RealPigment

Paint compositing by example

Jingwan Lu, Stephen DiVerdi, Willa A. Chen, Connelly Barnes, Adam Finkelstein

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

8 Citations (Scopus)

Abstract

The color of composited pigments in digital painting is generally computed one of two ways: either alpha blending in RGB, or the Kubelka-Munk equation (KM). The former fails to reproduce paint like appearances, while the latter is difficult to use. We present a data-driven pigment model that reproduces arbitrary compositing behavior by interpolating sparse samples in a high dimensional space. The input is an of a color chart, which provides the composition samples. We propose two different prediction algorithms, one doing simple interpolation using radial basis functions (RBF), and another that trains a parametric model based on the KM equation to compute novel values. We show that RBF is able to reproduce arbitrary compositing behaviors, even non-paint-like such as additive blending, while KM compositing is more robust to acquisition noise and can generalize results over a broader range of values.

Original languageEnglish (US)
Title of host publicationNPAR 2014 - Proceedings of the Workshop on Non-Photorealistic Animation and Rendering - Joint Symposium on Computational Aesthetics and Sketch-Based Interfaces and Modeling and Non-Photorealistic Animation and Rendering, Expressive 2014
EditorsStephen N. Spencer
PublisherAssociation for Computing Machinery
Pages21-30
Number of pages10
ISBN (Electronic)9781450330206
DOIs
StatePublished - Aug 8 2014
EventWorkshop on Non-Photorealistic Animation and Rendering, NPAR 2014 - 4th Joint Symposium on Computational Aesthetics and Sketch-Based Interfaces and Modeling and Non-Photorealistic Animation and Rendering, Expressive 2014 - Vancouver, Canada
Duration: Aug 8 2014Aug 10 2014

Publication series

NameNPAR Symposium on Non-Photorealistic Animation and Rendering
Volume2014-January

Other

OtherWorkshop on Non-Photorealistic Animation and Rendering, NPAR 2014 - 4th Joint Symposium on Computational Aesthetics and Sketch-Based Interfaces and Modeling and Non-Photorealistic Animation and Rendering, Expressive 2014
CountryCanada
CityVancouver
Period8/8/148/10/14

Fingerprint

Pigments
Paint
Color
Painting
Radial Functions
Basis Functions
Interpolation
Arbitrary
Parametric Model
Chemical analysis
Chart
Data-driven
High-dimensional
Interpolate
Model-based
Generalise
Prediction
Range of data
Model

All Science Journal Classification (ASJC) codes

  • Applied Mathematics
  • Computer Science Applications
  • Computer Graphics and Computer-Aided Design
  • Modeling and Simulation

Cite this

Lu, J., DiVerdi, S., Chen, W. A., Barnes, C., & Finkelstein, A. (2014). RealPigment: Paint compositing by example. In S. N. Spencer (Ed.), NPAR 2014 - Proceedings of the Workshop on Non-Photorealistic Animation and Rendering - Joint Symposium on Computational Aesthetics and Sketch-Based Interfaces and Modeling and Non-Photorealistic Animation and Rendering, Expressive 2014 (pp. 21-30). (NPAR Symposium on Non-Photorealistic Animation and Rendering; Vol. 2014-January). Association for Computing Machinery. https://doi.org/10.1145/2630397.2630401
Lu, Jingwan ; DiVerdi, Stephen ; Chen, Willa A. ; Barnes, Connelly ; Finkelstein, Adam. / RealPigment : Paint compositing by example. NPAR 2014 - Proceedings of the Workshop on Non-Photorealistic Animation and Rendering - Joint Symposium on Computational Aesthetics and Sketch-Based Interfaces and Modeling and Non-Photorealistic Animation and Rendering, Expressive 2014. editor / Stephen N. Spencer. Association for Computing Machinery, 2014. pp. 21-30 (NPAR Symposium on Non-Photorealistic Animation and Rendering).
@inproceedings{7e0858a6faef42848035b23c33e82062,
title = "RealPigment: Paint compositing by example",
abstract = "The color of composited pigments in digital painting is generally computed one of two ways: either alpha blending in RGB, or the Kubelka-Munk equation (KM). The former fails to reproduce paint like appearances, while the latter is difficult to use. We present a data-driven pigment model that reproduces arbitrary compositing behavior by interpolating sparse samples in a high dimensional space. The input is an of a color chart, which provides the composition samples. We propose two different prediction algorithms, one doing simple interpolation using radial basis functions (RBF), and another that trains a parametric model based on the KM equation to compute novel values. We show that RBF is able to reproduce arbitrary compositing behaviors, even non-paint-like such as additive blending, while KM compositing is more robust to acquisition noise and can generalize results over a broader range of values.",
author = "Jingwan Lu and Stephen DiVerdi and Chen, {Willa A.} and Connelly Barnes and Adam Finkelstein",
year = "2014",
month = "8",
day = "8",
doi = "https://doi.org/10.1145/2630397.2630401",
language = "English (US)",
series = "NPAR Symposium on Non-Photorealistic Animation and Rendering",
publisher = "Association for Computing Machinery",
pages = "21--30",
editor = "Spencer, {Stephen N.}",
booktitle = "NPAR 2014 - Proceedings of the Workshop on Non-Photorealistic Animation and Rendering - Joint Symposium on Computational Aesthetics and Sketch-Based Interfaces and Modeling and Non-Photorealistic Animation and Rendering, Expressive 2014",

}

Lu, J, DiVerdi, S, Chen, WA, Barnes, C & Finkelstein, A 2014, RealPigment: Paint compositing by example. in SN Spencer (ed.), NPAR 2014 - Proceedings of the Workshop on Non-Photorealistic Animation and Rendering - Joint Symposium on Computational Aesthetics and Sketch-Based Interfaces and Modeling and Non-Photorealistic Animation and Rendering, Expressive 2014. NPAR Symposium on Non-Photorealistic Animation and Rendering, vol. 2014-January, Association for Computing Machinery, pp. 21-30, Workshop on Non-Photorealistic Animation and Rendering, NPAR 2014 - 4th Joint Symposium on Computational Aesthetics and Sketch-Based Interfaces and Modeling and Non-Photorealistic Animation and Rendering, Expressive 2014, Vancouver, Canada, 8/8/14. https://doi.org/10.1145/2630397.2630401

RealPigment : Paint compositing by example. / Lu, Jingwan; DiVerdi, Stephen; Chen, Willa A.; Barnes, Connelly; Finkelstein, Adam.

NPAR 2014 - Proceedings of the Workshop on Non-Photorealistic Animation and Rendering - Joint Symposium on Computational Aesthetics and Sketch-Based Interfaces and Modeling and Non-Photorealistic Animation and Rendering, Expressive 2014. ed. / Stephen N. Spencer. Association for Computing Machinery, 2014. p. 21-30 (NPAR Symposium on Non-Photorealistic Animation and Rendering; Vol. 2014-January).

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

TY - GEN

T1 - RealPigment

T2 - Paint compositing by example

AU - Lu, Jingwan

AU - DiVerdi, Stephen

AU - Chen, Willa A.

AU - Barnes, Connelly

AU - Finkelstein, Adam

PY - 2014/8/8

Y1 - 2014/8/8

N2 - The color of composited pigments in digital painting is generally computed one of two ways: either alpha blending in RGB, or the Kubelka-Munk equation (KM). The former fails to reproduce paint like appearances, while the latter is difficult to use. We present a data-driven pigment model that reproduces arbitrary compositing behavior by interpolating sparse samples in a high dimensional space. The input is an of a color chart, which provides the composition samples. We propose two different prediction algorithms, one doing simple interpolation using radial basis functions (RBF), and another that trains a parametric model based on the KM equation to compute novel values. We show that RBF is able to reproduce arbitrary compositing behaviors, even non-paint-like such as additive blending, while KM compositing is more robust to acquisition noise and can generalize results over a broader range of values.

AB - The color of composited pigments in digital painting is generally computed one of two ways: either alpha blending in RGB, or the Kubelka-Munk equation (KM). The former fails to reproduce paint like appearances, while the latter is difficult to use. We present a data-driven pigment model that reproduces arbitrary compositing behavior by interpolating sparse samples in a high dimensional space. The input is an of a color chart, which provides the composition samples. We propose two different prediction algorithms, one doing simple interpolation using radial basis functions (RBF), and another that trains a parametric model based on the KM equation to compute novel values. We show that RBF is able to reproduce arbitrary compositing behaviors, even non-paint-like such as additive blending, while KM compositing is more robust to acquisition noise and can generalize results over a broader range of values.

UR - http://www.scopus.com/inward/record.url?scp=84930947186&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84930947186&partnerID=8YFLogxK

U2 - https://doi.org/10.1145/2630397.2630401

DO - https://doi.org/10.1145/2630397.2630401

M3 - Conference contribution

T3 - NPAR Symposium on Non-Photorealistic Animation and Rendering

SP - 21

EP - 30

BT - NPAR 2014 - Proceedings of the Workshop on Non-Photorealistic Animation and Rendering - Joint Symposium on Computational Aesthetics and Sketch-Based Interfaces and Modeling and Non-Photorealistic Animation and Rendering, Expressive 2014

A2 - Spencer, Stephen N.

PB - Association for Computing Machinery

ER -

Lu J, DiVerdi S, Chen WA, Barnes C, Finkelstein A. RealPigment: Paint compositing by example. In Spencer SN, editor, NPAR 2014 - Proceedings of the Workshop on Non-Photorealistic Animation and Rendering - Joint Symposium on Computational Aesthetics and Sketch-Based Interfaces and Modeling and Non-Photorealistic Animation and Rendering, Expressive 2014. Association for Computing Machinery. 2014. p. 21-30. (NPAR Symposium on Non-Photorealistic Animation and Rendering). https://doi.org/10.1145/2630397.2630401