Advanced search
1 file | 3.72 MB Add to list
Author
Organization
Abstract
Autoencoders have been widely used in applications with limited annotations to extract features in an unsupervised manner, pre-processing the data to be used in machine learning models. This is especially helpful in image processing for art investigation where annotated data is scarce and difficult to collect. We introduce a structural similarity index based loss function to train the autoencoder for image data. By extending the recently developed partial convolution to partial deconvolution, we construct a fully partial convolutional autoencoder (FP-CAE) and adapt it to multimodal data, typically utilized in art investigation. Experimental results on images of the Ghent Altarpieceshow that our method significantly suppresses edge artifacts and improves the overall reconstruction performance. The proposed FP-CAE can be used for data preprocessing in craquelure detection and other art investigation tasks in future studies.
Keywords
Autoencoder, Partial convolution, Multimodal data

Downloads

  • paper123.pdf
    • full text
    • |
    • open access
    • |
    • PDF
    • |
    • 3.72 MB

Citation

Please use this url to cite or link to this publication:

MLA
Xie, Xianghui, et al. “Partial Convolution Based Multimodal Autoencoder for Art Investigation.” Proceedings of the 31st Benelux Conference on Artificial Intelligence (BNAIC 2019) and the 28th Belgian Dutch Conference on Machine Learning (Benelearn 2019) , edited by Katrien Beuls et al., vol. 2491, 2019.
APA
Xie, X., Meeus, L., & Pizurica, A. (2019). Partial convolution based multimodal autoencoder for art investigation. In K. Beuls, B. Bogaerts, G. Bontempi, P. Geurts, N. Harley, B. Lebichot, … P. Van Eecke (Eds.), Proceedings of the 31st Benelux Conference on Artificial Intelligence (BNAIC 2019) and the 28th Belgian Dutch Conference on Machine Learning (Benelearn 2019) (Vol. 2491).
Chicago author-date
Xie, Xianghui, Laurens Meeus, and Aleksandra Pizurica. 2019. “Partial Convolution Based Multimodal Autoencoder for Art Investigation.” In Proceedings of the 31st Benelux Conference on Artificial Intelligence (BNAIC 2019) and the 28th Belgian Dutch Conference on Machine Learning (Benelearn 2019) , edited by Katrien Beuls, Bart Bogaerts, Gianluca Bontempi, Pierre Geurts, Nick Harley, Bertrand Lebichot, Tom Lenaerts, Gilles Louppe, and Paul Van Eecke. Vol. 2491.
Chicago author-date (all authors)
Xie, Xianghui, Laurens Meeus, and Aleksandra Pizurica. 2019. “Partial Convolution Based Multimodal Autoencoder for Art Investigation.” In Proceedings of the 31st Benelux Conference on Artificial Intelligence (BNAIC 2019) and the 28th Belgian Dutch Conference on Machine Learning (Benelearn 2019) , ed by. Katrien Beuls, Bart Bogaerts, Gianluca Bontempi, Pierre Geurts, Nick Harley, Bertrand Lebichot, Tom Lenaerts, Gilles Louppe, and Paul Van Eecke. Vol. 2491.
Vancouver
1.
Xie X, Meeus L, Pizurica A. Partial convolution based multimodal autoencoder for art investigation. In: Beuls K, Bogaerts B, Bontempi G, Geurts P, Harley N, Lebichot B, et al., editors. Proceedings of the 31st Benelux Conference on Artificial Intelligence (BNAIC 2019) and the 28th Belgian Dutch Conference on Machine Learning (Benelearn 2019) . 2019.
IEEE
[1]
X. Xie, L. Meeus, and A. Pizurica, “Partial convolution based multimodal autoencoder for art investigation,” in Proceedings of the 31st Benelux Conference on Artificial Intelligence (BNAIC 2019) and the 28th Belgian Dutch Conference on Machine Learning (Benelearn 2019) , 2019, vol. 2491.
@inproceedings{8636034,
  abstract     = {Autoencoders have been widely used in applications with limited annotations to extract features in an unsupervised manner, pre-processing the data to be used in machine learning models. This is especially helpful in image processing for art investigation where annotated data is scarce and difficult to collect. We introduce a structural similarity index based loss function to train the autoencoder for image data.  By extending the recently developed partial convolution to partial deconvolution, we construct a fully partial convolutional autoencoder (FP-CAE) and adapt it to multimodal data, typically utilized in art investigation. Experimental results on images of the Ghent Altarpieceshow that our method significantly suppresses edge artifacts and improves the overall reconstruction performance. The proposed FP-CAE can be used for data preprocessing in craquelure detection and other art investigation tasks in future studies.},
  articleno    = {123},
  author       = {Xie, Xianghui and Meeus, Laurens and Pizurica, Aleksandra},
  booktitle    = {Proceedings of the 31st Benelux Conference on Artificial Intelligence (BNAIC 2019) and the 28th Belgian Dutch Conference on Machine Learning (Benelearn 2019) },
  editor       = {Beuls, Katrien and Bogaerts, Bart and Bontempi, Gianluca and Geurts, Pierre and Harley, Nick and Lebichot, Bertrand and Lenaerts, Tom and Louppe, Gilles and Van Eecke, Paul},
  issn         = {1613-0073},
  keywords     = {Autoencoder,Partial convolution,Multimodal data},
  language     = {eng},
  pages        = {15},
  title        = {Partial convolution based multimodal autoencoder for art investigation},
  url          = {http://ceur-ws.org/Vol-2491/paper123.pdf},
  volume       = {2491},
  year         = {2019},
}