Synthesizing free-viewpoint images from multiple view videos in soccer stadium

Kunihiko Hayashi, Hideo Saito

Research output: Chapter in Book/Report/Conference proceedingConference contribution

30 Citations (Scopus)

Abstract

We propose a new method for synthesizing free-viewpoint images from multiple view videos in soccer stadium. The previous method[1] can synthesize virtual viewpoint images at only intermediate viewpoint between real cameras. The previous method estimates only pixel-wise correspondences between input images for synthesizing virtual viewpoint images without a 3D reconstruction. In this paper, we propose a new method that can synthesize virtual viewpoint images at free-viewpoints which are not limited at intermediate viewpoint by reconstructing simple 3D models. To reconstruct the object soccer scene, we perform camera calibration without marker objects because we can save labor to put such objects on the soccer ground. For reconstructing of dynamic regions, we employ a bill-board representation of dynamic regions. The position of the billboard is determined by tracking the players. The billboard representation can realize a simple 3D reconstruction and transform textures into the appropriate appearance of the aspect of dynamic regions at virtual viewpoint. We also introduce the choice of optimal cameras for rendering on the billboard. We choose the optimal camera that has the most similar texture at virtual viewpoint. By applying our method to a soccer scene, we can synthesize free-viewpoint images by employing only pixel values of real images.

Original languageEnglish
Title of host publicationProceedings - Computer Graphics, Imaging and Visualisation: Techniques and Applications, CGIV'06
Pages220-225
Number of pages6
Volume2006
DOIs
Publication statusPublished - 2006
EventInternational Conference on Computer Graphics, Imaging and Visualisation, CGIV'06 - Sydney, Australia
Duration: 2006 Jul 262006 Jul 28

Other

OtherInternational Conference on Computer Graphics, Imaging and Visualisation, CGIV'06
CountryAustralia
CitySydney
Period06/7/2606/7/28

Fingerprint

Stadiums
Cameras
Textures
Pixels
Calibration
Personnel

Keywords

  • Calibration
  • Free-viewpoint
  • Homography
  • Image processing
  • Multiple view videos
  • Projective matrix

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design
  • Software

Cite this

Hayashi, K., & Saito, H. (2006). Synthesizing free-viewpoint images from multiple view videos in soccer stadium. In Proceedings - Computer Graphics, Imaging and Visualisation: Techniques and Applications, CGIV'06 (Vol. 2006, pp. 220-225). [1663795] https://doi.org/10.1109/CGIV.2006.83

Synthesizing free-viewpoint images from multiple view videos in soccer stadium. / Hayashi, Kunihiko; Saito, Hideo.

Proceedings - Computer Graphics, Imaging and Visualisation: Techniques and Applications, CGIV'06. Vol. 2006 2006. p. 220-225 1663795.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Hayashi, K & Saito, H 2006, Synthesizing free-viewpoint images from multiple view videos in soccer stadium. in Proceedings - Computer Graphics, Imaging and Visualisation: Techniques and Applications, CGIV'06. vol. 2006, 1663795, pp. 220-225, International Conference on Computer Graphics, Imaging and Visualisation, CGIV'06, Sydney, Australia, 06/7/26. https://doi.org/10.1109/CGIV.2006.83
Hayashi K, Saito H. Synthesizing free-viewpoint images from multiple view videos in soccer stadium. In Proceedings - Computer Graphics, Imaging and Visualisation: Techniques and Applications, CGIV'06. Vol. 2006. 2006. p. 220-225. 1663795 https://doi.org/10.1109/CGIV.2006.83
Hayashi, Kunihiko ; Saito, Hideo. / Synthesizing free-viewpoint images from multiple view videos in soccer stadium. Proceedings - Computer Graphics, Imaging and Visualisation: Techniques and Applications, CGIV'06. Vol. 2006 2006. pp. 220-225
@inproceedings{513577ab205742c3a0fd416a91aff9c6,
title = "Synthesizing free-viewpoint images from multiple view videos in soccer stadium",
abstract = "We propose a new method for synthesizing free-viewpoint images from multiple view videos in soccer stadium. The previous method[1] can synthesize virtual viewpoint images at only intermediate viewpoint between real cameras. The previous method estimates only pixel-wise correspondences between input images for synthesizing virtual viewpoint images without a 3D reconstruction. In this paper, we propose a new method that can synthesize virtual viewpoint images at free-viewpoints which are not limited at intermediate viewpoint by reconstructing simple 3D models. To reconstruct the object soccer scene, we perform camera calibration without marker objects because we can save labor to put such objects on the soccer ground. For reconstructing of dynamic regions, we employ a bill-board representation of dynamic regions. The position of the billboard is determined by tracking the players. The billboard representation can realize a simple 3D reconstruction and transform textures into the appropriate appearance of the aspect of dynamic regions at virtual viewpoint. We also introduce the choice of optimal cameras for rendering on the billboard. We choose the optimal camera that has the most similar texture at virtual viewpoint. By applying our method to a soccer scene, we can synthesize free-viewpoint images by employing only pixel values of real images.",
keywords = "Calibration, Free-viewpoint, Homography, Image processing, Multiple view videos, Projective matrix",
author = "Kunihiko Hayashi and Hideo Saito",
year = "2006",
doi = "10.1109/CGIV.2006.83",
language = "English",
isbn = "0769526063",
volume = "2006",
pages = "220--225",
booktitle = "Proceedings - Computer Graphics, Imaging and Visualisation: Techniques and Applications, CGIV'06",

}

TY - GEN

T1 - Synthesizing free-viewpoint images from multiple view videos in soccer stadium

AU - Hayashi, Kunihiko

AU - Saito, Hideo

PY - 2006

Y1 - 2006

N2 - We propose a new method for synthesizing free-viewpoint images from multiple view videos in soccer stadium. The previous method[1] can synthesize virtual viewpoint images at only intermediate viewpoint between real cameras. The previous method estimates only pixel-wise correspondences between input images for synthesizing virtual viewpoint images without a 3D reconstruction. In this paper, we propose a new method that can synthesize virtual viewpoint images at free-viewpoints which are not limited at intermediate viewpoint by reconstructing simple 3D models. To reconstruct the object soccer scene, we perform camera calibration without marker objects because we can save labor to put such objects on the soccer ground. For reconstructing of dynamic regions, we employ a bill-board representation of dynamic regions. The position of the billboard is determined by tracking the players. The billboard representation can realize a simple 3D reconstruction and transform textures into the appropriate appearance of the aspect of dynamic regions at virtual viewpoint. We also introduce the choice of optimal cameras for rendering on the billboard. We choose the optimal camera that has the most similar texture at virtual viewpoint. By applying our method to a soccer scene, we can synthesize free-viewpoint images by employing only pixel values of real images.

AB - We propose a new method for synthesizing free-viewpoint images from multiple view videos in soccer stadium. The previous method[1] can synthesize virtual viewpoint images at only intermediate viewpoint between real cameras. The previous method estimates only pixel-wise correspondences between input images for synthesizing virtual viewpoint images without a 3D reconstruction. In this paper, we propose a new method that can synthesize virtual viewpoint images at free-viewpoints which are not limited at intermediate viewpoint by reconstructing simple 3D models. To reconstruct the object soccer scene, we perform camera calibration without marker objects because we can save labor to put such objects on the soccer ground. For reconstructing of dynamic regions, we employ a bill-board representation of dynamic regions. The position of the billboard is determined by tracking the players. The billboard representation can realize a simple 3D reconstruction and transform textures into the appropriate appearance of the aspect of dynamic regions at virtual viewpoint. We also introduce the choice of optimal cameras for rendering on the billboard. We choose the optimal camera that has the most similar texture at virtual viewpoint. By applying our method to a soccer scene, we can synthesize free-viewpoint images by employing only pixel values of real images.

KW - Calibration

KW - Free-viewpoint

KW - Homography

KW - Image processing

KW - Multiple view videos

KW - Projective matrix

UR - http://www.scopus.com/inward/record.url?scp=34247638654&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=34247638654&partnerID=8YFLogxK

U2 - 10.1109/CGIV.2006.83

DO - 10.1109/CGIV.2006.83

M3 - Conference contribution

AN - SCOPUS:34247638654

SN - 0769526063

SN - 9780769526065

VL - 2006

SP - 220

EP - 225

BT - Proceedings - Computer Graphics, Imaging and Visualisation: Techniques and Applications, CGIV'06

ER -