Accurate ball trajectory tracking and 3D visualization for computer-assisted sports broadcast

• Published in: Multimedia tools and applications Vol. 73; no. 3; pp. 1819 - 1842
• Main Authors: Labayen, Mikel, Olaizola, Igor G., Aginako, Naiara, Florez, Julian
• Format: Journal Article
• Language: English
• Published: Boston Springer US 01.12.2014; Springer; Springer Nature B.V
• Subjects: Algorithms; Analysis; Applied sciences; Artificial intelligence; Automation; Borders; Broadcasting; Camcorders; Cameras; Computer Communication Networks; Computer graphics; Computer Science; Computer science; control theory; systems; Computer systems and distributed systems. User interface; Computers; Data Structures and Information Theory; Electronics; Exact sciences and technology; Football; Integrated circuits; Integrated circuits by function (including memories and processors); Mathematical analysis; Multimedia; Multimedia computer applications; Multimedia Information Systems; Pattern recognition. Digital image processing. Computational geometry; Playgrounds; Real time; Semiconductor electronics. Microelectronics. Optoelectronics. Solid state devices; Software; Software engineering; Special Purpose and Application-Based Systems; Sports officiating; Studies; Television broadcasting; Television sports; Tennis; Three dimensional; Tracking control systems; Trajectories; Visual; Sports events broadcast; Camera calibration; Tracking; Segmentation; Computer graphics; Occlusion; Software development; Image processing; Polynomial approximation; Competitiveness; Hardware complexity; Image rendering; Image reconstruction; Visual system; Sport; Information communication; Graphical interface; Localization; Computer vision; Video cameras; Calibration; Real time; Requirement; Image segmentation; Data visualization; Tridimensional image; Border; Occultation; User behavior
• ISSN: 1380-7501; 1573-7721
• DOI: 10.1007/s11042-013-1558-x

The application of computer-aided controversial plays resolution in sport events significantly benefits organizers, referees and audience. Nowadays, especially in ball sports, very accurate technological solutions can be found. The main drawback of these systems is the need of complex and expensive hardware which makes them not affordable for less-known regional/traditional sports events. The lack of competitive systems with reduced hardware/software complexity and requirements motivates this research. Visual Analytics technologies permit system detecting the ball trajectory, solving with precision possible controversial plays. Ball is extracted from the video scene exploiting its shape features and velocity vector properties. Afterwards, its relative position to border line is calculated based on polynomial approximations. In order to enhance user visual experience, real-time rendering technologies are introduced to obtain virtual 3D reconstruction in quasi real-time. Comparing to other set ups, the main contribution of this work lays on the utilization of an unique camera per border line to extract 3D bounce point information. In addition, the system has no camera location/orientation limit, provided that line view is not occluded. Testing of the system has been done in real world scenarios, comparing the system output with referees’ judgment. Visual results of the system have been broadcasted during Basque Pelota matches.