| Title |
Analysis on the Mobile Healthcare Behavior using an Artificial Intelligence Based Pose Estimation |
| Authors |
하태용(Taeyong Ha) ; 이후진(Hoojin Lee) |
| DOI |
https://doi.org/10.5573/ieie.2020.57.1.63 |
| Keywords |
AI ; Deep Learning ; CNN ; Fitness ; PoseNet ; |
| Abstract |
Trends in managing personal health using smartphone devices and healthcare-related devices are receiving a lot of attention, but it is not easy to spread them to the real market for some time. Currently, users rely heavily on instructor training through video content or fitness applications. In this study, we use PoseNet of artificial intelligence(AI) convolutional neural network(CNN) based on the data used in the Leeds Sports Pose(LSP) dataset to analyze pose estimation and analyze human coordinates, and to applaud the result as a clap. The development language uses the Node.js program to implement the final momentum results into mobile healthcare. This provides motivation for the behavioral results of pose estimation to deviate from the dependency of intuition and error. Motion analysis based on more quantitative and faster pose estimation through body coordinates is now possible. In this regard, PoseNet can be a model that fits our purpose of developing a personal fitness program, being served as one of the many features of the healthcare platform. |