Optimization of paintbrush rendering of images by dynamic MCMC methods

T. Szirányi, Zoltán Tóth

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

6 Citations (Scopus)

Abstract

We have developed a new stochastic image rendering method for the compression, description and segmentation of images. This paintbrush-like image transformation is based on a random searching to insert brush-strokes into a generated image at decreasing scale of brush-sizes, without predefined models or interaction. We introduced a sequential multiscale image decomposition method, based on simulated rectangular-shaped paintbrush strokes. The resulting images look like good-quality paintings with well-defined contours, at an acceptable distortion compared to the original image. The image can be described with the parameters of the consecutive paintbrush strokes, resulting in a parameter-series that can be used for compression. The painting process can be applied for image representation, segmentation and contour detection. Our original method is based on stochastic exhaustive searching which takes a long time of convergence. In this paper we propose a modified algorithm of speed up of about 2x where the faster convergence is supported by a dynamic Metropolis Hastings rule.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
PublisherSpringer Verlag
Pages201-215
Number of pages15
Volume2134
ISBN (Print)3540425233, 9783540425236
DOIs
Publication statusPublished - 2001
Event3rd International Workshop on Energy Minimization Methods in Computer Vision and Pattern Recognition, EMMCVPR 2001 - Sophia Antipolis, France
Duration: Sep 3 2001Sep 5 2001

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume2134
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other3rd International Workshop on Energy Minimization Methods in Computer Vision and Pattern Recognition, EMMCVPR 2001
CountryFrance
CitySophia Antipolis
Period9/3/019/5/01

Fingerprint

MCMC Methods
Painting
Brushes
Rendering
Optimization
Stroke
Decomposition
Compression
Segmentation
Image Transformation
Image Decomposition
Metropolis-Hastings
Image Representation
Decomposition Method
Well-defined
Consecutive
Speedup
Series
Interaction

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Szirányi, T., & Tóth, Z. (2001). Optimization of paintbrush rendering of images by dynamic MCMC methods. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 2134, pp. 201-215). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 2134). Springer Verlag. https://doi.org/10.1007/3-540-44745-8_14

Optimization of paintbrush rendering of images by dynamic MCMC methods. / Szirányi, T.; Tóth, Zoltán.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 2134 Springer Verlag, 2001. p. 201-215 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 2134).

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

Szirányi, T & Tóth, Z 2001, Optimization of paintbrush rendering of images by dynamic MCMC methods. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 2134, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 2134, Springer Verlag, pp. 201-215, 3rd International Workshop on Energy Minimization Methods in Computer Vision and Pattern Recognition, EMMCVPR 2001, Sophia Antipolis, France, 9/3/01. https://doi.org/10.1007/3-540-44745-8_14
Szirányi T, Tóth Z. Optimization of paintbrush rendering of images by dynamic MCMC methods. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 2134. Springer Verlag. 2001. p. 201-215. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/3-540-44745-8_14
Szirányi, T. ; Tóth, Zoltán. / Optimization of paintbrush rendering of images by dynamic MCMC methods. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 2134 Springer Verlag, 2001. pp. 201-215 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{48b767f6018b47b890ea47100a1db27a,
title = "Optimization of paintbrush rendering of images by dynamic MCMC methods",
abstract = "We have developed a new stochastic image rendering method for the compression, description and segmentation of images. This paintbrush-like image transformation is based on a random searching to insert brush-strokes into a generated image at decreasing scale of brush-sizes, without predefined models or interaction. We introduced a sequential multiscale image decomposition method, based on simulated rectangular-shaped paintbrush strokes. The resulting images look like good-quality paintings with well-defined contours, at an acceptable distortion compared to the original image. The image can be described with the parameters of the consecutive paintbrush strokes, resulting in a parameter-series that can be used for compression. The painting process can be applied for image representation, segmentation and contour detection. Our original method is based on stochastic exhaustive searching which takes a long time of convergence. In this paper we propose a modified algorithm of speed up of about 2x where the faster convergence is supported by a dynamic Metropolis Hastings rule.",
author = "T. Szir{\'a}nyi and Zolt{\'a}n T{\'o}th",
year = "2001",
doi = "10.1007/3-540-44745-8_14",
language = "English",
isbn = "3540425233",
volume = "2134",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "201--215",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - GEN

T1 - Optimization of paintbrush rendering of images by dynamic MCMC methods

AU - Szirányi, T.

AU - Tóth, Zoltán

PY - 2001

Y1 - 2001

N2 - We have developed a new stochastic image rendering method for the compression, description and segmentation of images. This paintbrush-like image transformation is based on a random searching to insert brush-strokes into a generated image at decreasing scale of brush-sizes, without predefined models or interaction. We introduced a sequential multiscale image decomposition method, based on simulated rectangular-shaped paintbrush strokes. The resulting images look like good-quality paintings with well-defined contours, at an acceptable distortion compared to the original image. The image can be described with the parameters of the consecutive paintbrush strokes, resulting in a parameter-series that can be used for compression. The painting process can be applied for image representation, segmentation and contour detection. Our original method is based on stochastic exhaustive searching which takes a long time of convergence. In this paper we propose a modified algorithm of speed up of about 2x where the faster convergence is supported by a dynamic Metropolis Hastings rule.

AB - We have developed a new stochastic image rendering method for the compression, description and segmentation of images. This paintbrush-like image transformation is based on a random searching to insert brush-strokes into a generated image at decreasing scale of brush-sizes, without predefined models or interaction. We introduced a sequential multiscale image decomposition method, based on simulated rectangular-shaped paintbrush strokes. The resulting images look like good-quality paintings with well-defined contours, at an acceptable distortion compared to the original image. The image can be described with the parameters of the consecutive paintbrush strokes, resulting in a parameter-series that can be used for compression. The painting process can be applied for image representation, segmentation and contour detection. Our original method is based on stochastic exhaustive searching which takes a long time of convergence. In this paper we propose a modified algorithm of speed up of about 2x where the faster convergence is supported by a dynamic Metropolis Hastings rule.

UR - http://www.scopus.com/inward/record.url?scp=23044531801&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=23044531801&partnerID=8YFLogxK

U2 - 10.1007/3-540-44745-8_14

DO - 10.1007/3-540-44745-8_14

M3 - Conference contribution

AN - SCOPUS:23044531801

SN - 3540425233

SN - 9783540425236

VL - 2134

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 201

EP - 215

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

PB - Springer Verlag

ER -