Optimizing Through Learned Errors for Accurate Sports Field Registration

Wei Jiang, Juan Camilo Gamboa Higuera, Baptiste Angles, Weiwei Sun, Mehrsan Javan, Kwang Moo Yi; Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2020, pp. 201-210

Abstract


We propose an optimization-based framework to register sports field templates onto broadcast videos. For accurate registration we go beyond the prevalent feed-forward paradigm. Instead, we propose to train a deep network that regresses the registration error, and then register images by finding the registration parameters that minimize the regressed error. We demonstrate the effectiveness of our method by applying it to real-world sports broadcast videos, outperforming the state of the art. We further apply our method on a synthetic toy example and demonstrate that our method brings significant gains even when the problem is simplified and unlimited training data is available.

Related Material


[pdf]
[bibtex]
@InProceedings{Jiang_2020_WACV,
author = {Jiang, Wei and Higuera, Juan Camilo Gamboa and Angles, Baptiste and Sun, Weiwei and Javan, Mehrsan and Yi, Kwang Moo},
title = {Optimizing Through Learned Errors for Accurate Sports Field Registration},
booktitle = {Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)},
month = {March},
year = {2020}
}