VR-based Body Tracking for Homecare Training

Training helps to maintain and improve sufficientmuscle function, body control, and body coordination. Theseare important to reduce the risk of fracture incidents causedby falls, especially for the elderly or people recovering frominjury. Virtual reality training can offer a cost-effective and in-dividualized training experience. We present an application forthe HoloLens 2 to enable musculoskeletal training for elderlyand impaired persons to allow for autonomous training andautomatic progress evaluation. We designed a virtual downhillskiing scenario that is controlled by body movement to stimu-late balance and body control. By adapting the parameters of theski slope, we can tailor the intensity of the training to individualusers. In this work, we evaluate whether the movement data ofthe HoloLens 2 alone is sufficient to control and predict bodymovement and joint angles during musculoskeletal training. Werecord the movements of 10 healthy volunteers with externaltracking cameras and track a set of body and joint angles of theparticipant during training. We estimate correlation coefficientsand systematically analyze whether whole body movement canbe derived from the movement data of the HoloLens 2. No par-ticipant reports movement sickness effects and all were able toquickly interact and control their movement during skiing. Ourresults show a high correlation between HoloLens 2 movementdata and the external tracking of the upper body movement andjoint angles of the lower limbs