dc.contributor.author | Ciortan, Irina-Mihaela | |
dc.contributor.author | Arteaga, Yoko | |
dc.contributor.author | George, Sony | |
dc.contributor.author | Hardeberg, Jon Yngve | |
dc.date.accessioned | 2023-03-10T10:33:54Z | |
dc.date.available | 2023-03-10T10:33:54Z | |
dc.date.created | 2023-01-16T14:04:03Z | |
dc.date.issued | 2022 | |
dc.identifier.isbn | 978-3-031-20302-2 | |
dc.identifier.uri | https://hdl.handle.net/11250/3057655 | |
dc.description.abstract | The characterization of a painter’s style is useful for a series of applications, such as documenting art history, planning style-aware conservation and restoration, and discarding forgery attempts. In this work, we propose a method to assign paintings to the right artist with two strategies: traditional machine learning and deep learning. In particular, we quantify the visual characteristics of a painting at multiple scales, covering low-level as well as mid-level features (pyramid of histogram of oriented gradients, residual convolutional neural network features). We focus on coeval artists, representing Impressionism, Expressionism and Cubism art periods. Our results are consistent with state-of-the-art findings in art and computer vision literature. | en_US |
dc.language.iso | eng | en_US |
dc.publisher | Springer | en_US |
dc.relation.ispartof | The Future of Heritage Science and Technologies: ICT and Digital Heritage | |
dc.rights | Navngivelse 4.0 Internasjonal | * |
dc.rights.uri | http://creativecommons.org/licenses/by/4.0/deed.no | * |
dc.title | Multi-scale Painter Classification | en_US |
dc.title.alternative | Multi-scale Painter Classification | en_US |
dc.type | Chapter | en_US |
dc.description.version | acceptedVersion | en_US |
dc.identifier.cristin | 2107741 | |
cristin.ispublished | true | |
cristin.fulltext | postprint | |
cristin.qualitycode | 1 | |