A Depth Camera Motion Analysis Framework for Tele-rehabilitation: Motion Capture and Person-Centric Kinematics Analysis

Abstract

With increasing importance given to telerehabilitation, there is a growing need for accurate, low-cost, and portable motion capture systems that do not require specialist assessment venues. This paper proposes a novel framework for motion capture using only a single depth camera, which is portable and cost effective compared to most industry-standard optical systems, without compromising on accuracy. Novel signal processing and computer vision algorithms are proposed to determine motion patterns of interest from infrared and depth data. In order to demonstrate the proposed framework's suitability for rehabilitation, we developed a gait analysis application that depends on the underlying motion capture subsystem. Each subject's individual kinematics parameters, which are unique to that subject, are calculated and these are stored for monitoring individual progress of the clinical therapy. Experiments were conducted on 14 different subjects, 5 healthy and 9 stroke survivors. The results show very close agreement of the resulting relevant joint angles with a 12-camera based VICON system, a mean error of at most 1.75% in detecting gait events w.r.t the manually generated ground-truth, and significant performance improvements in terms of accuracy and execution time compared to a previous Kinect-based system.

Gait phase classification for in-home gait assessment

“…Experiments are conducted on 9 stroke patients and 6 health volunteers during a 4-meter walking test, resulting in 126 experiments with various walking speed direction and patterns. We acquire estimated trajectories of hip, knee, ankle, toes, heels joints using the motion capture system of [23].…”

“…Fig. 1 shows the gait events (names are in red on top of the skeletons) based on the trajectories of shoulder, hip, knee, ankle, toe, heel, toe1 & heel1 (opposite side) joints (shown as cycles) captured by a Kinect-based optical motion tracking system [23]. A typical gait phase representation plot is shown at the bottom of the figure based on the ratio of knee angle, where a unique colour is assigned to each gait phase (starting with P1=Loading Re-sponse1 to P9=Terminal Swing) during the gait cycle.…”

“…Based on trajectories of the joints of interest, obtained, for example, by tracking the markers placed on the joints in the recorded video, as in [23], gait parameters are calculated in each frame. For the jth gait cycle, let…”

“…Though we provide experimental results using the Microsoft Kinect v2-based motion capture system of [23], the proposed algorithms can be used with other motion capture systems as they only require as input accurate joint trajectories.…”

A Modular Data Glove System for Finger and Hand Motion Capture Based on Inertial Sensors

“…Two types of hand motion capture technologies have been reported in the literature: non-contact systems and contact systems [1-7, 9, 11, 12, 15]. Typical non-contact systems capture hand motions by using camera-based devices and image processing technology [9,11,15]. The users are not required to wear any devices, but such systems are easily affected by environmental conditions, such as illumination and occlusion.…”