Show simple item record

Files in this item

Thumbnail

Item metadata

dc.contributor.authorJiang, Ai
dc.contributor.authorNacenta, Miguel A.
dc.contributor.authorYe, Juan
dc.date.accessioned2022-10-11T10:30:02Z
dc.date.available2022-10-11T10:30:02Z
dc.date.issued2022-09-24
dc.identifier279836022
dc.identifier4cdf82f7-c432-4fb0-bc3e-19e248d183b3
dc.identifier85144345067
dc.identifier000888899400004
dc.identifier.citationJiang , A , Nacenta , M A & Ye , J 2022 , ' VisuaLizations As Intermediate Representations (VLAIR) : an approach for applying deep learning-based computer vision to non-image-based data ' , Visual Informatics , vol. 6 , no. 3 , pp. 35-50 . https://doi.org/10.1016/j.visinf.2022.05.001en
dc.identifier.issn2468-502X
dc.identifier.otherRIS: urn:353DE926F80691DA9AB5C1A6127AE563
dc.identifier.otherORCID: /0000-0002-2838-6836/work/120849895
dc.identifier.urihttps://hdl.handle.net/10023/26172
dc.descriptionFunding: We thank the China Scholarship Council (CSC) for financially supporting my PhD study at University of St Andrews, UK, and NSERC Discovery Grant 2020-04401 (Miguel Nacenta).en
dc.description.abstractDeep learning algorithms increasingly support automated systems in areas such as human activity recognition and purchase recommendation. We identify a current trend in which data is transformed first into abstract visualizations and then processed by a computer vision deep learning pipeline. We call this VisuaLization As Intermediate Representation (VLAIR) and believe that it can be instrumental to support accurate recognition in a number of fields while also enhancing humans’ ability to interpret deep learning models for debugging purposes or in personal use. In this paper we describe the potential advantages of this approach and explore various visualization mappings and deep learning architectures. We evaluate several VLAIR alternatives for a specific problem (human activity recognition in an apartment) and show that VLAIR attains classification accuracy above classical machine learning algorithms and several other non-image-based deep learning algorithms with several data representations.
dc.format.extent16
dc.format.extent3075265
dc.language.isoeng
dc.relation.ispartofVisual Informaticsen
dc.subjectInformation visualizationen
dc.subjectConvolutional neural networksen
dc.subjectHuman activity recognitionen
dc.subjectSmart homesen
dc.subjectData representationen
dc.subjectIntermediate representationsen
dc.subjectInterpretabilityen
dc.subjectMachine learningen
dc.subjectDeep learningen
dc.subjectQA75 Electronic computers. Computer scienceen
dc.subject3rd-NDASen
dc.subjectMCCen
dc.subject.lccQA75en
dc.titleVisuaLizations As Intermediate Representations (VLAIR) : an approach for applying deep learning-based computer vision to non-image-based dataen
dc.typeJournal articleen
dc.contributor.institutionUniversity of St Andrews. School of Computer Scienceen
dc.identifier.doi10.1016/j.visinf.2022.05.001
dc.description.statusPeer revieweden


This item appears in the following Collection(s)

Show simple item record