Details
| Original language | English |
|---|---|
| Pages (from-to) | 641-648 |
| Number of pages | 8 |
| Journal | ISPRS Journal of Photogrammetry and Remote Sensing |
| Volume | V-2-2020 |
| Publication status | Published - 3 Aug 2020 |
| Event | 2020 24th ISPRS Congress on Technical Commission II - Nice, Virtual, France Duration: 31 Aug 2020 → 2 Sept 2020 |
Abstract
This paper proposes several methods for training a Convolutional Neural Network (CNN) for learning the similarity between images of silk fabrics based on multiple semantic properties of the fabrics. In the context of the EU H2020 project SILKNOW (http://silknow.eu/), two variants of training were developed, one based on a Siamese CNN and one based on a triplet architecture. We propose different definitions of similarity and different loss functions for both training strategies, some of them also allowing the use of incomplete information about the training data. We assess the quality of the trained model by using the learned image features in a k-NN classification. We achieve overall accuracies of 93-95% and average F1-scores of 87-92%.
Keywords
- Convolutional Neural Networks, Cultural heritage, Image similarity, Incomplete training samples, Silk fabrics
ASJC Scopus subject areas
- Earth and Planetary Sciences(all)
- Earth and Planetary Sciences (miscellaneous)
- Environmental Science(all)
- Environmental Science (miscellaneous)
- Physics and Astronomy(all)
- Instrumentation
Cite this
- Standard
- Harvard
- Apa
- Vancouver
- BibTeX
- RIS
In: ISPRS Journal of Photogrammetry and Remote Sensing, Vol. V-2-2020, 03.08.2020, p. 641-648.
Research output: Contribution to journal › Conference article › Research › peer review
}
TY - JOUR
T1 - Assessing the Semantic Similarity of Images of Silk Fabrics Using Convolutional Neural Networks
AU - Clermont, D.
AU - Dorozynski, M.
AU - Wittich, D.
AU - Rottensteiner, F.
N1 - Funding Information: The research leading to these results is in the frame of the ”SILKNOW. Silk heritage in the Knowledge Society: from punched cards to big data, deep learning and visual/tangible simulations” project, which has received funding from the European Union’s Horizon 2020 research and innovation program under grant agreement No. 769504. We would also like to thank the Centre de Documentació i Museu Tèxtil, in particular Sílvia Saladrigas Cheng, for providing the data for this research and for giving us the permission to reproduce some of their images.
PY - 2020/8/3
Y1 - 2020/8/3
N2 - This paper proposes several methods for training a Convolutional Neural Network (CNN) for learning the similarity between images of silk fabrics based on multiple semantic properties of the fabrics. In the context of the EU H2020 project SILKNOW (http://silknow.eu/), two variants of training were developed, one based on a Siamese CNN and one based on a triplet architecture. We propose different definitions of similarity and different loss functions for both training strategies, some of them also allowing the use of incomplete information about the training data. We assess the quality of the trained model by using the learned image features in a k-NN classification. We achieve overall accuracies of 93-95% and average F1-scores of 87-92%.
AB - This paper proposes several methods for training a Convolutional Neural Network (CNN) for learning the similarity between images of silk fabrics based on multiple semantic properties of the fabrics. In the context of the EU H2020 project SILKNOW (http://silknow.eu/), two variants of training were developed, one based on a Siamese CNN and one based on a triplet architecture. We propose different definitions of similarity and different loss functions for both training strategies, some of them also allowing the use of incomplete information about the training data. We assess the quality of the trained model by using the learned image features in a k-NN classification. We achieve overall accuracies of 93-95% and average F1-scores of 87-92%.
KW - Convolutional Neural Networks
KW - Cultural heritage
KW - Image similarity
KW - Incomplete training samples
KW - Silk fabrics
UR - http://www.scopus.com/inward/record.url?scp=85091092275&partnerID=8YFLogxK
U2 - 10.5194/isprs-annals-V-2-2020-641-2020
DO - 10.5194/isprs-annals-V-2-2020-641-2020
M3 - Conference article
AN - SCOPUS:85091092275
VL - V-2-2020
SP - 641
EP - 648
JO - ISPRS Journal of Photogrammetry and Remote Sensing
JF - ISPRS Journal of Photogrammetry and Remote Sensing
SN - 0924-2716
T2 - 2020 24th ISPRS Congress on Technical Commission II
Y2 - 31 August 2020 through 2 September 2020
ER -