Parametric texture model based on joint statistics of complex wavelet coefficients

Javier Portilla, Eero Simoncelli

Research output: Contribution to journalArticle

Abstract

We present a universal statistical model for texture images in the context of an overcomplete complex wavelet transform. The model is parameterized by a set of statistics computed on pairs of coefficients corresponding to basis functions at adjacent spatial locations, orientations, and scales. We develop an efficient algorithm for synthesizing random images subject to these constraints, by iteratively projecting onto the set of images satisfying each constraint, and we use this to test the perceptual validity of the model. In particular, we demonstrate the necessity of subgroups of the parameter set by showing examples of texture synthesis that fail when those parameters are removed from the set. We also demonstrate the power of our model by successfully synthesizing examples drawn from a diverse collection of artificial and natural textures.

Original languageEnglish (US)
Pages (from-to)49-71
Number of pages23
JournalInternational Journal of Computer Vision
Volume40
Issue number1
DOIs
StatePublished - Oct 2000

Fingerprint

Textures
Statistics
Wavelet transforms
Statistical Models

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Vision and Pattern Recognition
  • Control and Systems Engineering

Cite this

Parametric texture model based on joint statistics of complex wavelet coefficients. / Portilla, Javier; Simoncelli, Eero.

In: International Journal of Computer Vision, Vol. 40, No. 1, 10.2000, p. 49-71.

Research output: Contribution to journalArticle

@article{2910a62b800740cfb450df22b079575a,
title = "Parametric texture model based on joint statistics of complex wavelet coefficients",
abstract = "We present a universal statistical model for texture images in the context of an overcomplete complex wavelet transform. The model is parameterized by a set of statistics computed on pairs of coefficients corresponding to basis functions at adjacent spatial locations, orientations, and scales. We develop an efficient algorithm for synthesizing random images subject to these constraints, by iteratively projecting onto the set of images satisfying each constraint, and we use this to test the perceptual validity of the model. In particular, we demonstrate the necessity of subgroups of the parameter set by showing examples of texture synthesis that fail when those parameters are removed from the set. We also demonstrate the power of our model by successfully synthesizing examples drawn from a diverse collection of artificial and natural textures.",
author = "Javier Portilla and Eero Simoncelli",
year = "2000",
month = "10",
doi = "10.1023/A:1026553619983",
language = "English (US)",
volume = "40",
pages = "49--71",
journal = "International Journal of Computer Vision",
issn = "0920-5691",
publisher = "Springer Netherlands",
number = "1",

}

TY - JOUR

T1 - Parametric texture model based on joint statistics of complex wavelet coefficients

AU - Portilla, Javier

AU - Simoncelli, Eero

PY - 2000/10

Y1 - 2000/10

N2 - We present a universal statistical model for texture images in the context of an overcomplete complex wavelet transform. The model is parameterized by a set of statistics computed on pairs of coefficients corresponding to basis functions at adjacent spatial locations, orientations, and scales. We develop an efficient algorithm for synthesizing random images subject to these constraints, by iteratively projecting onto the set of images satisfying each constraint, and we use this to test the perceptual validity of the model. In particular, we demonstrate the necessity of subgroups of the parameter set by showing examples of texture synthesis that fail when those parameters are removed from the set. We also demonstrate the power of our model by successfully synthesizing examples drawn from a diverse collection of artificial and natural textures.

AB - We present a universal statistical model for texture images in the context of an overcomplete complex wavelet transform. The model is parameterized by a set of statistics computed on pairs of coefficients corresponding to basis functions at adjacent spatial locations, orientations, and scales. We develop an efficient algorithm for synthesizing random images subject to these constraints, by iteratively projecting onto the set of images satisfying each constraint, and we use this to test the perceptual validity of the model. In particular, we demonstrate the necessity of subgroups of the parameter set by showing examples of texture synthesis that fail when those parameters are removed from the set. We also demonstrate the power of our model by successfully synthesizing examples drawn from a diverse collection of artificial and natural textures.

UR - http://www.scopus.com/inward/record.url?scp=0034291204&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0034291204&partnerID=8YFLogxK

U2 - 10.1023/A:1026553619983

DO - 10.1023/A:1026553619983

M3 - Article

VL - 40

SP - 49

EP - 71

JO - International Journal of Computer Vision

JF - International Journal of Computer Vision

SN - 0920-5691

IS - 1

ER -