Super-resolution by prediction based sub-pel motion estimation

Damith J. Mudugamuwa; Xiangjian He; Daming Wei; Chung Hyun Ahn

doi:10.1109/IVCNZ.2009.5378397

Super-resolution by prediction based sub-pel motion estimation

Damith J. Mudugamuwa, Xiangjian He, Daming Wei, Chung Hyun Ahn

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

2 Citations (Scopus)

Abstract

Super-resolution reconstruction produces high-resolution images from a set of low-resolution images of the same scene. In the last two and a half decades, many super-resolution algorithms have been proposed. These algorithms are usually very sensitive to their assumed models of data and noise, and also computationally expensive for many practical applications. In this paper we adopt computationally efficient prediction based sub-pel motion estimation to produce a fast super-resolution reconstruction that can also accommodate generic motion patterns. The proposed algorithm adaptively exploits the available high frequency content in adjacent video frames to generate high resolution video frames. Initial experiments showed promising results of around 2dB, PSNR improvement, over single frame bi-linear interpolation.

Original language	English
Title of host publication	2009 24th International Conference Image and Vision Computing New Zealand, IVCNZ 2009 - Conference Proceedings
Pages	282-287
Number of pages	6
DOIs	https://doi.org/10.1109/IVCNZ.2009.5378397
Publication status	Published - 2009
Externally published	Yes
Event	2009 24th International Conference Image and Vision Computing New Zealand, IVCNZ 2009 - Wellington, New Zealand Duration: 23 Nov 2009 → 25 Nov 2009

Publication series

Name	2009 24th International Conference Image and Vision Computing New Zealand, IVCNZ 2009 - Conference Proceedings

Conference

Conference	2009 24th International Conference Image and Vision Computing New Zealand, IVCNZ 2009
Country/Territory	New Zealand
City	Wellington
Period	23/11/09 → 25/11/09

Keywords

Image de-blurring
Image registration
Image restoration
Motion estimation
Super-resolution

ASJC Scopus subject areas

Computational Theory and Mathematics
Computer Vision and Pattern Recognition

Access to Document

10.1109/IVCNZ.2009.5378397

Cite this

Mudugamuwa, D. J., He, X., Wei, D., & Ahn, C. H. (2009). Super-resolution by prediction based sub-pel motion estimation. In 2009 24th International Conference Image and Vision Computing New Zealand, IVCNZ 2009 - Conference Proceedings (pp. 282-287). Article 5378397 (2009 24th International Conference Image and Vision Computing New Zealand, IVCNZ 2009 - Conference Proceedings). https://doi.org/10.1109/IVCNZ.2009.5378397

@inproceedings{73d0c471aee946ac8cbc0d345e7fab44,

title = "Super-resolution by prediction based sub-pel motion estimation",

abstract = "Super-resolution reconstruction produces high-resolution images from a set of low-resolution images of the same scene. In the last two and a half decades, many super-resolution algorithms have been proposed. These algorithms are usually very sensitive to their assumed models of data and noise, and also computationally expensive for many practical applications. In this paper we adopt computationally efficient prediction based sub-pel motion estimation to produce a fast super-resolution reconstruction that can also accommodate generic motion patterns. The proposed algorithm adaptively exploits the available high frequency content in adjacent video frames to generate high resolution video frames. Initial experiments showed promising results of around 2dB, PSNR improvement, over single frame bi-linear interpolation.",

keywords = "Image de-blurring, Image registration, Image restoration, Motion estimation, Super-resolution",

author = "Mudugamuwa, {Damith J.} and Xiangjian He and Daming Wei and Ahn, {Chung Hyun}",

year = "2009",

doi = "10.1109/IVCNZ.2009.5378397",

language = "English",

isbn = "9781424446988",

series = "2009 24th International Conference Image and Vision Computing New Zealand, IVCNZ 2009 - Conference Proceedings",

pages = "282--287",

booktitle = "2009 24th International Conference Image and Vision Computing New Zealand, IVCNZ 2009 - Conference Proceedings",

note = "2009 24th International Conference Image and Vision Computing New Zealand, IVCNZ 2009 ; Conference date: 23-11-2009 Through 25-11-2009",

}

Mudugamuwa, DJ, He, X, Wei, D & Ahn, CH 2009, Super-resolution by prediction based sub-pel motion estimation. in 2009 24th International Conference Image and Vision Computing New Zealand, IVCNZ 2009 - Conference Proceedings., 5378397, 2009 24th International Conference Image and Vision Computing New Zealand, IVCNZ 2009 - Conference Proceedings, pp. 282-287, 2009 24th International Conference Image and Vision Computing New Zealand, IVCNZ 2009, Wellington, New Zealand, 23/11/09. https://doi.org/10.1109/IVCNZ.2009.5378397

Super-resolution by prediction based sub-pel motion estimation. / Mudugamuwa, Damith J.; He, Xiangjian; Wei, Daming et al.
2009 24th International Conference Image and Vision Computing New Zealand, IVCNZ 2009 - Conference Proceedings. 2009. p. 282-287 5378397 (2009 24th International Conference Image and Vision Computing New Zealand, IVCNZ 2009 - Conference Proceedings).

Research output: Chapter in Book/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Super-resolution by prediction based sub-pel motion estimation

AU - Mudugamuwa, Damith J.

AU - He, Xiangjian

AU - Wei, Daming

AU - Ahn, Chung Hyun

PY - 2009

Y1 - 2009

N2 - Super-resolution reconstruction produces high-resolution images from a set of low-resolution images of the same scene. In the last two and a half decades, many super-resolution algorithms have been proposed. These algorithms are usually very sensitive to their assumed models of data and noise, and also computationally expensive for many practical applications. In this paper we adopt computationally efficient prediction based sub-pel motion estimation to produce a fast super-resolution reconstruction that can also accommodate generic motion patterns. The proposed algorithm adaptively exploits the available high frequency content in adjacent video frames to generate high resolution video frames. Initial experiments showed promising results of around 2dB, PSNR improvement, over single frame bi-linear interpolation.

AB - Super-resolution reconstruction produces high-resolution images from a set of low-resolution images of the same scene. In the last two and a half decades, many super-resolution algorithms have been proposed. These algorithms are usually very sensitive to their assumed models of data and noise, and also computationally expensive for many practical applications. In this paper we adopt computationally efficient prediction based sub-pel motion estimation to produce a fast super-resolution reconstruction that can also accommodate generic motion patterns. The proposed algorithm adaptively exploits the available high frequency content in adjacent video frames to generate high resolution video frames. Initial experiments showed promising results of around 2dB, PSNR improvement, over single frame bi-linear interpolation.

KW - Image de-blurring

KW - Image registration

KW - Image restoration

KW - Motion estimation

KW - Super-resolution

UR - http://www.scopus.com/inward/record.url?scp=77951165879&partnerID=8YFLogxK

U2 - 10.1109/IVCNZ.2009.5378397

DO - 10.1109/IVCNZ.2009.5378397

M3 - Conference contribution

AN - SCOPUS:77951165879

SN - 9781424446988

T3 - 2009 24th International Conference Image and Vision Computing New Zealand, IVCNZ 2009 - Conference Proceedings

SP - 282

EP - 287

BT - 2009 24th International Conference Image and Vision Computing New Zealand, IVCNZ 2009 - Conference Proceedings

T2 - 2009 24th International Conference Image and Vision Computing New Zealand, IVCNZ 2009

Y2 - 23 November 2009 through 25 November 2009

ER -

Mudugamuwa DJ, He X, Wei D, Ahn CH. Super-resolution by prediction based sub-pel motion estimation. In 2009 24th International Conference Image and Vision Computing New Zealand, IVCNZ 2009 - Conference Proceedings. 2009. p. 282-287. 5378397. (2009 24th International Conference Image and Vision Computing New Zealand, IVCNZ 2009 - Conference Proceedings). doi: 10.1109/IVCNZ.2009.5378397

Super-resolution by prediction based sub-pel motion estimation

Abstract

Publication series

Conference

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this