Fast search of a similar patch for self-similarity based image super resolution

Jun Sang Yoo, Ji Hoon Choi, Kang Sun Choi, Dae Yeol Lee, Hui Yong Kim, Jong-Ok Kim

Research output: Contribution to journalArticle

Abstract

In the self-similarity super resolution (SR) approach, similar examples are searched across down-scales in the image pyramid, and the computations of searching similar examples are very heavy. This makes it difficult to work in a real-time way under common software implementation. Therefore, the search process should be further accelerated at an algorithm level. Cauchy-Schwarz inequality has been used previously for fast vector quantization (VQ) encoding. The candidate patches in the search region of SR are analogous to the code-words in the VQ, and Cauchy- Schwarz inequality is exploited to exclude implausible candidate patches early. Consequently, significant acceleration of the similar patch search process is achieved. The proposed method can easily make an optimal trade-off between running speed and visual quality by appropriately configuring the bypass-threshold.

Original languageEnglish
Pages (from-to)2194-2198
Number of pages5
JournalIEICE Transactions on Information and Systems
VolumeE99D
Issue number8
DOIs
Publication statusPublished - 2016 Aug 1

Fingerprint

Vector quantization

Keywords

  • Fast search
  • Kick-out condition
  • Self-similarity
  • Super-resolution

ASJC Scopus subject areas

  • Software
  • Hardware and Architecture
  • Computer Vision and Pattern Recognition
  • Artificial Intelligence
  • Electrical and Electronic Engineering

Cite this

Fast search of a similar patch for self-similarity based image super resolution. / Yoo, Jun Sang; Choi, Ji Hoon; Choi, Kang Sun; Lee, Dae Yeol; Kim, Hui Yong; Kim, Jong-Ok.

In: IEICE Transactions on Information and Systems, Vol. E99D, No. 8, 01.08.2016, p. 2194-2198.

Research output: Contribution to journalArticle

Yoo, Jun Sang ; Choi, Ji Hoon ; Choi, Kang Sun ; Lee, Dae Yeol ; Kim, Hui Yong ; Kim, Jong-Ok. / Fast search of a similar patch for self-similarity based image super resolution. In: IEICE Transactions on Information and Systems. 2016 ; Vol. E99D, No. 8. pp. 2194-2198.
@article{250dd7dc844642f9b17535d8d37f68e3,
title = "Fast search of a similar patch for self-similarity based image super resolution",
abstract = "In the self-similarity super resolution (SR) approach, similar examples are searched across down-scales in the image pyramid, and the computations of searching similar examples are very heavy. This makes it difficult to work in a real-time way under common software implementation. Therefore, the search process should be further accelerated at an algorithm level. Cauchy-Schwarz inequality has been used previously for fast vector quantization (VQ) encoding. The candidate patches in the search region of SR are analogous to the code-words in the VQ, and Cauchy- Schwarz inequality is exploited to exclude implausible candidate patches early. Consequently, significant acceleration of the similar patch search process is achieved. The proposed method can easily make an optimal trade-off between running speed and visual quality by appropriately configuring the bypass-threshold.",
keywords = "Fast search, Kick-out condition, Self-similarity, Super-resolution",
author = "Yoo, {Jun Sang} and Choi, {Ji Hoon} and Choi, {Kang Sun} and Lee, {Dae Yeol} and Kim, {Hui Yong} and Jong-Ok Kim",
year = "2016",
month = "8",
day = "1",
doi = "10.1587/transinf.2016EDL8049",
language = "English",
volume = "E99D",
pages = "2194--2198",
journal = "IEICE Transactions on Information and Systems",
issn = "0916-8532",
publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
number = "8",

}

TY - JOUR

T1 - Fast search of a similar patch for self-similarity based image super resolution

AU - Yoo, Jun Sang

AU - Choi, Ji Hoon

AU - Choi, Kang Sun

AU - Lee, Dae Yeol

AU - Kim, Hui Yong

AU - Kim, Jong-Ok

PY - 2016/8/1

Y1 - 2016/8/1

N2 - In the self-similarity super resolution (SR) approach, similar examples are searched across down-scales in the image pyramid, and the computations of searching similar examples are very heavy. This makes it difficult to work in a real-time way under common software implementation. Therefore, the search process should be further accelerated at an algorithm level. Cauchy-Schwarz inequality has been used previously for fast vector quantization (VQ) encoding. The candidate patches in the search region of SR are analogous to the code-words in the VQ, and Cauchy- Schwarz inequality is exploited to exclude implausible candidate patches early. Consequently, significant acceleration of the similar patch search process is achieved. The proposed method can easily make an optimal trade-off between running speed and visual quality by appropriately configuring the bypass-threshold.

AB - In the self-similarity super resolution (SR) approach, similar examples are searched across down-scales in the image pyramid, and the computations of searching similar examples are very heavy. This makes it difficult to work in a real-time way under common software implementation. Therefore, the search process should be further accelerated at an algorithm level. Cauchy-Schwarz inequality has been used previously for fast vector quantization (VQ) encoding. The candidate patches in the search region of SR are analogous to the code-words in the VQ, and Cauchy- Schwarz inequality is exploited to exclude implausible candidate patches early. Consequently, significant acceleration of the similar patch search process is achieved. The proposed method can easily make an optimal trade-off between running speed and visual quality by appropriately configuring the bypass-threshold.

KW - Fast search

KW - Kick-out condition

KW - Self-similarity

KW - Super-resolution

UR - http://www.scopus.com/inward/record.url?scp=84983088223&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84983088223&partnerID=8YFLogxK

U2 - 10.1587/transinf.2016EDL8049

DO - 10.1587/transinf.2016EDL8049

M3 - Article

AN - SCOPUS:84983088223

VL - E99D

SP - 2194

EP - 2198

JO - IEICE Transactions on Information and Systems

JF - IEICE Transactions on Information and Systems

SN - 0916-8532

IS - 8

ER -