mnmDTW: An extension to Dynamic Time Warping for Camera-based Movement Error Localization
This addresses the need for precise error localization in exercise analysis, but it is incremental as it extends an existing method.
The paper tackled the problem of localizing movement errors in exercise videos by using a modified Dynamic Time Warping (DTW) method to measure deviations from a gold standard, showing that mistakes are clearly visible and localizable.
In this proof of concept, we use Computer Vision (CV) methods to extract pose information out of exercise videos. We then employ a modified version of Dynamic Time Warping (DTW) to calculate the deviation from a gold standard execution of the exercise. Specifically, we calculate the distance between each body part individually to get a more precise measure for exercise accuracy. We can show that exercise mistakes are clearly visible, identifiable and localizable through this metric.