Title :
Analysis of children´s posture for the bodily kinesthetic test
Author :
Dongjin Lee;Woo-han Yun;Chankyu Park;Hosub Yoon;Jaehong Kim
Author_Institution :
Human Robot Interaction Research Section, Electronics and Telecommunications Research Institute, Deajeon, Korea
Abstract :
In this paper, we present a children´s posture analysis system using the Kinect while they are performing the yoga poses with the Nao robot. We collected posture data from eighteen children who were kindergaten kids and their posture accuracy was annotated by an expert into four different levels. For a feature extraction, we apply a method of an angular skeleton representation in every frame and those features are aggregated with several different functions over the entire sequence of each pose. We show results of the posture classification and analysis.
Keywords :
"Feature extraction","Torso","Robot kinematics","Support vector machines","Skeleton","Real-time systems"
Conference_Titel :
Ubiquitous Robots and Ambient Intelligence (URAI), 2015 12th International Conference on
DOI :
10.1109/URAI.2015.7358886