A Trajectory-Based Ball Detection and Tracking Algorithm in Broadcast Tennis Video (MP-P6)

Author(s) : Xinguo Yu (Institute for Infocomm Research, Singapore) Chern-Horng Sim (National University of Singapore, Singapore) Jenny Ran Wang (School of Computer Science and Engineering, The University of New South Wales, Sydney 2052, Australia) Loong Fah Cheong (National University of Singapore, Singapore)

Abstract : Ball locations over frames facilitate tennis video analysis to a great extent. But so far no algorithm is able to obtain satisfactory result in locating the ball in broadcast tennis video. This paper presents a trajectory-based algorithm to detect and track the ball in broadcast tennis video. Unlike the object-based algorithm, it does not decide whether an object is the ball. Instead it decides whether a candidate trajectory is a ball trajectory. This algorithm is able to obtain ball locations for most frames in a broadcast tennis video, achieved based on these four ideas, namely, (1) an anti-model method to produce ball candidates for each frame, (2) a trajectory-based scheme to generate, identify, and extend the ball trajectories from the produced candidates, (3) a method to infer the ball locations according to the player’s location and the points of hitting, (4) a method to compute the ball locations not visible in the video based on the obtained ball locations. The experimental results show that our algorithm obtains the ball locations for above 96% frames in enough accuracy.

Menu