Holistic Figure Skating Classification Using Pose-Estimated Videos and Convolutional Neural Networks

DOI

10.2991/978-2-38476-384-9_109How to use a DOI?

Keywords

Figure skating; pose estimation; convolutional neural networks; holistic classification; sports analytics; performance analysis; machine learning

Abstract

Figure skating is a sport where the overall impression of a performance is paramount. While individual elements like jumps and spins are scored, the holistic quality of the program, including artistry, choreography, and skater’s interpretation of the music, significantly impacts the final score. This research proposes a novel approach to classify figure skating programs holistically using pose-estimated videos and convolutional neural networks (CNNs). By converting videos into pose estimations, we capture the essential movement patterns of the skaters while reducing computational complexity. A CNN is then trained on these pose estimations to classify programs into different categories, such as style, level of difficulty, or competitive success. This approach could provide valuable insights for coaches, judges, and skaters themselves, helping them to better understand the nuances of successful performances and potentially leading to more objective and comprehensive judging criteria. Additionally, this research could pave the way for automated scoring systems that consider the holistic aspects of figure skating, complementing existing systems that focus on individual elements.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Hanshu Wang
PY  - 2025
DA  - 2025/04/03
TI  - Holistic Figure Skating Classification Using Pose-Estimated Videos and Convolutional Neural Networks
BT  - Proceedings of the 2024 3rd International Conference on Educational Science and Social Culture (ESSC 2024)
PB  - Atlantis Press
SP  - 961
EP  - 967
SN  - 2352-5398
UR  - https://doi.org/10.2991/978-2-38476-384-9_109
DO  - 10.2991/978-2-38476-384-9_109
ID  - Wang2025
ER  -