Abstract
This paper presents a complete, general and modular system which after a simple previous configuration is able to detect and track each player on the court or field. The presented multi-camera system is based on a mono-camera object detection and tracking system originally designed for video surveillance applications. Target sports of the developed system are team sports (e.g., basketball, soccer). The main objective of this paper is to present a semi-supervised system able to detect and track the players in multi-camera sports videos, focusing on the fusion of different tracks of detected blobs in order to match tracks across cameras. The proposed system is simpler than other systems from the state of the art, can operate in real time and has margin to be improved and to reduce supervision adding additional complexity. In addition to the detection and tracking system, an evaluation system has been designed to obtain quantitative results of the system performance.
Similar content being viewed by others
Notes
In this case the tracking is defined as the set of blobs that is fused with another set of blobs
In the ISSIA soccer dataset, this ID is extracted from the ground truth tracking of each of the 6 cameras. The blob ID for the blobs of a player is the same for the six tracking files.
The LOA, iLOA and eLOA are “classes”, which are instantiated as specific lists (e.g., FCeLOA, RiLOA) during the different fusions performed (see section 5.2).
A document with the corrections is available in the created web page.
References
Anjum N, Cavallaro A (2009) Trajectory association and fusion across partialy overlapping cameras. AVSS, pp 201–206
Bebie T, Bieri H (1998) SoccerMan-reconstructing soccer games from video sequences. Image processingl 1:898–902
Choi S, Seo Y, Kim H, Hong KS (1997) Where are the ball and players? Soccer game analysis with colorbased tracking and image mosaic. In: Proc. of ICIAP, pp 196–203
D’Orazio T, Leo M, Mosca N, Spagnolo P, Mazzeo PL (2009) A semi-automatic system for ground truth generation of soccer video sequences. AVSS, pp 559–564
de Meneses YL, Roduit P, Luisier F, Jacot J (2005) Trajectory analysis for sport and video surveillance. Electron Lett Comput Vis Image Anal 5(3):148–156
Du W, Hayet JB, Piater J, Verly J (2006) Collaborative multi-camera tracking of athletes in team sports. Workshop on Computer Vision Based Anal in Sports Environments (CVBASE), pp 2–13
Figueroa PJ, Leite NJ, Barros RML (2006) Tracking soccer players aiming their kinematical motion analysis. Trans Comput Vis Image Underst 101(2):122–135
Figueroa P, Leite N, Barros R, Cohen I, Medioni G (2004) Tracking soccer players using the graph representation. In: Proc. of ICPR 4:787–790
Hartley R, Zisserman A (2003) A multiple view geometry in computer vision. Cambridge University Press
Huang Y, Llach J, Bhagavathy S (2007) Players and ball detection in soccer videos based on color segmentation and shape analysis. Lecture Notes in Computer Science 4577:416–425
Junejo IN, Foroosh H (2007) Trajectory rectification and path modeling for video surveillance. In: Proc. of ICCV, pp 1–7
Kang J, Cohen I, Medioni G (2004) Tracking people in crowded scenes across multiple cameras. In: Proc. of ACCV
Kayumbi G, Anjum N, Cavallaro A (2008) Global trajectory reconstruction from distributed visual sensors. In: Proc. of ICDSC, pp 1–8
Kayumbi G, Mazzeo PL, Spagnolo P, Taj M, Cavallaro A (2008) Distributed visual sensing for virtual top-view trajectory generation in football videos. In: Proc. of CIVR
Martín R, Martínez JM (2013) An automatic system for sports analytics in multi-camera tennis videos. In: Proc. of AMMDS-AVSS (in press)
Misu T, Gohshi S, Izumi Y, Fujita Y, Naemura M (2004) Robust tracking of athletes using multiple features of multiple views. In: Proc. of WSCG, pp 285–292
Nummiaro K, Koller-Meier E, Svoboda T, Roth D, Van Gool J-L (2003) Color-based object tracking in multi-camera environments. In: Proc. of DAGM, pp 591–599
Poppe C, Bruyne SD, Verstockt S, de Walle RV (2010) Multi-camera analysis of soccer sequences. In: Proc. of AVSS, pp 26–31
Sachiko I, Hideo S (2004) Parallel tracking of all soccer players by integrating detected positions in multiple view images. In: Proc. of ICPR
SanMiguel JC, Martínez JM (2012) A semantic-based probabilistic approach for real-time video event recognition. Comp Vis Image Underst 116(9):937–952
Sheikh YA, Shah M (2008) Trajectory association across multiple airborne cameras. Trans Pattern Anal Mach Intell 30(2):361–367
Taj M, Cavallaro A (2009) Multi-camera track-before-detect. In: Proc. of ICDSC
Tong X, et al (2004) An effective and fast soccer ball detection and tracking method. In: Proc of ICPR 4:795–798
Xinguo Y, Farin D (2005) Current and emerging topics in sports video processing. In: Proc. of ICME
Xu M, Orwell J, Jones G (2004) Tracking football players with multiple cameras. In: Proc. of ICIP 5:2909–2912
Acknowledgments
This work has been partially supported by the Spanish Government (TEC2011-25995).
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Martín, R., Martínez, J.M. A semi-supervised system for players detection and tracking in multi-camera soccer videos. Multimed Tools Appl 73, 1617–1642 (2014). https://doi.org/10.1007/s11042-013-1659-6
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-013-1659-6