A Comparison of Dual-Kinect and Vicon

The robotics community is increasingly making use of human motion capture and joint tracking analysis. This study aims to assess the potential of employing a camera system to study upper-body joint motions by comparing the joint tracking abilities of a dual-Kinect system and a Vicon 3D Motion Capture system. Such motion data can be used to train a robot with data collected from humans performing representative motions. An example of such robotic applications is shown in Figure 1, which shows a robot being used in place of a police officer to direct pedestrians and traffic. Note that work related musculoskeletal disorders (WRMSDs) are a major issue plaguing traffic policemen and factory workers [1,2]. Muscular fatigue is induced due to long working hours, as well as incorrect or sub-optimal motion techniques [3]. Vicon systems use high definition cameras which, while being accurate, are expensive and have mobility constraints. On the other hand, the Microsoft Kinect V2 is a non-invasive, low-cost camera primarily used in the video game industry which can be used for human-body motion analysis. A primary differentiating factor between Kinect and Vicon is the necessity of retroreflective markers in the Vicon system. However, the Kinect does not require markers for human-body tracking because a proprietary Microsoft Software Development Kit (SDK) possesses the ability to track human body joints. Furthermore, the Vicon system is composed of multiple cameras set up around the perimeter of the measurement workspace. On the other hand, the Kinect is a single sensor package that only sees what is directly in front of it.