Discrete particle swarm optimization for player trajectory extraction in soccer broadcast videos

Broadcasted soccer videos demand automatic semantic representation for eventand tactic analyses. player tracking is an important step which can be further processedand analyzed by sport experts to evaluate player or team performance. a novel scheme forplayer tracking in soccer broadcast videos is proposed in this research. following playerdetection using adaboost, player labeling, occlusion handling and mosaic construction,player tracking is formulated as an optimization problem allowing player trajectories tobe extracted using particle swarm optimization (pso). pso is an optimization methodinspired by the ocking behavior of birds, which was originally customized for continuousfunction value optimization. in this paper, a new application of discrete particle swarmoptimization for player tracking in soccer videos is proposed. updating equations for theparticle swarm optimization algorithm are modied based on problem characteristics anddiscrete variables to extract the player trajectory. experimental results show that themodied pso is promising in solving soccer player tracking problems.