A review on soccer player tracking techniques based on extracted features

Abstract

Playing games specially soccer is one of the most likable activities which performed by a large mass base in most countries. All the broadcasted videos of these games are taken as an extensive challenging by researchers in computer vision field. The topics which get the big chance of importance in industry are the tracking of both player and field. This can be used to describe the match for people who are visually impaired. In this review, we summarize some of the recent researcher's work related to different player tracking techniques according to specific features extracted from the scene to build a robust object (player and/or ball) tracking, such as motion, color, edges, etc., which considered as the most commonly used in object tracking. Also, we compare some of these techniques according to two groups of publicly available data sets in order to study the effect of selecting these features on the tracking accuracy.