Vision-based real-time speed tracking

Motion-based features of a human can depict the approximate moving direction of the body while walking or slow running. But nearly all the motion-based features are not robust in capturing the speed of a moving person. In this research, a real-time vision-based method is proposed to compute the speed of a player during a soccer game using the image frames from video cameras placed in all the corners and mid-points of a soccer field. The coach´s main concern is to increase the dynamism and aggression among the teammates. Calculating the speed of each player helps the coach in determining a player´s performance (while team selection), and for repositioning a player during a game.