TY - JOUR AU - Madaras, Martin AU - Riečický, Adam AU - Mesároš, Michal AU - Piovarči, Michal AU - Ďurikovič, Roman PY - 2019 DA - 2019// TI - Position Estimation and Calibration of Inertial Motion Capture Systems Using Single Camera JO - Journal of Virtual Reality and Broadcasting VL - 15(2018) IS - 3 KW - Motion Capture KW - glove skeleton calibration KW - inertial suit KW - optical system KW - optical-inertial synchronization KW - silhouette tracking AB - The paper proposes a hybrid system for position estimationof a motion capture suit and gloves as wellas a method for an automatic skeleton calibration formotion capture gloves. The skeleton calibration workswith a single image scan of the hand where the skeletonis fitted. The position estimation is based on a synchronizationof an inertial motion capture system anda single camera optical setup. The proposed synchronizationuses an iterative optimization of an energy potentialin image space, minimizing the error betweenthe camera image and a rendered virtual representationof the scene. For each frame, an input skeletonpose from the mocap suit is used to render a silhouetteof a subject. Moreover, the local neighborhood aroundthe last known position is searched by matching thesilhouette to the distance transform representation ofthe camera image based on Chamfer matching. Usingthe combination of the camera tracking and the inertialmotion capture suit, it is possible to retrieve the position of the joints that are hidden from the camera view.Using the proposed hybrid technique, it is possible tocapture the position even if it cannot be captured bythe suit sensors. Our system can be used for both realtimetracking and off-line post-processing of alreadycaptured mocap data. SN - 1860-2037 UR - http://nbn-resolving.de/urn:nbn:de:0009-6-48214 DO - 10.20385/1860-2037/15.2018.3 ID - madaras2019 ER -