Human Gait Analysis and Activity Recognition: A Review
Neha Gaud, Maya Rathore, Ugrasen Suman
Abstract
Gait refers the walking pattern of human being. It is the manifestation of change in the joint angles of lower extremity. Locomotion enables the person to perform the daily life activities, to maneuver over the discontinue surfaces. In this paper, we are presenting the systematic survey of bipedal legged locomotion and activity recognition. Human walking plays crucial role in our daily life activities. It provides the much needed mobility to perform daily life activities efficiently. Recognition of human activity is an intriguing issue that can be solved in a variety of ways. The majority of researchers recommended effective deep learning techniques such as CNN, inception CNN, LSTM, Bi-LSTM, and hybrid approaches. Human walk pattern reflects the health condition of any person. Various clinical examinations are performed through human walk to check their clinical success, post treatment recovery and rehabilitation. It also helps early diagnosis of various walking impairment at early stage. The main objective of this research is to review the different state of art gait analysis & activity recognition approaches for human health issues using various wearable sensors and other techniques. The discussion held in his paper would help the coming researchers in the field of human gait analysis and activity recognition.