Texture Stationarization: Turning Photos into Tileable Textures

Joep Moritz, Stuart James, Tom Fincham Haines, Tobias Ritschel, Tim Weyrich

Research output: Contribution to conferencePaper

1 Citation (Scopus)
104 Downloads (Pure)

Abstract

Texture synthesis has grown into a mature field in computer graphics, allowing the synthesis of naturalistic textures and images from photographic exemplars. Surprisingly little work, however, has been dedicated to synthesizing tileable textures, that is, textures that when laid out in a regular grid of tiles form a homogeneous appearance suitable for use in memory-sensitive real-time graphics applications. One of the key challenges in doing so is that most natural input exemplars exhibit uneven spatial variations that, when tiled, show as repetitive patterns. We propose an approach to synthesize tileable textures while enforcing stationarity properties that effectively mask repetitions while maintaining the unique characteristics of the exemplar. We explore a number of alternative measures for texture stationarity and show how each measure can be integrated into a standard texture synthesis method (PatchMatch) to enforce stationarity at user-controlled scales. We demonstrate the efficacy of our approach using a database of 118 exemplar images, both from publicly available sources as well as new ones captured under uncontrolled conditions, and we quantitatively analyze alternative stationarity measures for their robustness across many test runs using
different random seeds. In conclusion, we suggest a novel synthesis approach that employs local histogram matching to reliably turn input photographs of natural surfaces into tiles well suited for artifact-free tiling.
Original languageEnglish
Pages177-188
Number of pages12
DOIs
Publication statusPublished - 2017
EventEurographics 2017: Annual Conference of the European Association for Computer Graphics -
Duration: 24 Apr 201728 Apr 2017

Conference

ConferenceEurographics 2017: Annual Conference of the European Association for Computer Graphics
Period24/04/1728/04/17

Fingerprint

Textures
Tile
Computer graphics
Seed
Masks
Data storage equipment

Cite this

Moritz, J., James, S., Fincham Haines, T., Ritschel, T., & Weyrich, T. (2017). Texture Stationarization: Turning Photos into Tileable Textures. 177-188. Paper presented at Eurographics 2017: Annual Conference of the European Association for Computer Graphics, . https://doi.org/10.1111/cgf.13117

Texture Stationarization: Turning Photos into Tileable Textures. / Moritz, Joep; James, Stuart; Fincham Haines, Tom; Ritschel, Tobias; Weyrich, Tim.

2017. 177-188 Paper presented at Eurographics 2017: Annual Conference of the European Association for Computer Graphics, .

Research output: Contribution to conferencePaper

Moritz, J, James, S, Fincham Haines, T, Ritschel, T & Weyrich, T 2017, 'Texture Stationarization: Turning Photos into Tileable Textures' Paper presented at Eurographics 2017: Annual Conference of the European Association for Computer Graphics, 24/04/17 - 28/04/17, pp. 177-188. https://doi.org/10.1111/cgf.13117
Moritz J, James S, Fincham Haines T, Ritschel T, Weyrich T. Texture Stationarization: Turning Photos into Tileable Textures. 2017. Paper presented at Eurographics 2017: Annual Conference of the European Association for Computer Graphics, . https://doi.org/10.1111/cgf.13117
Moritz, Joep ; James, Stuart ; Fincham Haines, Tom ; Ritschel, Tobias ; Weyrich, Tim. / Texture Stationarization: Turning Photos into Tileable Textures. Paper presented at Eurographics 2017: Annual Conference of the European Association for Computer Graphics, .12 p.
@conference{234f5fa8e7ba42c99b74163de0eb721b,
title = "Texture Stationarization: Turning Photos into Tileable Textures",
abstract = "Texture synthesis has grown into a mature field in computer graphics, allowing the synthesis of naturalistic textures and images from photographic exemplars. Surprisingly little work, however, has been dedicated to synthesizing tileable textures, that is, textures that when laid out in a regular grid of tiles form a homogeneous appearance suitable for use in memory-sensitive real-time graphics applications. One of the key challenges in doing so is that most natural input exemplars exhibit uneven spatial variations that, when tiled, show as repetitive patterns. We propose an approach to synthesize tileable textures while enforcing stationarity properties that effectively mask repetitions while maintaining the unique characteristics of the exemplar. We explore a number of alternative measures for texture stationarity and show how each measure can be integrated into a standard texture synthesis method (PatchMatch) to enforce stationarity at user-controlled scales. We demonstrate the efficacy of our approach using a database of 118 exemplar images, both from publicly available sources as well as new ones captured under uncontrolled conditions, and we quantitatively analyze alternative stationarity measures for their robustness across many test runs usingdifferent random seeds. In conclusion, we suggest a novel synthesis approach that employs local histogram matching to reliably turn input photographs of natural surfaces into tiles well suited for artifact-free tiling.",
author = "Joep Moritz and Stuart James and {Fincham Haines}, Tom and Tobias Ritschel and Tim Weyrich",
year = "2017",
doi = "10.1111/cgf.13117",
language = "English",
pages = "177--188",
note = "Eurographics 2017: Annual Conference of the European Association for Computer Graphics ; Conference date: 24-04-2017 Through 28-04-2017",

}

TY - CONF

T1 - Texture Stationarization: Turning Photos into Tileable Textures

AU - Moritz, Joep

AU - James, Stuart

AU - Fincham Haines, Tom

AU - Ritschel, Tobias

AU - Weyrich, Tim

PY - 2017

Y1 - 2017

N2 - Texture synthesis has grown into a mature field in computer graphics, allowing the synthesis of naturalistic textures and images from photographic exemplars. Surprisingly little work, however, has been dedicated to synthesizing tileable textures, that is, textures that when laid out in a regular grid of tiles form a homogeneous appearance suitable for use in memory-sensitive real-time graphics applications. One of the key challenges in doing so is that most natural input exemplars exhibit uneven spatial variations that, when tiled, show as repetitive patterns. We propose an approach to synthesize tileable textures while enforcing stationarity properties that effectively mask repetitions while maintaining the unique characteristics of the exemplar. We explore a number of alternative measures for texture stationarity and show how each measure can be integrated into a standard texture synthesis method (PatchMatch) to enforce stationarity at user-controlled scales. We demonstrate the efficacy of our approach using a database of 118 exemplar images, both from publicly available sources as well as new ones captured under uncontrolled conditions, and we quantitatively analyze alternative stationarity measures for their robustness across many test runs usingdifferent random seeds. In conclusion, we suggest a novel synthesis approach that employs local histogram matching to reliably turn input photographs of natural surfaces into tiles well suited for artifact-free tiling.

AB - Texture synthesis has grown into a mature field in computer graphics, allowing the synthesis of naturalistic textures and images from photographic exemplars. Surprisingly little work, however, has been dedicated to synthesizing tileable textures, that is, textures that when laid out in a regular grid of tiles form a homogeneous appearance suitable for use in memory-sensitive real-time graphics applications. One of the key challenges in doing so is that most natural input exemplars exhibit uneven spatial variations that, when tiled, show as repetitive patterns. We propose an approach to synthesize tileable textures while enforcing stationarity properties that effectively mask repetitions while maintaining the unique characteristics of the exemplar. We explore a number of alternative measures for texture stationarity and show how each measure can be integrated into a standard texture synthesis method (PatchMatch) to enforce stationarity at user-controlled scales. We demonstrate the efficacy of our approach using a database of 118 exemplar images, both from publicly available sources as well as new ones captured under uncontrolled conditions, and we quantitatively analyze alternative stationarity measures for their robustness across many test runs usingdifferent random seeds. In conclusion, we suggest a novel synthesis approach that employs local histogram matching to reliably turn input photographs of natural surfaces into tiles well suited for artifact-free tiling.

UR - https://doi.org/10.1111/cgf.13117

U2 - 10.1111/cgf.13117

DO - 10.1111/cgf.13117

M3 - Paper

SP - 177

EP - 188

ER -