Twin deep convolutional neural network for example-based image colorization

Domonkos Varga, T. Szirányi

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

1 Citation (Scopus)

Abstract

This paper deals with the colorization of grayscale images. Recent papers have shown remarkable results on image colorization utilizing various deep architectures. Unlike previous methods, we perform colorization using a deep architecture and a reference image. Our architecture utilizes two parallel Convolutional Neural Networks which have the same structure. One CNN, which uses the reference image, helps the other CNN in color prediction for the input image. On the other hand, the second CNN, which uses the input image, helps to identify the areas which holds essential information about the color scheme of the scene. Comprehensive experiments and qualitative and quantitative evaluations were conducted on the images of SUN database and on other images. Quantitative evaluations are based on Peak Signal-to-Noise Ratio (PSNR) and on Quaternion Structural Similarity (QSSIM).

Original languageEnglish
Title of host publicationComputer Analysis of Images and Patterns - 17th International Conference, CAIP 2017, Proceedings
PublisherSpringer Verlag
Pages184-195
Number of pages12
ISBN (Print)9783319646886
DOIs
Publication statusPublished - Jan 1 2017
Event17th International Conference on Computer Analysis of Images and Patterns, CAIP 2017 - Ystad, Sweden
Duration: Aug 22 2017Aug 24 2017

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume10424 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other17th International Conference on Computer Analysis of Images and Patterns, CAIP 2017
CountrySweden
CityYstad
Period8/22/178/24/17

Fingerprint

Neural Networks
Color
Neural networks
Signal to noise ratio
Quantitative Evaluation
Experiments
Structural Similarity
Quaternion
Prediction
Experiment
Architecture

Keywords

  • Convolutional neural network
  • Deep learning
  • Image colorization

ASJC Scopus subject areas

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Varga, D., & Szirányi, T. (2017). Twin deep convolutional neural network for example-based image colorization. In Computer Analysis of Images and Patterns - 17th International Conference, CAIP 2017, Proceedings (pp. 184-195). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10424 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-64689-3_15

Twin deep convolutional neural network for example-based image colorization. / Varga, Domonkos; Szirányi, T.

Computer Analysis of Images and Patterns - 17th International Conference, CAIP 2017, Proceedings. Springer Verlag, 2017. p. 184-195 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 10424 LNCS).

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

Varga, D & Szirányi, T 2017, Twin deep convolutional neural network for example-based image colorization. in Computer Analysis of Images and Patterns - 17th International Conference, CAIP 2017, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 10424 LNCS, Springer Verlag, pp. 184-195, 17th International Conference on Computer Analysis of Images and Patterns, CAIP 2017, Ystad, Sweden, 8/22/17. https://doi.org/10.1007/978-3-319-64689-3_15
Varga D, Szirányi T. Twin deep convolutional neural network for example-based image colorization. In Computer Analysis of Images and Patterns - 17th International Conference, CAIP 2017, Proceedings. Springer Verlag. 2017. p. 184-195. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-64689-3_15
Varga, Domonkos ; Szirányi, T. / Twin deep convolutional neural network for example-based image colorization. Computer Analysis of Images and Patterns - 17th International Conference, CAIP 2017, Proceedings. Springer Verlag, 2017. pp. 184-195 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{2bb2b7bcd77141ffbd52d5f20c2b4624,
title = "Twin deep convolutional neural network for example-based image colorization",
abstract = "This paper deals with the colorization of grayscale images. Recent papers have shown remarkable results on image colorization utilizing various deep architectures. Unlike previous methods, we perform colorization using a deep architecture and a reference image. Our architecture utilizes two parallel Convolutional Neural Networks which have the same structure. One CNN, which uses the reference image, helps the other CNN in color prediction for the input image. On the other hand, the second CNN, which uses the input image, helps to identify the areas which holds essential information about the color scheme of the scene. Comprehensive experiments and qualitative and quantitative evaluations were conducted on the images of SUN database and on other images. Quantitative evaluations are based on Peak Signal-to-Noise Ratio (PSNR) and on Quaternion Structural Similarity (QSSIM).",
keywords = "Convolutional neural network, Deep learning, Image colorization",
author = "Domonkos Varga and T. Szir{\'a}nyi",
year = "2017",
month = "1",
day = "1",
doi = "10.1007/978-3-319-64689-3_15",
language = "English",
isbn = "9783319646886",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "184--195",
booktitle = "Computer Analysis of Images and Patterns - 17th International Conference, CAIP 2017, Proceedings",

}

TY - GEN

T1 - Twin deep convolutional neural network for example-based image colorization

AU - Varga, Domonkos

AU - Szirányi, T.

PY - 2017/1/1

Y1 - 2017/1/1

N2 - This paper deals with the colorization of grayscale images. Recent papers have shown remarkable results on image colorization utilizing various deep architectures. Unlike previous methods, we perform colorization using a deep architecture and a reference image. Our architecture utilizes two parallel Convolutional Neural Networks which have the same structure. One CNN, which uses the reference image, helps the other CNN in color prediction for the input image. On the other hand, the second CNN, which uses the input image, helps to identify the areas which holds essential information about the color scheme of the scene. Comprehensive experiments and qualitative and quantitative evaluations were conducted on the images of SUN database and on other images. Quantitative evaluations are based on Peak Signal-to-Noise Ratio (PSNR) and on Quaternion Structural Similarity (QSSIM).

AB - This paper deals with the colorization of grayscale images. Recent papers have shown remarkable results on image colorization utilizing various deep architectures. Unlike previous methods, we perform colorization using a deep architecture and a reference image. Our architecture utilizes two parallel Convolutional Neural Networks which have the same structure. One CNN, which uses the reference image, helps the other CNN in color prediction for the input image. On the other hand, the second CNN, which uses the input image, helps to identify the areas which holds essential information about the color scheme of the scene. Comprehensive experiments and qualitative and quantitative evaluations were conducted on the images of SUN database and on other images. Quantitative evaluations are based on Peak Signal-to-Noise Ratio (PSNR) and on Quaternion Structural Similarity (QSSIM).

KW - Convolutional neural network

KW - Deep learning

KW - Image colorization

UR - http://www.scopus.com/inward/record.url?scp=85028507525&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85028507525&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-64689-3_15

DO - 10.1007/978-3-319-64689-3_15

M3 - Conference contribution

SN - 9783319646886

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

SP - 184

EP - 195

BT - Computer Analysis of Images and Patterns - 17th International Conference, CAIP 2017, Proceedings

PB - Springer Verlag

ER -