Human Motion Tracking Using 3D Image Features with a Long Short-Term Memory Mechanism Model-An Example of Forward Reaching.

Kai-Yu Chen,Hui-Min Lee,Li-Wei Chou,Yi-Shu Zhou,Ying-Hui Lai,Shih-Tsang Tang,Cheng-Hung Lin,Shuenn-Tsong Young

doi:10.3390/s22010292

Abstract

Human motion tracking is widely applied to rehabilitation tasks, and inertial measurement unit (IMU) sensors are a well-known approach for recording motion behavior. IMU sensors can provide accurate information regarding three-dimensional (3D) human motion. However, IMU sensors must be attached to the body, which can be inconvenient or uncomfortable for users. To alleviate this issue, a visual-based tracking system from two-dimensional (2D) RGB images has been studied extensively in recent years and proven to have a suitable performance for human motion tracking. However, the 2D image system has its limitations. Specifically, human motion consists of spatial changes, and the 3D motion features predicted from the 2D images have limitations. In this study, we propose a deep learning (DL) human motion tracking technology using 3D image features with a deep bidirectional long short-term memory (DBLSTM) mechanism model. The experimental results show that, compared with the traditional 2D image system, the proposed system provides improved human motion tracking ability with RMSE in acceleration less than 0.5 (m/s2) X, Y, and Z directions. These findings suggest that the proposed model is a viable approach for future human motion tracking applications.

Highlights

Our main contributions of this study are as follows: (1) We investigate whether the time-related deep learning model structure, deep bidirectional long short-term memory (DBLSTM), is advantageous compared to several traditional models in human motion tracking tasks
coefficient of multiple correlations (CMC) > 0.9 and percent root mean square error (PRMSE) < 5%, which means that using deep learning (DL) methods can obtain good performance in human motion tracking tasks
This study proposed a DL-based system to track human motion using 3D image features with a DBLSTM model

Summary

Introduction

Rehabilitation is becoming an increasingly important issue owing to the rise in elderly population. According to the World Population Ageing 2020 report [1], 727 million people were aged 65 years or older around the world, and the number of the elderly was projected to rise to 1.5 billion by 2050. The World Health Organization (WHO) indicates that 15 million people suffer from stroke each year [2], with 75% of those being elderly [3]. This means that healthcare, such as motion rehabilitation to regain motor function, must be valued and improved

Objectives

Methods

Results

Conclusion