TileGAN: Synthesis of large-scale non-homogeneous textures

Anna Frühstück; Ibraheem Alhashim; Peter Wonka

doi:10.1145/3306346.3322993

TileGAN: Synthesis of large-scale non-homogeneous textures

Anna Frühstück, Ibraheem Alhashim, Peter Wonka

Research output: Contribution to journal › Article › peer-review

55 Scopus citations

Abstract

We tackle the problem of texture synthesis in the setting where many input images are given and a large-scale output is required. We build on recent generative adversarial networks and propose two extensions in this paper. First, we propose an algorithm to combine outputs of GANs trained on a smaller resolution to produce a large-scale plausible texture map with virtually no boundary artifacts. Second, we propose a user interface to enable artistic control. Our quantitative and qualitative results showcase the generation of synthesized high-resolution maps consisting of up to hundreds of megapixels as a case in point.

Original language	English (US)
Pages (from-to)	1-11
Number of pages	11
Journal	ACM Transactions on Graphics
Volume	38
Issue number	4
DOIs	https://doi.org/10.1145/3306346.3322993
State	Published - Jan 1 2019

Bibliographical note

KAUST Repository Item: Exported on 2020-10-01
Acknowledged KAUST grant number(s): URF/1/3426-01-01
Acknowledgements: We would like to thank Tero Karras and his collaborators [2018a] for making their source code available. This work was supported by the KAUST Office of Sponsored Research (OSR) under Award No. URF/1/3730-01-01 and URF/1/3426-01-01.

Access to Document

10.1145/3306346.3322993

https://repository.kaust.edu.sa/bitstream/10754/660123/1/Articlefile1.pdf

Cite this

@article{50e07598862c4b8593f13bf3d96ff21f,

title = "TileGAN: Synthesis of large-scale non-homogeneous textures",

abstract = "We tackle the problem of texture synthesis in the setting where many input images are given and a large-scale output is required. We build on recent generative adversarial networks and propose two extensions in this paper. First, we propose an algorithm to combine outputs of GANs trained on a smaller resolution to produce a large-scale plausible texture map with virtually no boundary artifacts. Second, we propose a user interface to enable artistic control. Our quantitative and qualitative results showcase the generation of synthesized high-resolution maps consisting of up to hundreds of megapixels as a case in point.",

author = "Anna Fr{\"u}hst{\"u}ck and Ibraheem Alhashim and Peter Wonka",

note = "KAUST Repository Item: Exported on 2020-10-01 Acknowledged KAUST grant number(s): URF/1/3426-01-01 Acknowledgements: We would like to thank Tero Karras and his collaborators [2018a] for making their source code available. This work was supported by the KAUST Office of Sponsored Research (OSR) under Award No. URF/1/3730-01-01 and URF/1/3426-01-01.",

year = "2019",

month = jan,

day = "1",

doi = "10.1145/3306346.3322993",

language = "English (US)",

volume = "38",

pages = "1--11",

journal = "ACM Transactions on Graphics",

issn = "0730-0301",

publisher = "Association for Computing Machinery",

number = "4",

}

TY - JOUR

T1 - TileGAN: Synthesis of large-scale non-homogeneous textures

AU - Frühstück, Anna

AU - Alhashim, Ibraheem

AU - Wonka, Peter

N1 - KAUST Repository Item: Exported on 2020-10-01 Acknowledged KAUST grant number(s): URF/1/3426-01-01 Acknowledgements: We would like to thank Tero Karras and his collaborators [2018a] for making their source code available. This work was supported by the KAUST Office of Sponsored Research (OSR) under Award No. URF/1/3730-01-01 and URF/1/3426-01-01.

PY - 2019/1/1

Y1 - 2019/1/1

N2 - We tackle the problem of texture synthesis in the setting where many input images are given and a large-scale output is required. We build on recent generative adversarial networks and propose two extensions in this paper. First, we propose an algorithm to combine outputs of GANs trained on a smaller resolution to produce a large-scale plausible texture map with virtually no boundary artifacts. Second, we propose a user interface to enable artistic control. Our quantitative and qualitative results showcase the generation of synthesized high-resolution maps consisting of up to hundreds of megapixels as a case in point.

AB - We tackle the problem of texture synthesis in the setting where many input images are given and a large-scale output is required. We build on recent generative adversarial networks and propose two extensions in this paper. First, we propose an algorithm to combine outputs of GANs trained on a smaller resolution to produce a large-scale plausible texture map with virtually no boundary artifacts. Second, we propose a user interface to enable artistic control. Our quantitative and qualitative results showcase the generation of synthesized high-resolution maps consisting of up to hundreds of megapixels as a case in point.

UR - http://hdl.handle.net/10754/660123

UR - http://dl.acm.org/citation.cfm?doid=3306346.3322993

UR - http://www.scopus.com/inward/record.url?scp=85073889112&partnerID=8YFLogxK

U2 - 10.1145/3306346.3322993

DO - 10.1145/3306346.3322993

M3 - Article

SN - 0730-0301

VL - 38

SP - 1

EP - 11

JO - ACM Transactions on Graphics

JF - ACM Transactions on Graphics

IS - 4

ER -

TileGAN: Synthesis of large-scale non-homogeneous textures

Abstract

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this