Fast segmentation of exercise repetitions enabling real-time patient analysis and feedback

We present a segmentation algorithm capable of segmenting exercise repetitions in real-time. This approach does not require training data and can use simulated or movement data from a healthy person to serve as the template. This approach is invariant to low range of motion, instability in movements and sensor noise while remaining selective to different exercises. This algorithm enables responsive feedback for technology-assisted physical rehabilitation systems. We evaluate the algorithm against a healthy population and stroke survivor population performing stroke rehabilitation exercises captured on a consumer level depth sensor. We show the algorithm can consistently achieve correct and fast segmentation.