Deep learning in Human Gait Recognition: An Overview

Abstract

One of the rudimentary forms of human locomotion is Gait that roves the center of mass (COM) of the body in different directions. The human gait is comprised of details of the personals that consist of movements related to patterns and intentions. The detection of human gait is through either wearable sensors or an electromyography signal that shows the most promising potential as a therapeutic method. While analyzing Gait, a large number of gait parameters are required which are interdependent, and this makes it difficult to interpret as a large number of parameters have a large amount of data. These data are mostly collected from various clinical laboratories. The current paper discusses human gait and the recognition of these events based on various deep learning models. Adding to this, issues and challenges that are related to Gait are also elaborated with prominent techniques used in gait recognition.