Temporal segmentation of association football from TV broadcasting

Author

Siles, Francisco

Author_Institution

PRIS-Lab.: Pattern Recognition & Intell. Syst. Lab., Univ. de Costa Rica, San Jose, Costa Rica

fYear

2013

fDate

19-21 June 2013

Firstpage

39

Lastpage

43

Abstract

This paper presents the first module of our system Ace for the automated interpretation of association football games. This module is in charge of the temporal segmentation of the input broadcasting TV video into scenes. The temporal segmentation is based on a new devised dissimilarity measure using the Bhattacharyya distance of the hue and local spatial variance histograms, to incorporate chromatic and structural properties of the images. Great results of extended experiments conducted on the last FIFA´s World Cup are presented.

Keywords

image segmentation; sport; television broadcasting; video signal processing; ACE; Bhattacharyya distance; FIFA World Cup; TV broadcasting; TV video; association football; automated association football game interpretation; chromatic properties; devised dissimilarity measure; spatial variance histograms; structural properties; temporal segmentation; Cameras; Equations; Feature extraction; Games; Histograms; Image color analysis; Visualization;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Engineering Systems (INES), 2013 IEEE 17th International Conference on

Conference_Location

San Jose

Print_ISBN

978-1-4799-0828-8

Type

conf

DOI

10.1109/INES.2013.6632843

Filename

6632843