Soccer Field Localization from a Single Image

Namdar Homayounfar,Sanja Fidler,Raquel Urtasun

DOI: https://doi.org/10.48550/arXiv.1604.02715

2016-04-11

Abstract:In this work, we propose a novel way of efficiently localizing a soccer field from a single broadcast image of the game. Related work in this area relies on manually annotating a few key frames and extending the localization to similar images, or installing fixed specialized cameras in the stadium from which the layout of the field can be obtained. In contrast, we formulate this problem as a branch and bound inference in a Markov random field where an energy function is defined in terms of field cues such as grass, lines and circles. Moreover, our approach is fully automatic and depends only on single images from the broadcast video of the game. We demonstrate the effectiveness of our method by applying it to various games and obtain promising results. Finally, we posit that our approach can be applied easily to other sports such as hockey and basketball.

Computer Vision and Pattern Recognition

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to efficiently locate the football field in a single broadcast image. Specifically, existing methods usually rely on manually annotating key frames or installing fixed dedicated cameras in the stadium to obtain the field layout. However, these methods either require a great deal of human intervention or have high requirements for equipment, which limits their wide application. This paper proposes a new method. By constructing a Markov random field (MRF) and using an energy function to define field cues (such as grass, lines and circles), the goal of automatically locating the football field from a single broadcast image is achieved. This method is fully automated and only depends on a single image in the game video, having high practicality and flexibility. In addition, the author also points out that this method can be easily extended to other sports, such as ice hockey and basketball.