Tampere University of Technology

TUTCRIS Research Portal

Image interpolation based on non-local geometric similarities and directional gradients

Research output: Contribution to journalArticleScientificpeer-review

Standard

Image interpolation based on non-local geometric similarities and directional gradients. / Zhu, Shuyuan; Zeng, Bing; Zeng, Liaoyuan; Gabbouj, Moncef.

In: IEEE Transactions on Multimedia, Vol. 18, No. 9, 01.09.2016, p. 1707-1719.

Research output: Contribution to journalArticleScientificpeer-review

Harvard

Zhu, S, Zeng, B, Zeng, L & Gabbouj, M 2016, 'Image interpolation based on non-local geometric similarities and directional gradients', IEEE Transactions on Multimedia, vol. 18, no. 9, pp. 1707-1719. https://doi.org/10.1109/TMM.2016.2593039

APA

Vancouver

Author

Zhu, Shuyuan ; Zeng, Bing ; Zeng, Liaoyuan ; Gabbouj, Moncef. / Image interpolation based on non-local geometric similarities and directional gradients. In: IEEE Transactions on Multimedia. 2016 ; Vol. 18, No. 9. pp. 1707-1719.

Bibtex - Download

@article{82789a52a02b4cc1be5bc9d22cd8829a,
title = "Image interpolation based on non-local geometric similarities and directional gradients",
abstract = "Image interpolation offers an efficient way to compose a high-resolution (HR) image from the observed low-resolution (LR) image. Advanced interpolation techniques design the interpolation weighting coefficients by solving a minimum mean-square-error (MMSE) problem in which the local geometric similarity is often considered. However, using local geometric similarities cannot usually make the MMSE-based interpolation as reliable as expected. To solve this problem, we propose a robust interpolation scheme by using the nonlocal geometric similarities to construct the HR image. In our proposed method, the MMSE-based interpolation weighting coefficients are generated by solving a regularized least squares problem that is built upon a number of dual-reference patches drawn from the given LR image and regularized by the directional gradients of these patches. Experimental results demonstrate that our proposed method offers a remarkable quality improvement as compared to some state-of-the-art methods, both objectively and subjectively.",
keywords = "Directional gradient, geometric similarity, image interpolation, minimum mean square error (MMSE)",
author = "Shuyuan Zhu and Bing Zeng and Liaoyuan Zeng and Moncef Gabbouj",
year = "2016",
month = "9",
day = "1",
doi = "10.1109/TMM.2016.2593039",
language = "English",
volume = "18",
pages = "1707--1719",
journal = "IEEE Transactions on Multimedia",
issn = "1520-9210",
publisher = "Institute of Electrical and Electronics Engineers",
number = "9",

}

RIS (suitable for import to EndNote) - Download

TY - JOUR

T1 - Image interpolation based on non-local geometric similarities and directional gradients

AU - Zhu, Shuyuan

AU - Zeng, Bing

AU - Zeng, Liaoyuan

AU - Gabbouj, Moncef

PY - 2016/9/1

Y1 - 2016/9/1

N2 - Image interpolation offers an efficient way to compose a high-resolution (HR) image from the observed low-resolution (LR) image. Advanced interpolation techniques design the interpolation weighting coefficients by solving a minimum mean-square-error (MMSE) problem in which the local geometric similarity is often considered. However, using local geometric similarities cannot usually make the MMSE-based interpolation as reliable as expected. To solve this problem, we propose a robust interpolation scheme by using the nonlocal geometric similarities to construct the HR image. In our proposed method, the MMSE-based interpolation weighting coefficients are generated by solving a regularized least squares problem that is built upon a number of dual-reference patches drawn from the given LR image and regularized by the directional gradients of these patches. Experimental results demonstrate that our proposed method offers a remarkable quality improvement as compared to some state-of-the-art methods, both objectively and subjectively.

AB - Image interpolation offers an efficient way to compose a high-resolution (HR) image from the observed low-resolution (LR) image. Advanced interpolation techniques design the interpolation weighting coefficients by solving a minimum mean-square-error (MMSE) problem in which the local geometric similarity is often considered. However, using local geometric similarities cannot usually make the MMSE-based interpolation as reliable as expected. To solve this problem, we propose a robust interpolation scheme by using the nonlocal geometric similarities to construct the HR image. In our proposed method, the MMSE-based interpolation weighting coefficients are generated by solving a regularized least squares problem that is built upon a number of dual-reference patches drawn from the given LR image and regularized by the directional gradients of these patches. Experimental results demonstrate that our proposed method offers a remarkable quality improvement as compared to some state-of-the-art methods, both objectively and subjectively.

KW - Directional gradient

KW - geometric similarity

KW - image interpolation

KW - minimum mean square error (MMSE)

U2 - 10.1109/TMM.2016.2593039

DO - 10.1109/TMM.2016.2593039

M3 - Article

VL - 18

SP - 1707

EP - 1719

JO - IEEE Transactions on Multimedia

JF - IEEE Transactions on Multimedia

SN - 1520-9210

IS - 9

ER -