Show simple item record  

dc.contributor.authorVetrova, Varvaraen_NZ
dc.contributor.authorCoup, Sheldonen_NZ
dc.contributor.authorFrank, Eibeen_NZ
dc.contributor.authorCree, Michael J.en_NZ
dc.coverage.spatialAuckland, New Zealanden_NZ
dc.date.accessioned2019-08-20T21:35:14Z
dc.date.available2018-01-01en_NZ
dc.date.available2019-08-20T21:35:14Z
dc.date.issued2018en_NZ
dc.identifier.citationVetrova, V., Coup, S., Frank, E., & Cree, M. J. (2018). Hidden features: Experiments with feature transfer for fine-grained multi-class and one-class image categorization. In 2018 International conference on image and vision computing New Zealand (IVCNZ). Auckland, New Zealand: IEEE.en
dc.identifier.issn2151-2191en_NZ
dc.identifier.urihttps://hdl.handle.net/10289/12793
dc.description.abstractCan we apply out-of-the box feature transfer using pre-trained convolutional neural networks in fine-grained multiclass image categorization tasks? What is the effect of (a) domainspecific fine-tuning and (b) a special-purpose network architecture designed and trained specifically for the target domain? How do these approaches perform in one-class classification? We investigate these questions by tackling two biological object recognition tasks: classification of “cryptic” plants of genus Coprosma and identification of New Zealand moth species. We compare results based on out-of-the-box features extracted using a pre-trained state-of-the-art network to those obtained by finetuning to the target domain, and also evaluate features learned using a simple Siamese network trained only on data from the target domain. For each extracted feature set, we test a number of classifiers, e.g., support vector machines. In addition to multiclass classification, we also consider one-class classification, a scenario that is particularly relevant to biosecurity applications. In the multi-class setting, we find that out-of-the-box lowlevel features extracted from the generic pre-trained network yield high accuracy (90.76%) when coupled with a simple LDA classifier. Fine-tuning improves accuracy only slightly (to 91.6%). Interestingly, features extracted from the much simpler Siamese network trained on data from the target domain lead to comparable results (90.8%). In the one-class classification setting, we note high variability in the area under the ROC curve across feature sets, opening up the possibility of considering an ensemble approach.
dc.format.mimetypeapplication/pdf
dc.language.isoen
dc.publisherIEEEen_NZ
dc.rightsThis is an author’s accepted version of a paper published in the Proceedings: 2018 International conference on image and vision computing New Zealand (IVCNZ). © 2018 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.
dc.sourceInternational Conference on Image and Vision Computing New Zealand (IVCNZ)en_NZ
dc.subjectScience & Technologyen_NZ
dc.subjectTechnologyen_NZ
dc.subjectComputer Science, Artificial Intelligenceen_NZ
dc.subjectImaging Science & Photographic Technologyen_NZ
dc.subjectComputer Scienceen_NZ
dc.subjectfine-grained categorizationen_NZ
dc.subjectfeature transferen_NZ
dc.subjectone-class classificationen_NZ
dc.subjectSiamese networksen_NZ
dc.subjectSIMILARITYen_NZ
dc.subjectMachine learning
dc.titleHidden features: Experiments with feature transfer for fine-grained multi-class and one-class image categorizationen_NZ
dc.typeConference Contribution
dc.relation.isPartOf2018 International conference on image and vision computing New Zealand (IVCNZ)en_NZ
pubs.elements-id235901
pubs.finish-date2018-11-21en_NZ
pubs.publication-statusPublisheden_NZ
pubs.start-date2018-11-19en_NZ


Files in this item

This item appears in the following Collection(s)

Show simple item record