Abstract art by shape classification

Yi-Zhe Song, David Pickup, Chuan Li, Paul Rosin, Peter Hall

Research output: Contribution to journalArticle

  • 8 Citations

Abstract

This paper shows that classifying shapes is a tool useful in Non-Photorealistic rendering from photographs. Our classifier inputs regions from an image segmentation hierarchy and outputs the “best” fitting simple shape such as a circle, square or triangle. Other approaches to NPR have recognised the benefits of segmentation, but none have classified the shape of segments. By doing so, we can create artwork of a more abstract nature, emulating the style of modern artists such as Matisse and other artists who favoured shape simplification in their artwork. The classifier chooses the shape that “best” represents the region. Since the classifier is trained by a user, the ‘best shape’ has a subjective quality that can over-ride measurements such as minimum error and more importantly captures user preferences. Once trained, the system is fully automatic, although simple user interaction is also possible to allow for differences in individual tastes. A gallery of results shows how this classifier contributes to NPR from images by producing abstract artwork.
LanguageEnglish
Pages1252-1263
Number of pages12
JournalIEEE Transactions on Visualization and Computer Graphics
Volume19
Issue number8
Early online date14 Feb 2013
DOIs
StatusPublished - Aug 2013

Fingerprint

Classifiers
Image segmentation

ASJC Scopus subject areas

  • Computer Graphics and Computer-Aided Design

Cite this

Abstract art by shape classification. / Song, Yi-Zhe; Pickup, David; Li, Chuan; Rosin, Paul; Hall, Peter.

In: IEEE Transactions on Visualization and Computer Graphics, Vol. 19, No. 8, 08.2013, p. 1252-1263.

Research output: Contribution to journalArticle

Song, Y-Z, Pickup, D, Li, C, Rosin, P & Hall, P 2013, 'Abstract art by shape classification' IEEE Transactions on Visualization and Computer Graphics, vol. 19, no. 8, pp. 1252-1263. DOI: 10.1109/TVCG.2013.13
Song Y-Z, Pickup D, Li C, Rosin P, Hall P. Abstract art by shape classification. IEEE Transactions on Visualization and Computer Graphics. 2013 Aug;19(8):1252-1263. Available from, DOI: 10.1109/TVCG.2013.13
Song, Yi-Zhe ; Pickup, David ; Li, Chuan ; Rosin, Paul ; Hall, Peter. / Abstract art by shape classification. In: IEEE Transactions on Visualization and Computer Graphics. 2013 ; Vol. 19, No. 8. pp. 1252-1263
@article{c91374cc982e4688912f00e21167e324,
title = "Abstract art by shape classification",
abstract = "This paper shows that classifying shapes is a tool useful in Non-Photorealistic rendering from photographs. Our classifier inputs regions from an image segmentation hierarchy and outputs the “best” fitting simple shape such as a circle, square or triangle. Other approaches to NPR have recognised the benefits of segmentation, but none have classified the shape of segments. By doing so, we can create artwork of a more abstract nature, emulating the style of modern artists such as Matisse and other artists who favoured shape simplification in their artwork. The classifier chooses the shape that “best” represents the region. Since the classifier is trained by a user, the ‘best shape’ has a subjective quality that can over-ride measurements such as minimum error and more importantly captures user preferences. Once trained, the system is fully automatic, although simple user interaction is also possible to allow for differences in individual tastes. A gallery of results shows how this classifier contributes to NPR from images by producing abstract artwork.",
author = "Yi-Zhe Song and David Pickup and Chuan Li and Paul Rosin and Peter Hall",
year = "2013",
month = "8",
doi = "10.1109/TVCG.2013.13",
language = "English",
volume = "19",
pages = "1252--1263",
journal = "IEEE Transactions on Visualization and Computer Graphics",
issn = "1077-2626",
publisher = "IEEE",
number = "8",

}

TY - JOUR

T1 - Abstract art by shape classification

AU - Song,Yi-Zhe

AU - Pickup,David

AU - Li,Chuan

AU - Rosin,Paul

AU - Hall,Peter

PY - 2013/8

Y1 - 2013/8

N2 - This paper shows that classifying shapes is a tool useful in Non-Photorealistic rendering from photographs. Our classifier inputs regions from an image segmentation hierarchy and outputs the “best” fitting simple shape such as a circle, square or triangle. Other approaches to NPR have recognised the benefits of segmentation, but none have classified the shape of segments. By doing so, we can create artwork of a more abstract nature, emulating the style of modern artists such as Matisse and other artists who favoured shape simplification in their artwork. The classifier chooses the shape that “best” represents the region. Since the classifier is trained by a user, the ‘best shape’ has a subjective quality that can over-ride measurements such as minimum error and more importantly captures user preferences. Once trained, the system is fully automatic, although simple user interaction is also possible to allow for differences in individual tastes. A gallery of results shows how this classifier contributes to NPR from images by producing abstract artwork.

AB - This paper shows that classifying shapes is a tool useful in Non-Photorealistic rendering from photographs. Our classifier inputs regions from an image segmentation hierarchy and outputs the “best” fitting simple shape such as a circle, square or triangle. Other approaches to NPR have recognised the benefits of segmentation, but none have classified the shape of segments. By doing so, we can create artwork of a more abstract nature, emulating the style of modern artists such as Matisse and other artists who favoured shape simplification in their artwork. The classifier chooses the shape that “best” represents the region. Since the classifier is trained by a user, the ‘best shape’ has a subjective quality that can over-ride measurements such as minimum error and more importantly captures user preferences. Once trained, the system is fully automatic, although simple user interaction is also possible to allow for differences in individual tastes. A gallery of results shows how this classifier contributes to NPR from images by producing abstract artwork.

UR - http://www.scopus.com/inward/record.url?scp=84879223353&partnerID=8YFLogxK

UR - http://dx.doi.org/10.1109/TVCG.2013.13

U2 - 10.1109/TVCG.2013.13

DO - 10.1109/TVCG.2013.13

M3 - Article

VL - 19

SP - 1252

EP - 1263

JO - IEEE Transactions on Visualization and Computer Graphics

T2 - IEEE Transactions on Visualization and Computer Graphics

JF - IEEE Transactions on Visualization and Computer Graphics

SN - 1077-2626

IS - 8

ER -