Learning large-scale automatic image colorization

Aditya Deshpande, Jason Rock, David Alexander Forsyth

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

Abstract

We describe an automated method for image colorization that learns to colorize from examples. Our method exploits a LEARCH framework to train a quadratic objective function in the chromaticity maps, comparable to a Gaussian random field. The coefficients of the objective function are conditioned on image features, using a random forest. The objective function admits correlations on long spatial scales, and can control spatial error in the colorization of the image. Images are then colorized by minimizing this objective function. We demonstrate that our method strongly outperforms a natural baseline on large-scale experiments with images of real scenes using a demanding loss function. We demonstrate that learning a model that is conditioned on scene produces improved results. We show how to incorporate a desired color histogram into the objective function, and that doing so can lead to further improvements in results.

Original languageEnglish (US)
Title of host publication2015 International Conference on Computer Vision, ICCV 2015
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages567-575
Number of pages9
ISBN (Electronic)9781467383912
DOIs
StatePublished - Feb 17 2015
Event15th IEEE International Conference on Computer Vision, ICCV 2015 - Santiago, Chile
Duration: Dec 11 2015Dec 18 2015

Publication series

NameProceedings of the IEEE International Conference on Computer Vision
Volume2015 International Conference on Computer Vision, ICCV 2015
ISSN (Print)1550-5499

Other

Other15th IEEE International Conference on Computer Vision, ICCV 2015
CountryChile
CitySantiago
Period12/11/1512/18/15

Fingerprint

Color
Experiments

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition

Cite this

Deshpande, A., Rock, J., & Forsyth, D. A. (2015). Learning large-scale automatic image colorization. In 2015 International Conference on Computer Vision, ICCV 2015 (pp. 567-575). [7410429] (Proceedings of the IEEE International Conference on Computer Vision; Vol. 2015 International Conference on Computer Vision, ICCV 2015). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/ICCV.2015.72

Learning large-scale automatic image colorization. / Deshpande, Aditya; Rock, Jason; Forsyth, David Alexander.

2015 International Conference on Computer Vision, ICCV 2015. Institute of Electrical and Electronics Engineers Inc., 2015. p. 567-575 7410429 (Proceedings of the IEEE International Conference on Computer Vision; Vol. 2015 International Conference on Computer Vision, ICCV 2015).

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

Deshpande, A, Rock, J & Forsyth, DA 2015, Learning large-scale automatic image colorization. in 2015 International Conference on Computer Vision, ICCV 2015., 7410429, Proceedings of the IEEE International Conference on Computer Vision, vol. 2015 International Conference on Computer Vision, ICCV 2015, Institute of Electrical and Electronics Engineers Inc., pp. 567-575, 15th IEEE International Conference on Computer Vision, ICCV 2015, Santiago, Chile, 12/11/15. https://doi.org/10.1109/ICCV.2015.72
Deshpande A, Rock J, Forsyth DA. Learning large-scale automatic image colorization. In 2015 International Conference on Computer Vision, ICCV 2015. Institute of Electrical and Electronics Engineers Inc. 2015. p. 567-575. 7410429. (Proceedings of the IEEE International Conference on Computer Vision). https://doi.org/10.1109/ICCV.2015.72
Deshpande, Aditya ; Rock, Jason ; Forsyth, David Alexander. / Learning large-scale automatic image colorization. 2015 International Conference on Computer Vision, ICCV 2015. Institute of Electrical and Electronics Engineers Inc., 2015. pp. 567-575 (Proceedings of the IEEE International Conference on Computer Vision).
@inproceedings{e579c9e4fee042b2a608df02bc1405cd,
title = "Learning large-scale automatic image colorization",
abstract = "We describe an automated method for image colorization that learns to colorize from examples. Our method exploits a LEARCH framework to train a quadratic objective function in the chromaticity maps, comparable to a Gaussian random field. The coefficients of the objective function are conditioned on image features, using a random forest. The objective function admits correlations on long spatial scales, and can control spatial error in the colorization of the image. Images are then colorized by minimizing this objective function. We demonstrate that our method strongly outperforms a natural baseline on large-scale experiments with images of real scenes using a demanding loss function. We demonstrate that learning a model that is conditioned on scene produces improved results. We show how to incorporate a desired color histogram into the objective function, and that doing so can lead to further improvements in results.",
author = "Aditya Deshpande and Jason Rock and Forsyth, {David Alexander}",
year = "2015",
month = "2",
day = "17",
doi = "10.1109/ICCV.2015.72",
language = "English (US)",
series = "Proceedings of the IEEE International Conference on Computer Vision",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "567--575",
booktitle = "2015 International Conference on Computer Vision, ICCV 2015",
address = "United States",

}

TY - GEN

T1 - Learning large-scale automatic image colorization

AU - Deshpande, Aditya

AU - Rock, Jason

AU - Forsyth, David Alexander

PY - 2015/2/17

Y1 - 2015/2/17

N2 - We describe an automated method for image colorization that learns to colorize from examples. Our method exploits a LEARCH framework to train a quadratic objective function in the chromaticity maps, comparable to a Gaussian random field. The coefficients of the objective function are conditioned on image features, using a random forest. The objective function admits correlations on long spatial scales, and can control spatial error in the colorization of the image. Images are then colorized by minimizing this objective function. We demonstrate that our method strongly outperforms a natural baseline on large-scale experiments with images of real scenes using a demanding loss function. We demonstrate that learning a model that is conditioned on scene produces improved results. We show how to incorporate a desired color histogram into the objective function, and that doing so can lead to further improvements in results.

AB - We describe an automated method for image colorization that learns to colorize from examples. Our method exploits a LEARCH framework to train a quadratic objective function in the chromaticity maps, comparable to a Gaussian random field. The coefficients of the objective function are conditioned on image features, using a random forest. The objective function admits correlations on long spatial scales, and can control spatial error in the colorization of the image. Images are then colorized by minimizing this objective function. We demonstrate that our method strongly outperforms a natural baseline on large-scale experiments with images of real scenes using a demanding loss function. We demonstrate that learning a model that is conditioned on scene produces improved results. We show how to incorporate a desired color histogram into the objective function, and that doing so can lead to further improvements in results.

UR - http://www.scopus.com/inward/record.url?scp=84973922960&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84973922960&partnerID=8YFLogxK

U2 - 10.1109/ICCV.2015.72

DO - 10.1109/ICCV.2015.72

M3 - Conference contribution

AN - SCOPUS:84973922960

T3 - Proceedings of the IEEE International Conference on Computer Vision

SP - 567

EP - 575

BT - 2015 International Conference on Computer Vision, ICCV 2015

PB - Institute of Electrical and Electronics Engineers Inc.

ER -