Stability of Kinect for range of motion analysis in static stretching exercises

Physical rehabilitation aims people with physical impairments to enhance and restore their functional ability. The Microsoft Kinect v1 and v2 technologies apply depth information and machine vision techniques to generate 3D coordinates of a set of anatomical landmarks on the human body regarded as Kinect joints. Trigonometry relationship between Kinect joints can be used to extract body Range of Motion (ROM). The purpose of this study was to evaluate stability of Kinect for ROM measurement during static stretching exercises. According to the literature, the stability of Kinect in static exercises has been reported to a limited extent. 13 healthy men participated in this study and performed 5 exercises in 2 different distances from the cameras. Exercises were recorded by Kinect v1 and Kinect v2, concurrently. The stability of Kinect was also evaluated for 5 ROMs including: elbow flexion, shoulder abduction, wrist pronation, wrist flexion, and wrist ulnar deviation. Maximum and average joint displacement errors were used for stability analysis. Results showed that Kinect v2 is more stable compared to Kinect v1. Kinect v2 joints showed displacement error of more than 15 mm for wrist. For the other joints, Kinect showed an average displacement error of less than 10 mm.