A Remote Exercise Training, Monitoring And Assessment Based On Virtual Reality Environment And Depth Sensor

Today, the use of technology has an effect on people s lives and from a physical activity point of view, it causes the lack of physical activity and adverse outcomes in health. We proposed an exercise training platform under monitoring a virtual trainer. Also, the platform can assess exercise performance based on virtual reality environment and depth sensor. This platform consists of a depth sensor, exercise database, virtual environment, avatars and processing tools (training and monitoring). For depth sensor, we used Kinect to acquire posture, position, and motion data. Exercise database was collected from an expert trainer as an offline process but for training and monitoring, the platform performs as an online process. V-realm Builder, Poser and Matlab were used to model, edit, design and display virtual reality environment. Mean square error is a comparison criterion between user movement and virtual trainer for monitoring. For this study, different avatars were created with a variety of age and gender that user can select an appropriate choice to interact more and better with the virtual environment. This platform can be used in everywhere and for all people with different ages and physical conditions. In addition, movements can be watched on the screen as a mirror. Furthermore, the platform in this study is a low-cost approach