Automatic Soccer Player Tracking in Single Camera with Robust Occlusion Handling Using Attribute Matching

This paper presents an automatic method to track soccer players in soccer video recorded from a single camera where the occurrence of pan-tilt-zoom can take place. The automatic object tracking is intended to support texture extraction in a free viewpoint video authoring application for soccer video. To ensure that the identity of the tracked object can be correctly obtained, background segmentation is performed and automatically removes commercial billboards whenever it overlaps with the soccer player. Next, object tracking is performed by an attribute matching algorithm for all objects in the temporal domain to find and maintain the correlation of the detected objects. The attribute matching process finds the best match between two objects in different frames according to their pre-determined attributes: position, size, dominant color and motion information. Utilizing these attributes, the experimental results show that the tracking process can handle occlusion problems such as occlusion involving more than three objects and occluded objects with similar color and moving direction, as well as correctly identify objects in the presence of camera movements. key words: free viewpoint, attribute matching, automatic object tracking, soccer video

[1]  Shihong Lao,et al.  Multiple Player Tracking in Sports Video: A Dual-Mode Two-Way Bayesian Inference Approach With Progressive Observation Modeling , 2011, IEEE Transactions on Image Processing.

[2]  Masaki Hayashi,et al.  Multiple players tracking and identification using group detection and player number recognition in sports video , 2013, IECON 2013 - 39th Annual Conference of the IEEE Industrial Electronics Society.

[3]  Hiroshi Sankoh,et al.  Free-viewpoint Video Rendering in Large Outdoor Space such as Soccer Stadium based on Object Extraction and Tracking Technology , 2014 .

[4]  M. M. Naushad Ali,et al.  An Efficient Algorithm for Detection of Soccer Ball and Players , 2012 .

[5]  James J. Little,et al.  Learning to Track and Identify Players from Broadcast Sports Videos , 2013, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  I. Kitahara,et al.  Live mixed-reality 3D video in soccer stadium , 2003, The Second IEEE and ACM International Symposium on Mixed and Augmented Reality, 2003. Proceedings..

[7]  Yu Huang,et al.  Players and Ball Detection in Soccer Videos Based on Color Segmentation and Shape Analysis , 2007, MCAM.

[8]  M. Heydari,et al.  An MLP-based player detection and tracking in broadcast soccer video , 2012, 2012 International Conference of Robotics and Artificial Intelligence.

[9]  R. Anido,et al.  Distributed real-time soccer tracking , 2004, VSSN '04.

[10]  Shigeyuki Sakazawa,et al.  Robust background subtraction method based on 3D model projections with likelihood , 2010, 2010 IEEE International Workshop on Multimedia Signal Processing.

[11]  Markus H. Gross,et al.  Articulated Billboards for Video‐based Rendering , 2010, Comput. Graph. Forum.

[12]  J.-Y. Bouguet,et al.  Pyramidal implementation of the lucas kanade feature tracker , 1999 .

[13]  Hirokatsu Kataoka,et al.  Player position estimation by monocular camera for soccer video analysis , 2011, SICE Annual Conference 2011.

[14]  Yongduek Seo,et al.  Tracking Soccer Ball in TV Broadcast Video , 2005, ICIAP.

[15]  Takeo Kanade,et al.  Virtualized Reality: Constructing Virtual Worlds from Real Scenes , 1997, IEEE Multim..

[16]  F. Meyer,et al.  Color image segmentation , 1992 .

[17]  Xiaokang Yang,et al.  Camshift Guided Particle Filter for Visual Tracking , 2007, SiPS.

[18]  Shamik Sural,et al.  Graph-Based Multiplayer Detection and Tracking in Broadcast Soccer Videos , 2008, IEEE Transactions on Multimedia.

[19]  Tetsuya Takiguchi,et al.  3D tracking of soccer players using time-situation graph in monocular image sequence , 2012, Proceedings of the 21st International Conference on Pattern Recognition (ICPR2012).

[20]  Irfan A. Essa,et al.  Player localization using multiple static cameras for sports visualization , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[21]  Jia Liu,et al.  Automatic player labeling, tracking and field registration and trajectory mapping in broadcast soccer video , 2011, TIST.

[22]  Jake K. Aggarwal,et al.  Tracking soccer players using broadcast TV images , 2005, IEEE Conference on Advanced Video and Signal Based Surveillance, 2005..

[23]  Hiroshi Sankoh,et al.  Dynamic camera calibration method for free-viewpoint experience in sport videos , 2012, ACM Multimedia.

[24]  Slawomir Mackowiak,et al.  Football Player Detection in Video Broadcast , 2010, ICCVG.

[25]  Hideo Saito,et al.  Synthesizing Free-Viewpoing Images from Multiple View Videos in Soccer StadiumADIUM , 2006, International Conference on Computer Graphics, Imaging and Visualisation (CGIV'06).

[26]  Jia Liu,et al.  Automatic Player Detection, Labeling and Tracking in Broadcast Soccer Video , 2007, BMVC.

[27]  Shih-Fu Chang,et al.  VisualSEEk: a fully automated content-based image query system , 1997, MULTIMEDIA '96.

[28]  Yoshimitsu Aoki,et al.  Football players and ball trajectories projection from single camera's image , 2011, 2011 17th Korea-Japan Joint Workshop on Frontiers of Computer Vision (FCV).