An Unsupervised Approach for Segmentation and Clustering of Soccer Players

In this work we consider the problem of soccer team discrimination. The approach we propose starts from the monocular images acquired by a still camera. The first step is the soccer player detection, performed by means of background subtraction. An algorithm based on pixels energy content has been implemented in order to detect moving objects. The use of energy information, combined with a temporal sliding window procedure, allows to be substantially independent from motion hypothesis. Colour histograms in RGB space are extracted from each player, and provided to the unsupervised classification phase. This is composed by two distinct modules: firstly, a modified version of the BSAS clustering algorithm builds the clusters for each class of objects. Then, at runtime, each player is classified by evaluating its distance, in the features space, from the classes previously detected. Algorithms have been tested on different real soccer matches of the Italian Serie A.

[1]  Shih-Fu Chang,et al.  Structure analysis of soccer video with domain knowledge and hidden Markov models , 2004, Pattern Recognit. Lett..

[2]  Jeff B. Pelz,et al.  Image Segmentation By Local Morphological Granulometries , 1989, 12th Canadian Symposium on Remote Sensing Geoscience and Remote Sensing Symposium,.

[3]  Josef Kittler,et al.  Minimum error thresholding , 1986, Pattern Recognit..

[4]  Michael Beetz,et al.  Camera-based observation of football games for analyzing multi-agent activities , 2006, AAMAS '06.

[5]  Shih-Fu Chang,et al.  Real-time view recognition and event detection for sports video , 2004, J. Vis. Commun. Image Represent..

[6]  Yoshihiro Fujita,et al.  Robust Tracking of Athletes Using Multiple Features of Multiple Views , 2004, WSCG.

[7]  Takeo Kanade,et al.  Advances in Cooperative Multi-Sensor Video Surveillance , 1999 .

[8]  Justus H. Piater,et al.  A modular multi-camera framework for team sports tracking , 2005, IEEE Conference on Advanced Video and Signal Based Surveillance, 2005..

[9]  Kentaro Toyama,et al.  Wallflower: principles and practice of background maintenance , 1999, Proceedings of the Seventh IEEE International Conference on Computer Vision.

[10]  Edward R. Dougherty,et al.  Morphological pattern-spectrum classification of noisy shapes: Exterior granulometries , 1995, Pattern Recognit..

[11]  A. Murat Tekalp,et al.  Automatic Soccer Video Analysis and Summarization , 2003, IS&T/SPIE Electronic Imaging.

[12]  Ming Xu,et al.  Architecture and algorithms for tracking football players with multiple cameras , 2005 .

[13]  Carlos Orrite-Uruñuela,et al.  Detected motion classification with a double-background and a Neighborhood-based difference , 2003, Pattern Recognit. Lett..

[14]  Ahmed Bouridane,et al.  Automatic classification and recognition of shoeprints , 1999 .

[15]  Zhifei Xu,et al.  Segmentation of players and team discrimination in soccer videos , 2005, Proceedings of 2005 IEEE International Workshop on VLSI Design and Video Technology, 2005..

[16]  Chng Eng Siong,et al.  A Player-Possession Acquisition System for Broadcast Soccer Video , 2005, 2005 IEEE International Conference on Multimedia and Expo.

[17]  A.M. Tekalp,et al.  Robust dominant color region detection and color-based applications for sports video , 2003, Proceedings 2003 International Conference on Image Processing (Cat. No.03CH37429).

[18]  Petros Maragos,et al.  Pattern Spectrum and Multiscale Shape Representation , 1989, IEEE Trans. Pattern Anal. Mach. Intell..

[19]  E. Dougherty,et al.  Gray-scale morphological granulometric texture classification , 1994 .

[20]  Alberto Del Bimbo,et al.  Semantic annotation of soccer videos: automatic highlights identification , 2003, Comput. Vis. Image Underst..

[21]  Jeff B. Pelz,et al.  Morphological texture-based maximum-likelihood pixel classification based on local granulometric moments , 1992, Pattern Recognit..

[22]  John Flynn,et al.  Automated processing of shoeprint images based on the Fourier transform for use in forensic science , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[23]  Edward R. Dougherty,et al.  The granulometric size density in filter optimization , 1999, XII Brazilian Symposium on Computer Graphics and Image Processing (Cat. No.PR00481).

[24]  Nicolas Vandenbroucke,et al.  Color image segmentation by pixel classification in an adapted hybrid color space. Application to soccer image analysis , 2003, Comput. Vis. Image Underst..

[25]  Yoshinori Izumi,et al.  Morphological segmentation of sport scenes using color information , 2000, IEEE Trans. Broadcast..

[26]  B. Ripley,et al.  Pattern Recognition , 1968, Nature.

[27]  Joo-Hwee Lim,et al.  Team possession analysis for broadcast soccer video based on ball trajectory , 2003, Fourth International Conference on Information, Communications and Signal Processing, 2003 and the Fourth Pacific Rim Conference on Multimedia. Proceedings of the 2003 Joint.

[28]  T. Kanade,et al.  Color information for region segmentation , 1980 .

[29]  Justus H. Piater,et al.  Robust Non-Rigid Object Tracking Using Point Distribution Models , 2005, BMVC.