Masked modeling-based action event prediction considering bidirectional time-series in soccer

Ryota Goka; Yuya Moroto; Keisuke Maeda; Takahiro Ogawa; Huang-Chia Shih; Miki Haseyama

doi:10.1117/12.3019688

2 May 2024 Masked modeling-based action event prediction considering bidirectional time-series in soccer

Ryota Goka, Yuya Moroto, Keisuke Maeda, Takahiro Ogawa, Huang-Chia Shih, Miki Haseyama

Proceedings Volume 13164, International Workshop on Advanced Imaging Technology (IWAIT) 2024; 1316434 (2024) https://doi.org/10.1117/12.3019688
Event: International Workshop on Advanced Imaging Technology (IWAIT) 2024, 2024, Langkawi, Malaysia

Abstract

This paper presents an action event prediction based on masked modeling for bidirectional time-series analysis in soccer. Since optimal action events should be selected based on changes in match situations, it is important to consider bidirectional time-series changes in data. To predict the next action event with the consideration of the bidirectional time-series, the proposed method learns the contexts of action event sequences by predicting the masked action events from the preceding and following contexts. The prediction accuracy of our method is improved from that of the unidirectional method, which shows the effectiveness of taking the bidirectional time series into account in soccer.

(2024) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Ryota Goka, Yuya Moroto, Keisuke Maeda, Takahiro Ogawa, Huang-Chia Shih, and Miki Haseyama "Masked modeling-based action event prediction considering bidirectional time-series in soccer", Proc. SPIE 13164, International Workshop on Advanced Imaging Technology (IWAIT) 2024, 1316434 (2 May 2024); https://doi.org/10.1117/12.3019688

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available