English

mnmDTW: An extension to Dynamic Time Warping for Camera-based Movement Error Localization

Computer Vision and Pattern Recognition 2023-10-16 v1 Artificial Intelligence

Abstract

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.

Keywords

Cite

@article{arxiv.2310.09170,
  title  = {mnmDTW: An extension to Dynamic Time Warping for Camera-based Movement Error Localization},
  author = {Sebastian Dill and Maurice Rohr},
  journal= {arXiv preprint arXiv:2310.09170},
  year   = {2023}
}

Comments

Poster Prague 2023 Conference, 4 pages

R2 v1 2026-06-28T12:49:58.114Z