A review of machine learning network in human motion biomechanics

Abstract

Human motion analysis is fundamental in many real applications such as surveillance and monitoring, human-machine interface, medical motion analysis and diagnosis. With the increasing amount of data in biomechanics research, it is becoming increasingly important to automatically analyse and understand object motions from large amount of footage and sensor data. The modalities for capturing the gait data are grouped according to the sensing technology: video sequences, wearable sensors, and floor sensors, as well as the publicly available datasets. In order to extract the essence of these data and make research more efficient, modern machine learning techniques are starting to complement traditional statistical tools. The purpose of this review is to familiarise the readers with key directions of implementation of machine learning techniques for gait analysis. The essential human gait parameters are briefly reviewed, followed by a detailed review of the-state-of-the art in machine learning for the human gait analysis. The machine learning framework used for human analysis, such as support vector machine (SVM), Hidden Markov Model (HMM), Bayesian Network Classifier (BN), Convolutional Neural Network (CNN), Recurrent Neural Network (RNN), Long-Short Term Memory (LSTM) and Generative Adversarial Networks (GANs), shall too be discussed here. Finally, the challenges and future direction of machine learning’s application in motion analysis are outlined and discussed.