What is it about?
We provide a dataset of speed climbing performances represented as skeleton time series data. We propose and evaluate similarity measures inspired in Dynamic Time Warping (DTW) to effectively match speed climbing runs for similarity.
Featured Image
Photo by roya ann miller on Unsplash
Why is it important?
Current research focuses on precise sensor data that are hardly applicable in official sports events and competitions. We showcase a proof of concept that justifies the usability of the 2D skeleton time series obtained from an ordinary video.
Perspectives
We see great potential for further extensions of this work, by visually analyzing nuances in pair of speed climbing performances, e.g., to identify weak (major delays, inefficient trajectory) and strong parts of the climb, comparison of arbitrary performance with a world record, etc.
Petr Elias
Masarykova Univerzita
Read the Original
This page is a summary of: SPEED21: Speed Climbing Motion Dataset, October 2021, ACM (Association for Computing Machinery),
DOI: 10.1145/3475722.3482795.
You can read the full text:
Resources
Contributors
The following have contributed to this page
