Soccer Field Dataset

This labeled soccer field dataset is compiled to classify different areas of the soccer field. It is gathered from IPL Azadi Soccer Dataset, Toronto Dataset, UBC Dataset, and Youtube. This collection contains three classes of:

- penalty area,

- center area, and

- middle area (between these two areas).

Since the dataset is provided for the VGG16 CNN network, the dimension of all video frames is 224x224. As the number of frames in some areas is less than others, one can use a simple solution (such as applying a random noise, a random rotation, a random brightness, and a random affine transformation, and so forth) to enrich it.

To complete this dataset or for more information please contact skasaei[at]sharif[dot]edu.

Related Publications

  • M. Fotouhi, S. Fouladi, and S. Kasaei. Projection Matrix by Orthogonal Vanishing Points. Springer, Multimedia Tools & Applications, 76(15):16189-16223, August 2017. [BibTeX]
  • A. Bozorgpour, M. Fotouhi, and S. Kasaei. Robust Homography Optimization in Soccer Scenes. In The 23rd Iranian Conference on Electrical Engineering (ICEE), pages 787-792, Tehran, Iran, May 2015. [BibTeX]
  • M. Fotouhi, A. Bozorgpour, and S. Kasaei. Automatic Soccer Field Line Recognition by Minimum Information. In International Symposium on Artificial Intelligence & Signal Processing (AISP), pages 136-142, Mashhad, Iran, March 2015. [BibTeX]