End-to-End Camera Calibration for Broadcast Videos
| Autor: | Panna Felsen, Sujoy Ganguly, Jennifer Hobbs, Xinyu Wei, Patrick Lucey, Long Sha |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
Computer science
business.industry Calibration (statistics) ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION 020207 software engineering Tracking system 02 engineering and technology Image segmentation 01 natural sciences 010309 optics End-to-end principle 0103 physical sciences 0202 electrical engineering electronic engineering information engineering Segmentation Computer vision Artificial intelligence business Pose Homography (computer vision) Camera resectioning |
| Zdroj: | CVPR |
| DOI: | 10.1109/cvpr42600.2020.01364 |
| Popis: | The increasing number of vision-based tracking systems deployed in production have necessitated fast, robust camera calibration. In the domain of sport, the majority of current work focuses on sports where lines and intersections are easy to extract, and appearance is relatively consistent across venues. However, for more challenging sports like basketball, those techniques are not sufficient. In this paper, we propose an end-to-end approach for single moving camera calibration across challenging scenarios in sports. Our method contains three key modules: 1) area-based court segmentation, 2) camera pose estimation with embedded templates, 3) homography prediction via a spatial transform network (STN). All three modules are connected, enabling end-to-end training. We evaluate our method on a new college basketball dataset and demonstrate state of the art performance in variable and dynamic environments. We also validate our method on the World Cup 2014 dataset to show its competitive performance against the state-of-the-art methods. Lastly, we show that our method is two orders of magnitude faster than the previous state of the art on both datasets. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|