Motion Capture Data Analysis in the Instantaneous Frequency-Domain Using Hilbert-Huang Transform.

Ran Dong,Soichiro Ikuno,Dongsheng Cai

doi:10.3390/s20226534

Abstract

Motion capture data are widely used in different research fields such as medical, entertainment, and industry. However, most motion researches using motion capture data are carried out in the time-domain. To understand human motion complexities, it is necessary to analyze motion data in the frequency-domain. In this paper, to analyze human motions, we present a framework to transform motions into the instantaneous frequency-domain using the Hilbert-Huang transform (HHT). The empirical mode decomposition (EMD) that is a part of HHT decomposes nonstationary and nonlinear signals captured from the real-world experiments into pseudo monochromatic signals, so-called intrinsic mode function (IMF). Our research reveals that the multivariate EMD can decompose complicated human motions into a finite number of nonlinear modes (IMFs) corresponding to distinct motion primitives. Analyzing these decomposed motions in Hilbert spectrum, motion characteristics can be extracted and visualized in instantaneous frequency-domain. For example, we apply our framework to (1) a jump motion, (2) a foot-injured gait, and (3) a golf swing motion.

Highlights

A motion capture system records a finite number of marking positions in time near the important human joints and convert them into their joint angles approximately using a simple human skeletal system
Our results show that complicated motions can be decomposed into several distinct motion primitives such as jump motion, a gait of a foot-injured subject, and a golf swing motion in the instantaneous frequency-domain using our proposed framework
The intrinsic mode function (IMF) can decompose the chromatic signal into pseudo monochromatic signals

Summary

Introduction

A motion capture system records a finite number of marking positions in time near the important human joints and convert them into their joint angles approximately using a simple human skeletal system. There have been numerous researches on motion analysis based on motion capture data in the different research fields [1]. Kim et al [2] presented an analysis system and algorithm for golf swing based on an inertial sensor for sports motion analysis. Hssayeni et al [3] developed an analysis method for estimating Parkinsonian tremor using both gradient tree boosting ensemble model and long short-term memory (LSTM) deep learning model. Motion primitives are essential for deploying human activities [4], because intricated human movements are generated by combining and sequencing these motion primitives [5]

Methods

Discussion

Conclusion