TY - JOUR
T1 - Soccer match broadcast video analysis method based on detection and tracking
AU - Li, Hongyu
AU - Yang, Meng
AU - Yang, Chao
AU - Kang, Jianglang
AU - Suo, Xiang
AU - Meng, Weiliang
AU - Li, Zhen
AU - Mao, Lijuan
AU - Sheng, Bin
AU - Qi, Jun
N1 - Publisher Copyright:
© 2024 John Wiley & Sons Ltd.
PY - 2024/5
Y1 - 2024/5
N2 - We propose a comprehensive soccer match video analysis pipeline tailored for broadcast footage, which encompasses three pivotal stages: soccer field localization, player tracking, and soccer ball detection. Firstly, we introduce sports camera calibration to seamlessly map soccer field images from match videos onto a standardized two-dimensional soccer field template. This addresses the challenge of consistent analysis across video frames amid continuous camera angle changes. Secondly, given challenges such as occlusions, high-speed movements, and dynamic camera perspectives, obtaining accurate position data for players and the soccer ball is non-trivial. To mitigate this, we curate a large-scale, high-precision soccer ball detection dataset and devise a robust detection model, which achieved the (Formula presented.) of 80.9%. Additionally, we develop a high-speed, efficient, and lightweight tracking model to ensure precise player tracking. Through the integration of these modules, our pipeline focuses on real-time analysis of the current camera lens content during matches, facilitating rapid and accurate computation and analysis while offering intuitive visualizations.
AB - We propose a comprehensive soccer match video analysis pipeline tailored for broadcast footage, which encompasses three pivotal stages: soccer field localization, player tracking, and soccer ball detection. Firstly, we introduce sports camera calibration to seamlessly map soccer field images from match videos onto a standardized two-dimensional soccer field template. This addresses the challenge of consistent analysis across video frames amid continuous camera angle changes. Secondly, given challenges such as occlusions, high-speed movements, and dynamic camera perspectives, obtaining accurate position data for players and the soccer ball is non-trivial. To mitigate this, we curate a large-scale, high-precision soccer ball detection dataset and devise a robust detection model, which achieved the (Formula presented.) of 80.9%. Additionally, we develop a high-speed, efficient, and lightweight tracking model to ensure precise player tracking. Through the integration of these modules, our pipeline focuses on real-time analysis of the current camera lens content during matches, facilitating rapid and accurate computation and analysis while offering intuitive visualizations.
KW - field localization
KW - player tracking
KW - soccer ball detection
KW - video analysis
KW - visualizations
UR - http://www.scopus.com/inward/record.url?scp=85194554558&partnerID=8YFLogxK
U2 - 10.1002/cav.2259
DO - 10.1002/cav.2259
M3 - Article
AN - SCOPUS:85194554558
SN - 1546-4261
VL - 35
JO - Computer Animation and Virtual Worlds
JF - Computer Animation and Virtual Worlds
IS - 3
M1 - e2259
ER -