Poisson blended exemplar-based texture completion

Hoang M. Nguyen, Burkhard C. Wünsche, Patrice Delmas, Christof Lutteroth

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

1 Citation (Scopus)

Abstract



Image inpainting is the process of correcting undesirable changes to an image in an unobtrusive way. The existing literature in this research field describes predominantly techniques designed to correct narrow missing regions, which thus often produce undesirable results when the damaged region is large.

This paper presents a novel exemplar-based image inpainting technique for automatic filling-in missing region of an image. Our solution offers two major improvements compared to existing techniques. Patches for filling in missing regions are identified using an appearance space vector, which not only encodes pixel colours, but also colour gradients, feature distances and other measures for computing image similarity. In order to speed up the search for a matching patch we use a Principal Component Analysis to reduce the size of a feature vector used for patch comparison.

The second major improvement is the technique used combine patches filling in a missing region. In order to avoid visible seams we use a Poisson-guided interpolation to blend patches.

Our evaluation and comparison with existing techniques demonstrates significantly improved performance for inpainting missing image regions.
Original languageEnglish
Title of host publicationProceedings of the 37rd Australasian Computer Science Conference (ACSC), 2014
Subtitle of host publicationVolume 147
EditorsB. Thomas, D. Parry
Place of PublicationDarlinghurst, Australia
PublisherAssociation for Computing Machinery
Pages99-104
Number of pages6
ISBN (Print)9781921770302
Publication statusPublished - 20 Jan 2014

Fingerprint

Textures
Color
Vector spaces
Principal component analysis
Interpolation
Pixels

Cite this

Nguyen, H. M., Wünsche, B. C., Delmas, P., & Lutteroth, C. (2014). Poisson blended exemplar-based texture completion. In B. Thomas, & D. Parry (Eds.), Proceedings of the 37rd Australasian Computer Science Conference (ACSC), 2014: Volume 147 (pp. 99-104). Darlinghurst, Australia: Association for Computing Machinery.

Poisson blended exemplar-based texture completion. / Nguyen, Hoang M.; Wünsche, Burkhard C.; Delmas, Patrice; Lutteroth, Christof.

Proceedings of the 37rd Australasian Computer Science Conference (ACSC), 2014: Volume 147. ed. / B. Thomas; D. Parry. Darlinghurst, Australia : Association for Computing Machinery, 2014. p. 99-104.

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

Nguyen, HM, Wünsche, BC, Delmas, P & Lutteroth, C 2014, Poisson blended exemplar-based texture completion. in B Thomas & D Parry (eds), Proceedings of the 37rd Australasian Computer Science Conference (ACSC), 2014: Volume 147. Association for Computing Machinery, Darlinghurst, Australia, pp. 99-104.
Nguyen HM, Wünsche BC, Delmas P, Lutteroth C. Poisson blended exemplar-based texture completion. In Thomas B, Parry D, editors, Proceedings of the 37rd Australasian Computer Science Conference (ACSC), 2014: Volume 147. Darlinghurst, Australia: Association for Computing Machinery. 2014. p. 99-104
Nguyen, Hoang M. ; Wünsche, Burkhard C. ; Delmas, Patrice ; Lutteroth, Christof. / Poisson blended exemplar-based texture completion. Proceedings of the 37rd Australasian Computer Science Conference (ACSC), 2014: Volume 147. editor / B. Thomas ; D. Parry. Darlinghurst, Australia : Association for Computing Machinery, 2014. pp. 99-104
@inproceedings{748db425620e43a3bd192a664469ac78,
title = "Poisson blended exemplar-based texture completion",
abstract = "Image inpainting is the process of correcting undesirable changes to an image in an unobtrusive way. The existing literature in this research field describes predominantly techniques designed to correct narrow missing regions, which thus often produce undesirable results when the damaged region is large.This paper presents a novel exemplar-based image inpainting technique for automatic filling-in missing region of an image. Our solution offers two major improvements compared to existing techniques. Patches for filling in missing regions are identified using an appearance space vector, which not only encodes pixel colours, but also colour gradients, feature distances and other measures for computing image similarity. In order to speed up the search for a matching patch we use a Principal Component Analysis to reduce the size of a feature vector used for patch comparison.The second major improvement is the technique used combine patches filling in a missing region. In order to avoid visible seams we use a Poisson-guided interpolation to blend patches.Our evaluation and comparison with existing techniques demonstrates significantly improved performance for inpainting missing image regions.",
author = "Nguyen, {Hoang M.} and W{\"u}nsche, {Burkhard C.} and Patrice Delmas and Christof Lutteroth",
year = "2014",
month = "1",
day = "20",
language = "English",
isbn = "9781921770302",
pages = "99--104",
editor = "B. Thomas and D. Parry",
booktitle = "Proceedings of the 37rd Australasian Computer Science Conference (ACSC), 2014",
publisher = "Association for Computing Machinery",
address = "USA United States",

}

TY - GEN

T1 - Poisson blended exemplar-based texture completion

AU - Nguyen, Hoang M.

AU - Wünsche, Burkhard C.

AU - Delmas, Patrice

AU - Lutteroth, Christof

PY - 2014/1/20

Y1 - 2014/1/20

N2 - Image inpainting is the process of correcting undesirable changes to an image in an unobtrusive way. The existing literature in this research field describes predominantly techniques designed to correct narrow missing regions, which thus often produce undesirable results when the damaged region is large.This paper presents a novel exemplar-based image inpainting technique for automatic filling-in missing region of an image. Our solution offers two major improvements compared to existing techniques. Patches for filling in missing regions are identified using an appearance space vector, which not only encodes pixel colours, but also colour gradients, feature distances and other measures for computing image similarity. In order to speed up the search for a matching patch we use a Principal Component Analysis to reduce the size of a feature vector used for patch comparison.The second major improvement is the technique used combine patches filling in a missing region. In order to avoid visible seams we use a Poisson-guided interpolation to blend patches.Our evaluation and comparison with existing techniques demonstrates significantly improved performance for inpainting missing image regions.

AB - Image inpainting is the process of correcting undesirable changes to an image in an unobtrusive way. The existing literature in this research field describes predominantly techniques designed to correct narrow missing regions, which thus often produce undesirable results when the damaged region is large.This paper presents a novel exemplar-based image inpainting technique for automatic filling-in missing region of an image. Our solution offers two major improvements compared to existing techniques. Patches for filling in missing regions are identified using an appearance space vector, which not only encodes pixel colours, but also colour gradients, feature distances and other measures for computing image similarity. In order to speed up the search for a matching patch we use a Principal Component Analysis to reduce the size of a feature vector used for patch comparison.The second major improvement is the technique used combine patches filling in a missing region. In order to avoid visible seams we use a Poisson-guided interpolation to blend patches.Our evaluation and comparison with existing techniques demonstrates significantly improved performance for inpainting missing image regions.

M3 - Conference contribution

SN - 9781921770302

SP - 99

EP - 104

BT - Proceedings of the 37rd Australasian Computer Science Conference (ACSC), 2014

A2 - Thomas, B.

A2 - Parry, D.

PB - Association for Computing Machinery

CY - Darlinghurst, Australia

ER -