Using close range photogrammetric method to estimate kinetic variables in olympic-style weightlifting

Olympic-style weightlifting, or weightlifting, is an athletic discipline that has as a goal lift the heaviest amount of weight attached a weight set. Weightlifting complements other athletic disciplines to increase power and improve performance. Weight training exercises are also an important part of physical rehabilitation from muscle, joint, tendons, and ligaments injuries. To determinate its efficiency, optimize it, prevent lesions, or help in the rehab process indirect and direct methods are commonly used. Direct methods use data from muscular signals to estimate force, torque, and kinematic variables whereas indirect methods like as close range photogrammetric get data directly from a video camera. In this paper, we present a non-invasive system to recover the positions and motion pathways of weight set during lifting moment. Active and passive tracking marks are used to leverage the detection of points of interest. Kinematic parameters were collected and analyzed using moving pictures analysis system implemented in Processing and using OpenCV (Open Source Computer Vision) framework. The preliminary results indicate that our close range photogrammetric method is able to measure kinematic variable in weightlifting during physical or rehab training as well as during competition. Our experiment shows the error in the position measure is less than 2.24% in average.