Soccer On Your Tabletop

Konstantinos Rematas1   Ira Kemelmacher-Shlizerman1,2   Brian Curless1   Steve Seitz1,3
1University of Washington     2Facebook     Google3

IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018

Abstract

We present a system that transforms a monocular video of a soccer game into a moving 3D reconstruction, in which the players and field can be rendered interactively with a 3D viewer or through an Augmented Reality device. At the heart of our paper is an approach to estimate the depth map of each player, using a CNN that is trained on 3D player data extracted from soccer video games. We compare with state of the art body pose and depth estimation techniques, and show results on both synthetic ground truth benchmarks, and real YouTube soccer footage.

Paper

Paper pdf, to appear in CVPR 2018!

Supplementary material

@inproceedings{rematasCVPR18,
author = {Konstantinos Rematas and Ira Kemelmacher-Shlizerman and Brian Curless and Steve Seitz },
title = {Soccer On Your Tabletop},
year = {2018},
booktitle = {CVPR}
}

Code

On Github.

Press

Acknowledgments

This work is supported by NSF/Intel Visual and Experimental Computing Award #1538618 and the UW Reality Lab.

Contact

Konstantinos Rematas krematas@cs.washington.edu