Soccer video analysis : player tracking

• Main Author: Chua, Ian Jun Hao
• Other Authors: Wang Gang
• Format: Final Year Project
• Language: English
• Published: 2016
• Subjects: DRNTU::Engineering
• Online Access: http://hdl.handle.net/10356/67581
• Institution: Nanyang Technological University

In the sports entertainment industry, analysis of videos has become a necessity for any team to compete in the highest standards. Popular sports associations such as the Barclays’ Premier League (BPL) and the National Basketball Association (NBA) have utilized video analyzing software to gather useful information for their own team, such as player movements, team formations and ball trajectories. This benefits the team greatly as they can better understand their opponents before the game by studying their formation and tactics or analyze their own players’ common mistakes during a game. However, current codes are not in real time. The aim of this project is to improve on the existing codes so that it can run in real time. Furthermore, additional improvements such as real-time offside tracking can also be implemented on the existing codes. This would help eliminate human error during a game and ultimately improve the overall quality of a sports game. The project is broken down into three main parts: camera registration, player tracking and ball tracking. The author is in charge of player tracking. The goal of this project is to ultimately implement this technology in soccer matches.