Gait Parameter Monitoring Using A Wearable Stereo Camera System

Abstract

An infrared (IR) stereo camera is mounted on each bracelet and tasked with tracking a marker on the bracelet on the other ankle. Based on the imaged trajectories of both markers, a computer vision algorithm that we developed estimates the length, width, and height of the gait of a person. A prototype of the proposed gait monitoring system has been tested in a laboratory setting. It is observed that the system achieves an average accuracy of 97.43%, 97.19%, and 96.11% in the measurements of gait length, width, and height, respectively. Compared to state-of-the-art gait sensing systems, the design presented in this work promises lower cost and ease of use, while delivering comparable sensing performance.