Exploiting self-similarities for single frame super-resolution

Chih Yuan Yang, Jia Bin Huang, Ming Hsuan Yang

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

108 Citations (Scopus)

Abstract

We propose a super-resolution method that exploits self-similarities and group structural information of image patches using only one single input frame. The super-resolution problem is posed as learning the mapping between pairs of low-resolution and high-resolution image patches. Instead of relying on an extrinsic set of training images as often required in example-based super-resolution algorithms, we employ a method that generates image pairs directly from the image pyramid of one single frame. The generated patch pairs are clustered for training a dictionary by enforcing group sparsity constraints underlying the image patches. Super-resolution images are then constructed using the learned dictionary. Experimental results show the proposed method is able to achieve the state-of-the-art performance.

Original languageEnglish
Title of host publicationComputer Vision, ACCV 2010 - 10th Asian Conference on Computer Vision, Revised Selected Papers
Pages497-510
Number of pages14
EditionPART 3
DOIs
Publication statusPublished - 2011 Mar 16
Event10th Asian Conference on Computer Vision, ACCV 2010 - Queenstown, New Zealand
Duration: 2010 Nov 82010 Nov 12

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
NumberPART 3
Volume6494 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference10th Asian Conference on Computer Vision, ACCV 2010
CountryNew Zealand
CityQueenstown
Period10/11/810/11/12

Fingerprint

Super-resolution
Self-similarity
Glossaries
Image resolution
Optical resolving power
Patch
Pyramid
Sparsity
High Resolution
Experimental Results

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Yang, C. Y., Huang, J. B., & Yang, M. H. (2011). Exploiting self-similarities for single frame super-resolution. In Computer Vision, ACCV 2010 - 10th Asian Conference on Computer Vision, Revised Selected Papers (PART 3 ed., pp. 497-510). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 6494 LNCS, No. PART 3). https://doi.org/10.1007/978-3-642-19318-7_39
Yang, Chih Yuan ; Huang, Jia Bin ; Yang, Ming Hsuan. / Exploiting self-similarities for single frame super-resolution. Computer Vision, ACCV 2010 - 10th Asian Conference on Computer Vision, Revised Selected Papers. PART 3. ed. 2011. pp. 497-510 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 3).
@inproceedings{7a8a73baad764cd7982a677040059787,
title = "Exploiting self-similarities for single frame super-resolution",
abstract = "We propose a super-resolution method that exploits self-similarities and group structural information of image patches using only one single input frame. The super-resolution problem is posed as learning the mapping between pairs of low-resolution and high-resolution image patches. Instead of relying on an extrinsic set of training images as often required in example-based super-resolution algorithms, we employ a method that generates image pairs directly from the image pyramid of one single frame. The generated patch pairs are clustered for training a dictionary by enforcing group sparsity constraints underlying the image patches. Super-resolution images are then constructed using the learned dictionary. Experimental results show the proposed method is able to achieve the state-of-the-art performance.",
author = "Yang, {Chih Yuan} and Huang, {Jia Bin} and Yang, {Ming Hsuan}",
year = "2011",
month = "3",
day = "16",
doi = "10.1007/978-3-642-19318-7_39",
language = "English",
isbn = "9783642193170",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
number = "PART 3",
pages = "497--510",
booktitle = "Computer Vision, ACCV 2010 - 10th Asian Conference on Computer Vision, Revised Selected Papers",
edition = "PART 3",

}

Yang, CY, Huang, JB & Yang, MH 2011, Exploiting self-similarities for single frame super-resolution. in Computer Vision, ACCV 2010 - 10th Asian Conference on Computer Vision, Revised Selected Papers. PART 3 edn, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), no. PART 3, vol. 6494 LNCS, pp. 497-510, 10th Asian Conference on Computer Vision, ACCV 2010, Queenstown, New Zealand, 10/11/8. https://doi.org/10.1007/978-3-642-19318-7_39

Exploiting self-similarities for single frame super-resolution. / Yang, Chih Yuan; Huang, Jia Bin; Yang, Ming Hsuan.

Computer Vision, ACCV 2010 - 10th Asian Conference on Computer Vision, Revised Selected Papers. PART 3. ed. 2011. p. 497-510 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 6494 LNCS, No. PART 3).

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

TY - GEN

T1 - Exploiting self-similarities for single frame super-resolution

AU - Yang, Chih Yuan

AU - Huang, Jia Bin

AU - Yang, Ming Hsuan

PY - 2011/3/16

Y1 - 2011/3/16

N2 - We propose a super-resolution method that exploits self-similarities and group structural information of image patches using only one single input frame. The super-resolution problem is posed as learning the mapping between pairs of low-resolution and high-resolution image patches. Instead of relying on an extrinsic set of training images as often required in example-based super-resolution algorithms, we employ a method that generates image pairs directly from the image pyramid of one single frame. The generated patch pairs are clustered for training a dictionary by enforcing group sparsity constraints underlying the image patches. Super-resolution images are then constructed using the learned dictionary. Experimental results show the proposed method is able to achieve the state-of-the-art performance.

AB - We propose a super-resolution method that exploits self-similarities and group structural information of image patches using only one single input frame. The super-resolution problem is posed as learning the mapping between pairs of low-resolution and high-resolution image patches. Instead of relying on an extrinsic set of training images as often required in example-based super-resolution algorithms, we employ a method that generates image pairs directly from the image pyramid of one single frame. The generated patch pairs are clustered for training a dictionary by enforcing group sparsity constraints underlying the image patches. Super-resolution images are then constructed using the learned dictionary. Experimental results show the proposed method is able to achieve the state-of-the-art performance.

UR - http://www.scopus.com/inward/record.url?scp=79952529859&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79952529859&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-19318-7_39

DO - 10.1007/978-3-642-19318-7_39

M3 - Conference contribution

AN - SCOPUS:79952529859

SN - 9783642193170

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 497

EP - 510

BT - Computer Vision, ACCV 2010 - 10th Asian Conference on Computer Vision, Revised Selected Papers

ER -

Yang CY, Huang JB, Yang MH. Exploiting self-similarities for single frame super-resolution. In Computer Vision, ACCV 2010 - 10th Asian Conference on Computer Vision, Revised Selected Papers. PART 3 ed. 2011. p. 497-510. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); PART 3). https://doi.org/10.1007/978-3-642-19318-7_39