Presentation 2015-07-19
The classification of paintings by color statistics
Masatoshi Kitaguchi, Masahiro Wakabayashi, Hiromichi Sato, Tomoyuki Naito,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Several studies reported that statistics of visual properties of paintings might be distinctive among painters, and that it was possible to classify painters by those statistics. However, it is still unclear which visual properties are important for the classification. In this study, using general support vector machine (SVM), we tried to classify painters of our image database (1050 landscape paintings of several painters). For learning, we used nine color-statistics (HSV color statistics, XYZ color statistics, and so on). Cross-validation test revealed that the SVM correctly classified the painters with ~65-90% accuracy. We also found that mean brightness and entropy of brightness played roles for the image classification. Our results suggested that combinations of a few color statistics might be effective for painting classification.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Painting / Color Statistic / Color Space / Emotion / Sensibility / Neuroesthetics
Paper # HIP2015-60
Date of Issue 2015-07-11 (HIP)

Conference Information
Committee HIP
Conference Date 2015/7/18(2days)
Place (in Japanese) (See Japanese page)
Place (in English) Kyushu Sangyo University
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair Hideyuki Ando(Osaka Univ.)
Vice Chair Masahiro Ishii(Sapporo City Univ.) / Miyuki Kamachi(Kogakuin Univ.)
Secretary Masahiro Ishii(NICT) / Miyuki Kamachi(KDDI R&D Labs.)
Assistant Sinobu Kuroki(NTT) / Mutsumi Suganuma(Waseda Univ.)

Paper Information
Registration To Technical Committee on Human Information Processing
Language ENG-JTITLE
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) The classification of paintings by color statistics
Sub Title (in English)
Keyword(1) Painting
Keyword(2) Color Statistic
Keyword(3) Color Space
Keyword(4) Emotion
Keyword(5) Sensibility
Keyword(6) Neuroesthetics
1st Author's Name Masatoshi Kitaguchi
1st Author's Affiliation Osaka University(Osaka Univ.)
2nd Author's Name Masahiro Wakabayashi
2nd Author's Affiliation Osaka University(Osaka Univ.)
3rd Author's Name Hiromichi Sato
3rd Author's Affiliation Osaka University(Osaka Univ.)
4th Author's Name Tomoyuki Naito
4th Author's Affiliation Osaka University(Osaka Univ.)
Date 2015-07-19
Paper # HIP2015-60
Volume (vol) vol.115
Number (no) HIP-149
Page pp.pp.93-98(HIP),
#Pages 6
Date of Issue 2015-07-11 (HIP)