Comparative Analysis of Deep Learning-Based Action Recognition: The example of Table Tennis

DOI

10.2991/978-94-6463-512-6_29How to use a DOI?

Keywords

Action Recognition; Dual-Stream Models; Graph Convolutional Networks; Transformers

Abstract

With the advancement of deep learning technologies, computer vision has shown unprecedented potential in the field of action recognition. Particularly in table tennis, action recognition technologies not only help athletes improve their techniques but also provide real-time feedback during training and competitions. This study thoroughly investigates existing action recognition methods, including dual-stream models, Graph Convolutional Networks (GCNs), and Transformers, and highlights their applications in movement analysis during table tennis activities. The research focuses on revealing the accuracy and real-time capabilities of action recognition to better support coaches and athletes in understanding sports techniques. Additionally, this work introduces that Baidu has developed models capable of recognizing specific table tennis movements, such as serving and returning, with an accuracy rate exceeding 80%, significantly improving the quality and efficiency of training. This paper also discusses the prospective applications of action recognition technology in other sports, as well as potential challenges in future research. It is expected that these advancements will drive the development of sports technology, helping athletes and coaches achieve higher accomplishments through technological means.

Copyright

© 2024 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  - Xing Long
PY  - 2024
DA  - 2024/09/23
TI  - Comparative Analysis of Deep Learning-Based Action Recognition: The example of Table Tennis
BT  - Proceedings of the 2024 International Conference on Artificial Intelligence and Communication (ICAIC 2024)
PB  - Atlantis Press
SP  - 254
EP  - 267
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-512-6_29
DO  - 10.2991/978-94-6463-512-6_29
ID  - Long2024
ER  -