by coloring the environmental model.