Brushstroke based sparse hybrid convolutional neural networks for author classification of Chinese ink-wash paintings

Meijun Sun, Dong Zhang, Jinchang Ren, Zheng Wang, Jesse S. Jin

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

10 Citations (Scopus)

Abstract

A novel stroke based sparse hybrid convolutional neural networks (CNNs) method is proposed for author classification of Chinese ink-wash paintings (IWPs). As Chinese IWPs usually have many authors in several art styles, this differs from real images or western paintings and has led to a big challenge. In our work, we classify Chinese IWPs of different artists by analyzing a set of automatically extracted brushstrokes. A sparse hybrid CNNs in a deep-learning framework is then proposed to extract brushstroke features to replace the commonly used handcrafted ones such as edge, color, intensity and texture. Using 120 IWPs from six famous artists, promising results have been shown in successfully classifying authors in comparison to two other state-of-the-art approaches.

Original languageEnglish
Title of host publicationProceedings - International Conference on Image Processing, ICIP
PublisherIEEE
Pages626-630
Number of pages5
ISBN (Print)9781479983391
DOIs
Publication statusPublished - 9 Dec 2015
EventIEEE International Conference on Image Processing, ICIP 2015 - Quebec City, Canada
Duration: 27 Sep 201530 Sep 2015

Conference

ConferenceIEEE International Conference on Image Processing, ICIP 2015
CountryCanada
CityQuebec City
Period27/09/1530/09/15

Fingerprint

Painting
Ink
Neural networks
Textures
Color

Keywords

  • brushstroke analysis
  • Chinese ink-wash painting
  • convolutional neural networks
  • image classification
  • sparse coding

Cite this

Sun, M., Zhang, D., Ren, J., Wang, Z., & Jin, J. S. (2015). Brushstroke based sparse hybrid convolutional neural networks for author classification of Chinese ink-wash paintings. In Proceedings - International Conference on Image Processing, ICIP (pp. 626-630). IEEE. https://doi.org/10.1109/ICIP.2015.7350874
Sun, Meijun ; Zhang, Dong ; Ren, Jinchang ; Wang, Zheng ; Jin, Jesse S. / Brushstroke based sparse hybrid convolutional neural networks for author classification of Chinese ink-wash paintings. Proceedings - International Conference on Image Processing, ICIP. IEEE, 2015. pp. 626-630
@inproceedings{9ed84492fe484ab1abd1741ff320f0f8,
title = "Brushstroke based sparse hybrid convolutional neural networks for author classification of Chinese ink-wash paintings",
abstract = "A novel stroke based sparse hybrid convolutional neural networks (CNNs) method is proposed for author classification of Chinese ink-wash paintings (IWPs). As Chinese IWPs usually have many authors in several art styles, this differs from real images or western paintings and has led to a big challenge. In our work, we classify Chinese IWPs of different artists by analyzing a set of automatically extracted brushstrokes. A sparse hybrid CNNs in a deep-learning framework is then proposed to extract brushstroke features to replace the commonly used handcrafted ones such as edge, color, intensity and texture. Using 120 IWPs from six famous artists, promising results have been shown in successfully classifying authors in comparison to two other state-of-the-art approaches.",
keywords = "brushstroke analysis, Chinese ink-wash painting, convolutional neural networks, image classification, sparse coding",
author = "Meijun Sun and Dong Zhang and Jinchang Ren and Zheng Wang and Jin, {Jesse S.}",
year = "2015",
month = "12",
day = "9",
doi = "10.1109/ICIP.2015.7350874",
language = "English",
isbn = "9781479983391",
pages = "626--630",
booktitle = "Proceedings - International Conference on Image Processing, ICIP",
publisher = "IEEE",

}

Sun, M, Zhang, D, Ren, J, Wang, Z & Jin, JS 2015, Brushstroke based sparse hybrid convolutional neural networks for author classification of Chinese ink-wash paintings. in Proceedings - International Conference on Image Processing, ICIP. IEEE, pp. 626-630, IEEE International Conference on Image Processing, ICIP 2015, Quebec City, Canada, 27/09/15. https://doi.org/10.1109/ICIP.2015.7350874

Brushstroke based sparse hybrid convolutional neural networks for author classification of Chinese ink-wash paintings. / Sun, Meijun; Zhang, Dong; Ren, Jinchang; Wang, Zheng; Jin, Jesse S.

Proceedings - International Conference on Image Processing, ICIP. IEEE, 2015. p. 626-630.

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

TY - GEN

T1 - Brushstroke based sparse hybrid convolutional neural networks for author classification of Chinese ink-wash paintings

AU - Sun, Meijun

AU - Zhang, Dong

AU - Ren, Jinchang

AU - Wang, Zheng

AU - Jin, Jesse S.

PY - 2015/12/9

Y1 - 2015/12/9

N2 - A novel stroke based sparse hybrid convolutional neural networks (CNNs) method is proposed for author classification of Chinese ink-wash paintings (IWPs). As Chinese IWPs usually have many authors in several art styles, this differs from real images or western paintings and has led to a big challenge. In our work, we classify Chinese IWPs of different artists by analyzing a set of automatically extracted brushstrokes. A sparse hybrid CNNs in a deep-learning framework is then proposed to extract brushstroke features to replace the commonly used handcrafted ones such as edge, color, intensity and texture. Using 120 IWPs from six famous artists, promising results have been shown in successfully classifying authors in comparison to two other state-of-the-art approaches.

AB - A novel stroke based sparse hybrid convolutional neural networks (CNNs) method is proposed for author classification of Chinese ink-wash paintings (IWPs). As Chinese IWPs usually have many authors in several art styles, this differs from real images or western paintings and has led to a big challenge. In our work, we classify Chinese IWPs of different artists by analyzing a set of automatically extracted brushstrokes. A sparse hybrid CNNs in a deep-learning framework is then proposed to extract brushstroke features to replace the commonly used handcrafted ones such as edge, color, intensity and texture. Using 120 IWPs from six famous artists, promising results have been shown in successfully classifying authors in comparison to two other state-of-the-art approaches.

KW - brushstroke analysis

KW - Chinese ink-wash painting

KW - convolutional neural networks

KW - image classification

KW - sparse coding

UR - http://www.scopus.com/inward/record.url?scp=84956616844&partnerID=8YFLogxK

U2 - 10.1109/ICIP.2015.7350874

DO - 10.1109/ICIP.2015.7350874

M3 - Conference contribution book

SN - 9781479983391

SP - 626

EP - 630

BT - Proceedings - International Conference on Image Processing, ICIP

PB - IEEE

ER -

Sun M, Zhang D, Ren J, Wang Z, Jin JS. Brushstroke based sparse hybrid convolutional neural networks for author classification of Chinese ink-wash paintings. In Proceedings - International Conference on Image Processing, ICIP. IEEE. 2015. p. 626-630 https://doi.org/10.1109/ICIP.2015.7350874