Low-cost electromyograph combined with markerless pose detection

Abstract This papers presents a low-cost electromyograph combined with marker-less pose detection using computer vision. The developed and build three channel is tested by measuring the muscle activity of one leg, while subject performing squats. Simultaneously, camera records exercise subsequently image data evaluated OpenPose. We could show that this simple setup enables user to evaluate independent muscles as function knee angle. These results are in good agreement expected activity. sample-rate EMG device 2 kHz. overall cost under 100 €. To our knowledge, first work combining these two methods for dynamic exercises. method well customizable other sports due battery powered its handy size.