A Real-time 3D Motion Detection System Based on A Monocular Camera

Stroke, being the second most prevalent cause of mortality and the third leading contributor to disability, imposes a substantial financial burden on patients. In-home rehabilitation, as an emerging technology, provides healthcare professionals with access to time-sensitive data while enabling patients to undergo testing from their residences. A monocular camera motion detection system is proposed in this investigation to provide real-time and accurate motion information to patients and doctors to carry out rehabilitation treatment by detecting 3D motion keypoints. The experimental results show that the performance of the proposed system is within the satisfactory range in the model performance test and the comparison experiment with Kinect. The presented detection system can be utilized for initial assessment of patients in telerehabilitation systems, which is of significance to the rehabilitation of patients.