In the paper, we describe the technical details of a multi-player tracker system using tracking data obtained from a single low-cost stationary camera on field hockey games. Analyzing the tracking data of the players only from the transmitted video opens a multitude of applications that allows the cost of technology to be reduced. This method does not depend on the cooperation of the players (by using sensors) or their teams (by sharing data with a third party). The approach taken in this paper uses a variety of computer vision and tracking techniques. Making player tracking data more accessible lowers the barrier to entry for sports research and increases the period during which advanced analysis methods can be applied. The proposed system runs the full pipeline at 3 fps on a computer with a simple graphics card.
© Copyright 2022 Machine Learning and Data Mining for Sports Analytics. MLSA 2021. Communications in Computer and Information Science. Published by Springer. All rights reserved.