DeepPainter: Painter classification using deep convolutional autoencoders

Omid E. David; Nathan S. Netanyahu

doi:10.1007/978-3-319-44781-0_3

DeepPainter: Painter classification using deep convolutional autoencoders

Omid E. David, Nathan S. Netanyahu

University of Maryland, College Park

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

38 Scopus citations

Abstract

In this paper we describe the problem of painter classification, and propose a novel approach based on deep convolutional autoencoder neural networks. While previous approaches relied on image processing and manual feature extraction from paintings, our approach operates on the raw pixel level, without any preprocessing or manual feature extraction. We first train a deep convolutional autoencoder on a dataset of paintings, and subsequently use it to initialize a supervised convolutional neural network for the classification phase. The proposed approach substantially outperforms previous methods, improving the previous state-of-the-art for the 3-painter classification problem from 90.44% accuracy (previous state-of-the-art) to 96.52% accuracy, i.e., a 63% reduction in error rate.

Original language	English
Title of host publication	Artificial Neural Networks and Machine Learning - 25th International Conference on Artificial Neural Networks, ICANN 2016, Proceedings
Editors	Alessandro E.P. Villa, Paolo Masulli, Antonio Javier Pons Rivero
Publisher	Springer Verlag
Pages	20-28
Number of pages	9
ISBN (Print)	9783319447803
DOIs	https://doi.org/10.1007/978-3-319-44781-0_3
State	Published - 2016
Event	25th International Conference on Artificial Neural Networks and Machine Learning, ICANN 2016 - Barcelona, Spain Duration: 6 Sep 2016 → 9 Sep 2016

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	9887 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	25th International Conference on Artificial Neural Networks and Machine Learning, ICANN 2016
Country/Territory	Spain
City	Barcelona
Period	6/09/16 → 9/09/16

Bibliographical note

Publisher Copyright:
© Springer International Publishing Switzerland 2016.

Access to Document

10.1007/978-3-319-44781-0_3

Cite this

David, O. E., & Netanyahu, N. S. (2016). DeepPainter: Painter classification using deep convolutional autoencoders. In A. E. P. Villa, P. Masulli, & A. J. P. Rivero (Eds.), Artificial Neural Networks and Machine Learning - 25th International Conference on Artificial Neural Networks, ICANN 2016, Proceedings (pp. 20-28). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9887 LNCS). Springer Verlag. https://doi.org/10.1007/978-3-319-44781-0_3

David, Omid E. ; Netanyahu, Nathan S. / DeepPainter : Painter classification using deep convolutional autoencoders. Artificial Neural Networks and Machine Learning - 25th International Conference on Artificial Neural Networks, ICANN 2016, Proceedings. editor / Alessandro E.P. Villa ; Paolo Masulli ; Antonio Javier Pons Rivero. Springer Verlag, 2016. pp. 20-28 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{9d4073af6e4b45da9c39c6e0f42b476f,

title = "DeepPainter: Painter classification using deep convolutional autoencoders",

abstract = "In this paper we describe the problem of painter classification, and propose a novel approach based on deep convolutional autoencoder neural networks. While previous approaches relied on image processing and manual feature extraction from paintings, our approach operates on the raw pixel level, without any preprocessing or manual feature extraction. We first train a deep convolutional autoencoder on a dataset of paintings, and subsequently use it to initialize a supervised convolutional neural network for the classification phase. The proposed approach substantially outperforms previous methods, improving the previous state-of-the-art for the 3-painter classification problem from 90.44% accuracy (previous state-of-the-art) to 96.52% accuracy, i.e., a 63% reduction in error rate.",

author = "David, {Omid E.} and Netanyahu, {Nathan S.}",

note = "Publisher Copyright: {\textcopyright} Springer International Publishing Switzerland 2016.; 25th International Conference on Artificial Neural Networks and Machine Learning, ICANN 2016 ; Conference date: 06-09-2016 Through 09-09-2016",

year = "2016",

doi = "10.1007/978-3-319-44781-0_3",

language = "אנגלית",

isbn = "9783319447803",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Verlag",

pages = "20--28",

editor = "Villa, {Alessandro E.P.} and Paolo Masulli and Rivero, {Antonio Javier Pons}",

booktitle = "Artificial Neural Networks and Machine Learning - 25th International Conference on Artificial Neural Networks, ICANN 2016, Proceedings",

address = "גרמניה",

}

David, OE & Netanyahu, NS 2016, DeepPainter: Painter classification using deep convolutional autoencoders. in AEP Villa, P Masulli & AJP Rivero (eds), Artificial Neural Networks and Machine Learning - 25th International Conference on Artificial Neural Networks, ICANN 2016, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9887 LNCS, Springer Verlag, pp. 20-28, 25th International Conference on Artificial Neural Networks and Machine Learning, ICANN 2016, Barcelona, Spain, 6/09/16. https://doi.org/10.1007/978-3-319-44781-0_3

DeepPainter: Painter classification using deep convolutional autoencoders. / David, Omid E.; Netanyahu, Nathan S.
Artificial Neural Networks and Machine Learning - 25th International Conference on Artificial Neural Networks, ICANN 2016, Proceedings. ed. / Alessandro E.P. Villa; Paolo Masulli; Antonio Javier Pons Rivero. Springer Verlag, 2016. p. 20-28 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9887 LNCS).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - DeepPainter

T2 - 25th International Conference on Artificial Neural Networks and Machine Learning, ICANN 2016

AU - David, Omid E.

AU - Netanyahu, Nathan S.

N1 - Publisher Copyright: © Springer International Publishing Switzerland 2016.

PY - 2016

Y1 - 2016

N2 - In this paper we describe the problem of painter classification, and propose a novel approach based on deep convolutional autoencoder neural networks. While previous approaches relied on image processing and manual feature extraction from paintings, our approach operates on the raw pixel level, without any preprocessing or manual feature extraction. We first train a deep convolutional autoencoder on a dataset of paintings, and subsequently use it to initialize a supervised convolutional neural network for the classification phase. The proposed approach substantially outperforms previous methods, improving the previous state-of-the-art for the 3-painter classification problem from 90.44% accuracy (previous state-of-the-art) to 96.52% accuracy, i.e., a 63% reduction in error rate.

AB - In this paper we describe the problem of painter classification, and propose a novel approach based on deep convolutional autoencoder neural networks. While previous approaches relied on image processing and manual feature extraction from paintings, our approach operates on the raw pixel level, without any preprocessing or manual feature extraction. We first train a deep convolutional autoencoder on a dataset of paintings, and subsequently use it to initialize a supervised convolutional neural network for the classification phase. The proposed approach substantially outperforms previous methods, improving the previous state-of-the-art for the 3-painter classification problem from 90.44% accuracy (previous state-of-the-art) to 96.52% accuracy, i.e., a 63% reduction in error rate.

UR - http://www.scopus.com/inward/record.url?scp=84988409808&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-44781-0_3

DO - 10.1007/978-3-319-44781-0_3

M3 - ???researchoutput.researchoutputtypes.contributiontobookanthology.conference???

AN - SCOPUS:84988409808

SN - 9783319447803

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

SP - 20

EP - 28

BT - Artificial Neural Networks and Machine Learning - 25th International Conference on Artificial Neural Networks, ICANN 2016, Proceedings

A2 - Villa, Alessandro E.P.

A2 - Masulli, Paolo

A2 - Rivero, Antonio Javier Pons

PB - Springer Verlag

Y2 - 6 September 2016 through 9 September 2016

ER -

David OE, Netanyahu NS. DeepPainter: Painter classification using deep convolutional autoencoders. In Villa AEP, Masulli P, Rivero AJP, editors, Artificial Neural Networks and Machine Learning - 25th International Conference on Artificial Neural Networks, ICANN 2016, Proceedings. Springer Verlag. 2016. p. 20-28. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-319-44781-0_3

DeepPainter: Painter classification using deep convolutional autoencoders

Abstract

Publication series

Conference

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this