End of Degree Project Fernando Lago



News and Events 

"System implementation for automatic association of images of a soccer match with a synthetic model of the soccer field" 

Fernando Lago

E.T.S. Ing. Telecomunicación, Universidad Politécnica de Madrid, Julio 2016, "Sobresaliente, 10".

Advisor: Carlos Cuevas.

Today, the content broadcast on sporting events, such as soccer, have achieved great relevance. Because of this, there is an increasing interest in more detailed information, such as match statistics, analysis of specific team actions, etc. These statistics can be shown on the screen within a chart or, to provide a higher sense of realism, overlayed to the same soccer field. Nevertheless, the nature of the typically retransmitted images (perspective images that, in many times, show little parts of the soccer field) complicate the automatic extraction of the data, as well as its later viewing (through augmented reality) on the soccer field.

In order to carry out the automatic extraction of data it would be easier to have images without perspective such as, for example, aerial views of the soccer field. But this type of view, nowadays, still it is not typically used for retransmission of matches.

The aim of this Final Degree Project is to find associations between the images typically acquired by the television cameras and a synthetic model of the soccer field in which it will be easier to get the data (statistics) of the game. To do this association, it is necessary to analyze what soccer field area appears on the image. Once obtained the association between images, it will be easier the incorporation the overlayed information (augmented reality) on the original images.