Co-occurrence of deep convolutional features for image search
| dc.contributor.author | Forcén Carvalho, Juan Ignacio | |
| dc.contributor.author | Pagola Barrio, Miguel | |
| dc.contributor.author | Barrenechea Tartas, Edurne | |
| dc.contributor.author | Bustince Sola, Humberto | |
| dc.contributor.department | Estatistika, Informatika eta Matematika | eu |
| dc.contributor.department | Institute of Smart Cities - ISC | en |
| dc.contributor.department | Estadística, Informática y Matemáticas | es_ES |
| dc.date.accessioned | 2021-01-27T12:17:50Z | |
| dc.date.available | 2022-05-01T23:00:13Z | |
| dc.date.issued | 2020 | |
| dc.description.abstract | Image search can be tackled using deep features from pre-trained Convolutional Neural Networks (CNN). The feature map from the last convolutional layer of a CNN encodes descriptive information from which a discriminative global descriptor can be obtained. We propose a new representation of co-occurrences from deep convolutional features to extract additional relevant information from this last convolutional layer. Combining this co-occurrence map with the feature map, we achieve an improved image representation. We present two different methods to get the co-occurrence representation, the first one based on direct aggregation of activations, and the second one, based on a trainable co-occurrence representation. The image descriptors derived from our methodology improve the performance in very well-known image retrieval datasets as we prove in the experiments. | en |
| dc.embargo.lift | 2022-05-01 | |
| dc.embargo.terms | 2022-05-01 | |
| dc.format.extent | 30 p. | |
| dc.format.mimetype | application/pdf | en |
| dc.identifier.doi | 10.1016/j.imavis.2020.103909 | |
| dc.identifier.issn | 0262-8856 | |
| dc.identifier.uri | https://academica-e.unavarra.es/handle/2454/39071 | |
| dc.language.iso | eng | en |
| dc.publisher | Elsevier | en |
| dc.relation.ispartof | Image and Vision Computing, 2020, 97, 103909 | en |
| dc.relation.publisherversion | https://doi.org/10.1016/j.imavis.2020.103909 | |
| dc.rights | © 2020 Elsevier B.V. This manuscript version is made available under the CC-BY-NC-ND 4.0. | en |
| dc.rights.accessRights | info:eu-repo/semantics/openAccess | |
| dc.rights.uri | https://creativecommons.org/licenses/by-nc-nd/4.0/ | |
| dc.subject | Co-occurrence | en |
| dc.subject | Image retrieval | en |
| dc.subject | Feature aggregation | en |
| dc.subject | Pooling | en |
| dc.title | Co-occurrence of deep convolutional features for image search | en |
| dc.type | info:eu-repo/semantics/article | |
| dc.type.version | info:eu-repo/semantics/acceptedVersion | |
| dspace.entity.type | Publication | |
| relation.isAuthorOfPublication | 2980928f-fd1b-4d86-858b-f8c28e18b365 | |
| relation.isAuthorOfPublication | e5ab14f5-4f2e-4000-a415-0a7c3b28ec78 | |
| relation.isAuthorOfPublication | 4b0d58a5-e297-4e65-9a68-83649eb9c81b | |
| relation.isAuthorOfPublication | 1bdd7a0e-704f-48e5-8d27-4486444f82c9 | |
| relation.isAuthorOfPublication.latestForDiscovery | 2980928f-fd1b-4d86-858b-f8c28e18b365 |