Knee Up: an Exercise Game for Standing Knee Raises by Motion Capture with RGB-D Sensor

Abstract

In this work, we present the design and the implementation of Knee Up, an exergame that promotes knee health via standing knee raises exercises. It allows users to exercise at home and perform the exercise in a gaming environment. By motion capture using RGB-D data, estimated positions of the user's skeletal joints are acquired and processed in real-time. For the game, 3D virtual environments have been created with gamification elements so that users can enjoy more interactive and engaging exercise sessions. The game facilitates to evaluate the quality of performed standing knee raises exercise through a rule-based recognition algorithm and, in return, to provide timely feedback. A formal user study was conducted in order to assess the usability and gamification aspects of Knee Up. Study results demonstrate that Knee Up is generally well-received in terms of usability, engagement, ease of learning-to-play, and exercise sustainability; and validate that the rule-based recognition algorithm works satisfactorily well.