Fast gradient-based mesh generation method for the stereo image representation

Ilkwon Park, Hyeran Byun

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

1 Citation (Scopus)

Abstract

This paper proposes a fast gradient-based mesh generation method for the stereo image representation. In our approach, right image in stereo image and intermediate views are fundamentally predicted and synthesized from left image by 2D image warping using regular mesh and vice versa. To overcome texture distortion on object boundaries and preserve texture smoothness of homogenous areas, we propose node selection on the strong edges in the gradient map. Furthermore, the selected nodes are evaluated by stereo matching error and validated by cross validation for nodal disparity. Each node point is iteratively moving along the pixels with high gradient value instead every point to find the optimal node position. Therefore, our approach provides a fast mesh optimization as well as a reliable image quality. The experimental results show that the proposed approach provides computational efficiency and high PSNR of prediction image compared to previous methods.

Original languageEnglish
Title of host publication3DTV-CON 2009 - 3rd 3DTV-Conference
Subtitle of host publicationThe True Vision - Capture, Transmission and Display of 3D Video, Proceedings
DOIs
Publication statusPublished - 2009 Oct 19
Event3rd 3DTV-Conference: The True Vision - Capture, Transmission and Display of 3D Video, 3DTV-CON 2009 - Potsdam, Germany
Duration: 2009 May 42009 May 6

Other

Other3rd 3DTV-Conference: The True Vision - Capture, Transmission and Display of 3D Video, 3DTV-CON 2009
CountryGermany
CityPotsdam
Period09/5/409/5/6

Fingerprint

Mesh generation
Textures
Computational efficiency
Image quality
Pixels

All Science Journal Classification (ASJC) codes

  • Computer Graphics and Computer-Aided Design
  • Computer Networks and Communications
  • Computer Vision and Pattern Recognition
  • Human-Computer Interaction

Cite this

Park, I., & Byun, H. (2009). Fast gradient-based mesh generation method for the stereo image representation. In 3DTV-CON 2009 - 3rd 3DTV-Conference: The True Vision - Capture, Transmission and Display of 3D Video, Proceedings [5069672] https://doi.org/10.1109/3DTV.2009.5069672
Park, Ilkwon ; Byun, Hyeran. / Fast gradient-based mesh generation method for the stereo image representation. 3DTV-CON 2009 - 3rd 3DTV-Conference: The True Vision - Capture, Transmission and Display of 3D Video, Proceedings. 2009.
@inproceedings{1f36b2c872e04442a183d53ccdb98b8d,
title = "Fast gradient-based mesh generation method for the stereo image representation",
abstract = "This paper proposes a fast gradient-based mesh generation method for the stereo image representation. In our approach, right image in stereo image and intermediate views are fundamentally predicted and synthesized from left image by 2D image warping using regular mesh and vice versa. To overcome texture distortion on object boundaries and preserve texture smoothness of homogenous areas, we propose node selection on the strong edges in the gradient map. Furthermore, the selected nodes are evaluated by stereo matching error and validated by cross validation for nodal disparity. Each node point is iteratively moving along the pixels with high gradient value instead every point to find the optimal node position. Therefore, our approach provides a fast mesh optimization as well as a reliable image quality. The experimental results show that the proposed approach provides computational efficiency and high PSNR of prediction image compared to previous methods.",
author = "Ilkwon Park and Hyeran Byun",
year = "2009",
month = "10",
day = "19",
doi = "10.1109/3DTV.2009.5069672",
language = "English",
isbn = "9781424443185",
booktitle = "3DTV-CON 2009 - 3rd 3DTV-Conference",

}

Park, I & Byun, H 2009, Fast gradient-based mesh generation method for the stereo image representation. in 3DTV-CON 2009 - 3rd 3DTV-Conference: The True Vision - Capture, Transmission and Display of 3D Video, Proceedings., 5069672, 3rd 3DTV-Conference: The True Vision - Capture, Transmission and Display of 3D Video, 3DTV-CON 2009, Potsdam, Germany, 09/5/4. https://doi.org/10.1109/3DTV.2009.5069672

Fast gradient-based mesh generation method for the stereo image representation. / Park, Ilkwon; Byun, Hyeran.

3DTV-CON 2009 - 3rd 3DTV-Conference: The True Vision - Capture, Transmission and Display of 3D Video, Proceedings. 2009. 5069672.

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

TY - GEN

T1 - Fast gradient-based mesh generation method for the stereo image representation

AU - Park, Ilkwon

AU - Byun, Hyeran

PY - 2009/10/19

Y1 - 2009/10/19

N2 - This paper proposes a fast gradient-based mesh generation method for the stereo image representation. In our approach, right image in stereo image and intermediate views are fundamentally predicted and synthesized from left image by 2D image warping using regular mesh and vice versa. To overcome texture distortion on object boundaries and preserve texture smoothness of homogenous areas, we propose node selection on the strong edges in the gradient map. Furthermore, the selected nodes are evaluated by stereo matching error and validated by cross validation for nodal disparity. Each node point is iteratively moving along the pixels with high gradient value instead every point to find the optimal node position. Therefore, our approach provides a fast mesh optimization as well as a reliable image quality. The experimental results show that the proposed approach provides computational efficiency and high PSNR of prediction image compared to previous methods.

AB - This paper proposes a fast gradient-based mesh generation method for the stereo image representation. In our approach, right image in stereo image and intermediate views are fundamentally predicted and synthesized from left image by 2D image warping using regular mesh and vice versa. To overcome texture distortion on object boundaries and preserve texture smoothness of homogenous areas, we propose node selection on the strong edges in the gradient map. Furthermore, the selected nodes are evaluated by stereo matching error and validated by cross validation for nodal disparity. Each node point is iteratively moving along the pixels with high gradient value instead every point to find the optimal node position. Therefore, our approach provides a fast mesh optimization as well as a reliable image quality. The experimental results show that the proposed approach provides computational efficiency and high PSNR of prediction image compared to previous methods.

UR - http://www.scopus.com/inward/record.url?scp=70349897500&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70349897500&partnerID=8YFLogxK

U2 - 10.1109/3DTV.2009.5069672

DO - 10.1109/3DTV.2009.5069672

M3 - Conference contribution

AN - SCOPUS:70349897500

SN - 9781424443185

BT - 3DTV-CON 2009 - 3rd 3DTV-Conference

ER -

Park I, Byun H. Fast gradient-based mesh generation method for the stereo image representation. In 3DTV-CON 2009 - 3rd 3DTV-Conference: The True Vision - Capture, Transmission and Display of 3D Video, Proceedings. 2009. 5069672 https://doi.org/10.1109/3DTV.2009.5069672