Please use this identifier to cite or link to this item:
https://hdl.handle.net/20.500.13091/3669
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Örnek, Ahmet H. | - |
dc.contributor.author | Ceylan, Murat | - |
dc.date.accessioned | 2023-03-03T13:32:24Z | - |
dc.date.available | 2023-03-03T13:32:24Z | - |
dc.date.issued | 2022 | - |
dc.identifier.issn | 0765-0019 | - |
dc.identifier.issn | 1958-5608 | - |
dc.identifier.uri | https://doi.org/10.18280/ts.390529 | - |
dc.identifier.uri | https://hdl.handle.net/20.500.13091/3669 | - |
dc.description.abstract | Explaining the decision mechanism of Deep Convolutional Neural Networks (CNNs) is a new and challenging area because of the Black Box nature of CNN's. Class Activation Mapping (CAM) as a visual explainable method is used to highlight important regions of input images by using classification gradients. The lack of the current methods is to use all of the filters in the last convolutional layer which causes scattered and unfocused activation mapping. HayCAMas a novel visualization method provides better activation mapping and therefore better localization by using dimension reduction. It has been shown with mask detection use case that input images are fed into the CNN model and bounding boxes are drawn over the generated activation maps (i.e. weakly-supervised object detection) by three different CAM methods. IoU values are obtained as 0.1922 for GradCAM, 0.2472 for GradCAM++, 0.3386 for EigenCAM, and 0.3487 for the proposed HayCAM. The results show that HayCAM achieves the best activation mapping with dimension reduction. | en_US |
dc.description.sponsorship | Huawei Turkey RD Center | en_US |
dc.description.sponsorship | This study was supported by Huawei Turkey R&D Center. | en_US |
dc.language.iso | en | en_US |
dc.publisher | Int Information & Engineering Technology Assoc | en_US |
dc.relation.ispartof | Traitement Du Signal | en_US |
dc.rights | info:eu-repo/semantics/openAccess | en_US |
dc.subject | classification | en_US |
dc.subject | class activation mapping | en_US |
dc.subject | explainable artificial intelligence | en_US |
dc.subject | visual explanation | en_US |
dc.subject | weakly-supervised object detection | en_US |
dc.title | HayCAM: A Novel Visual Explanation for Deep Convolutional Neural Networks | en_US |
dc.type | Article | en_US |
dc.identifier.doi | 10.18280/ts.390529 | - |
dc.identifier.scopus | 2-s2.0-85150222892 | en_US |
dc.department | KTÜN | en_US |
dc.identifier.volume | 39 | en_US |
dc.identifier.issue | 5 | en_US |
dc.identifier.startpage | 1711 | en_US |
dc.identifier.endpage | 1719 | en_US |
dc.identifier.wos | WOS:000907630800022 | en_US |
dc.institutionauthor | … | - |
dc.relation.publicationcategory | Makale - Uluslararası Hakemli Dergi - Kurum Öğretim Eleman | en_US |
dc.identifier.scopusquality | Q3 | - |
item.grantfulltext | open | - |
item.openairetype | Article | - |
item.fulltext | With Fulltext | - |
item.openairecristype | http://purl.org/coar/resource_type/c_18cf | - |
item.languageiso639-1 | en | - |
item.cerifentitytype | Publications | - |
crisitem.author.dept | 02.04. Department of Electrical and Electronics Engineering | - |
Appears in Collections: | Scopus İndeksli Yayınlar Koleksiyonu / Scopus Indexed Publications Collections WoS İndeksli Yayınlar Koleksiyonu / WoS Indexed Publications Collections |
Files in This Item:
File | Size | Format | |
---|---|---|---|
39.05_29.pdf | 1.48 MB | Adobe PDF | View/Open |
CORE Recommender
WEB OF SCIENCETM
Citations
1
checked on Apr 27, 2024
Page view(s)
452
checked on Apr 29, 2024
Download(s)
130
checked on Apr 29, 2024
Google ScholarTM
Check
Altmetric
Items in GCRIS Repository are protected by copyright, with all rights reserved, unless otherwise indicated.