:: Volume 5, Issue 2 (9-2017) ::
jgit 2017, 5(2): 57-77 Back to browse issues page
Player Tracking using Graph and Artificial Intelligence methods in Soccer Broadcast Videos
Mehrtash Manafifard *, Hamid Ebadi, Hamid Abrishami Moghaddam
K.N.Toosi University of Technology
Abstract:   (2825 Views)
Player tracking in soccer broadcast videos can be further processed by coaches and experts to judge weaknesses and strengths of the players and the team. Following player detection by Adaboost, player labeling, occlusion handling and player localization, player trajectory is extracted using combination of graph with artificial bee colony (ABC) and particle swarm optimization (PSO) in this research. PSO and ABC are optimization method inspired by the flocking behavior of birds and bees which were originally customized for continuous function value optimization. However, the need for modifying the discrete version in different applications is inevitable. In this paper, a modified version of discrete PSO and ABC for player tracking is proposed. Moreover, a new method for registering frames to the field model based on line recognition is proposed to diminish the search space. Finally, the proposed algorithm is tested on seven shots from six different soccer broadcast videos. Experimental results show the capability of the proposed method for extracting player trajectory in soccer broadcast videos.
Keywords: Tracking, Football, Gragh, Artificial bee colony, Particle swarm optimization
Full-Text [PDF 1503 kb]   (1447 Downloads)    
Type of Study: Research | Subject: Aerial Photogrammetry
Received: 2017/10/8 | Accepted: 2017/10/8 | Published: 2017/10/8



XML   Persian Abstract   Print



Rights and permissions
Creative Commons License This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Volume 5, Issue 2 (9-2017) Back to browse issues page