# Localization of Ice-Rink for Broadcast Hockey Videos

arXiv: https://arxiv.org/abs/2104.10847 · HTML: https://mlsanity.com/paper/2104.10847/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 4 of its 13 references that are in the map.
- First posted on arXiv 22 Apr 2021.

## Builds on (most cited first, 4 of 4)

- [Sports Camera Calibration via Synthetic Data](https://mlsanity.com/paper/1810.10658/) (2018)
- [Optimizing Through Learned Errors for Accurate Sports Field Registration](https://mlsanity.com/paper/1909.08034/) (2019)
- [Real-Time Camera Pose Estimation for Sports Fields](https://mlsanity.com/paper/2003.14109/) (2020)
- [Automated Top View Registration of Broadcast Football Videos](https://mlsanity.com/paper/1703.01437/) (2017)

## Built on by (most cited first, 4 of 4)

- [BroadTrack: Broadcast Camera Tracking for Soccer](https://mlsanity.com/paper/2412.01721/) (2024)
- [KaliCalib: A Framework for Basketball Court Registration](https://mlsanity.com/paper/2209.07795/) (2022)
- [Can Geometry Save Central Views for Sports Field Registration?](https://mlsanity.com/paper/2504.20052/) (2025)
- [Towards Automated Key-Point Detection in Images with Partial Pool View](https://mlsanity.com/paper/2208.05641/) (2022)

## Abstract (arXiv)

In this work, an automatic and simple framework for hockey ice-rink localization from broadcast videos is introduced. First, video is broken into video-shots by a hierarchical partitioning of the video frames, and thresholding based on their histograms. To localize the frames on the ice-rink model, a ResNet18-based regressor is implemented and trained, which regresses to four control points on the model in a frame-by-frame fashion. This leads to the projection jittering problem in the video. To overcome this, in the inference phase, the trajectory of the control points on the ice-rink model are smoothed, for all the consecutive frames of a given video-shot, by convolving a Hann window with the achieved coordinates. Finally, the smoothed homography matrix is computed by using the direct linear transform on the four pairs of corresponding points. A hockey dataset for training and testing the regressor is gathered. The results show success of this simple and comprehensive procedure for localizing the hockey ice-rink and addressing the problem of jittering without affecting the accuracy of homography estimation.