FROM TRADITIONAL TO INTERDISCIPLINARY APPROACHES FOR INERTIAL BODY MOTION CAPTURE
Keywords: inertial motion capture, model-based sensor fusion, egocentric vision, optimal control simulation, human kinetics
AbstractInertial motion capture (mocap) is a widespread technology for capturing human motion outside the lab, e.g. for applications in sports, ergonomics, rehabilitation and personal fitness. Even though mature systems are commercially available, inertial mocap is still a subject of research due to a number of limitations: besides measurement errors and sparsity, also simplified body models and calibration routines, soft tissue artefacts and varying body shapes lead to limited precision and robustness compared to optical gold standard systems. The goal of the research group wearHEALTH at the TU Kaiserslautern is to tackle these challenges by bringing together ideas and approaches from different disciplines including biomechanics, sensor fusion, computer vision and (optimal control) simulation. In this talk, we will present an overview of our approaches and applications, starting from the more traditional ones.
Authors can retain copyright, while granting the International Society of Biomechanics in Sports (ISBS) the right of first publication.