A Few-Shot Attention Recurrent Residual U-Net for Crack Segmentation
Recent studies indicate that deep learning plays a crucial role in the automated visual inspection of road infrastructures. However, current learning schemes are static, implying no dynamic adaptation to users' feedback. To address this drawback, we present a few-shot learning paradigm for the...
Saved in:
Main Authors | , , , , , |
---|---|
Format | Journal Article |
Language | English |
Published |
02.03.2023
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Abstract | Recent studies indicate that deep learning plays a crucial role in the
automated visual inspection of road infrastructures. However, current learning
schemes are static, implying no dynamic adaptation to users' feedback. To
address this drawback, we present a few-shot learning paradigm for the
automated segmentation of road cracks, which is based on a U-Net architecture
with recurrent residual and attention modules (R2AU-Net). The retraining
strategy dynamically fine-tunes the weights of the U-Net as a few new rectified
samples are being fed into the classifier. Extensive experiments show that the
proposed few-shot R2AU-Net framework outperforms other state-of-the-art
networks in terms of Dice and IoU metrics, on a new dataset, named CrackMap,
which is made publicly available at https://github.com/ikatsamenis/CrackMap. |
---|---|
AbstractList | Recent studies indicate that deep learning plays a crucial role in the
automated visual inspection of road infrastructures. However, current learning
schemes are static, implying no dynamic adaptation to users' feedback. To
address this drawback, we present a few-shot learning paradigm for the
automated segmentation of road cracks, which is based on a U-Net architecture
with recurrent residual and attention modules (R2AU-Net). The retraining
strategy dynamically fine-tunes the weights of the U-Net as a few new rectified
samples are being fed into the classifier. Extensive experiments show that the
proposed few-shot R2AU-Net framework outperforms other state-of-the-art
networks in terms of Dice and IoU metrics, on a new dataset, named CrackMap,
which is made publicly available at https://github.com/ikatsamenis/CrackMap. |
Author | Bakalos, Nikolaos Katsamenis, Iason Doulamis, Anastasios Voulodimos, Athanasios Protopapadakis, Eftychios Doulamis, Nikolaos |
Author_xml | – sequence: 1 givenname: Iason surname: Katsamenis fullname: Katsamenis, Iason – sequence: 2 givenname: Eftychios surname: Protopapadakis fullname: Protopapadakis, Eftychios – sequence: 3 givenname: Nikolaos surname: Bakalos fullname: Bakalos, Nikolaos – sequence: 4 givenname: Anastasios surname: Doulamis fullname: Doulamis, Anastasios – sequence: 5 givenname: Nikolaos surname: Doulamis fullname: Doulamis, Nikolaos – sequence: 6 givenname: Athanasios surname: Voulodimos fullname: Voulodimos, Athanasios |
BackLink | https://doi.org/10.48550/arXiv.2303.01582$$DView paper in arXiv |
BookMark | eNotj8lOwzAURb2ABRQ-gBX-AQcP8bSMAgWkCqQO6-g1foaINkGuy_D3TQure690dKVzSc76oUdCbgQvSqc1v4P0030VUnFVcKGdvCD3FZ3iN1u8D5lWOWOfu6Gnc2z3KY1jbLsu7GFDV-wFM41DonWC9oMu8G07AnDkr8h5hM0Or_9zQpbTh2X9xGavj891NWNgrGQetHel5cK7tSyxxSCtidFycCDAuxjWMijOubFOKiURpfO6FMKAjCYINSG3f7cnjeYzdVtIv81RpznpqAMrPUVZ |
ContentType | Journal Article |
Copyright | http://creativecommons.org/licenses/by/4.0 |
Copyright_xml | – notice: http://creativecommons.org/licenses/by/4.0 |
DBID | AKY GOX |
DOI | 10.48550/arxiv.2303.01582 |
DatabaseName | arXiv Computer Science arXiv.org |
DatabaseTitleList | |
Database_xml | – sequence: 1 dbid: GOX name: arXiv.org url: http://arxiv.org/find sourceTypes: Open Access Repository |
DeliveryMethod | fulltext_linktorsrc |
ExternalDocumentID | 2303_01582 |
GroupedDBID | AKY GOX |
ID | FETCH-LOGICAL-a672-9a598470198b24eced276ff70a8a1a98fdb2d30006782332ee28954116a2f6d13 |
IEDL.DBID | GOX |
IngestDate | Mon Jan 08 05:39:47 EST 2024 |
IsDoiOpenAccess | true |
IsOpenAccess | true |
IsPeerReviewed | false |
IsScholarly | false |
Language | English |
LinkModel | DirectLink |
MergedId | FETCHMERGED-LOGICAL-a672-9a598470198b24eced276ff70a8a1a98fdb2d30006782332ee28954116a2f6d13 |
OpenAccessLink | https://arxiv.org/abs/2303.01582 |
ParticipantIDs | arxiv_primary_2303_01582 |
PublicationCentury | 2000 |
PublicationDate | 2023-03-02 |
PublicationDateYYYYMMDD | 2023-03-02 |
PublicationDate_xml | – month: 03 year: 2023 text: 2023-03-02 day: 02 |
PublicationDecade | 2020 |
PublicationYear | 2023 |
Score | 1.8782235 |
SecondaryResourceType | preprint |
Snippet | Recent studies indicate that deep learning plays a crucial role in the
automated visual inspection of road infrastructures. However, current learning
schemes... |
SourceID | arxiv |
SourceType | Open Access Repository |
SubjectTerms | Computer Science - Computer Vision and Pattern Recognition Computer Science - Learning |
Title | A Few-Shot Attention Recurrent Residual U-Net for Crack Segmentation |
URI | https://arxiv.org/abs/2303.01582 |
hasFullText | 1 |
inHoldings | 1 |
isFullTextHit | |
isPrint | |
link | http://utb.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwdV1LS8QwEB7WPXkRRWV9koPXaJu0aXosq3URXMHdhd6WPFXEVbr18fOdphW9eAtJCMyE5PsmmQfAmdCeOxF7qnViacKlpspzjcfdpIY546JQM_J2KiaL5KZKqwGQn1gYVX89fXT5gfX6AvkxP0fAknjJbjDWumxd31Xd52RIxdXP_52HHDN0_QGJchu2enZHim47dmDgVrtwWZDSfdLZ42tDiqbp_AvJffvQ3aZGwtY6RESRBZ26hiCLJONamWcycw8vfWjQag_m5dV8PKF98QKqRMZortIcL34kUFKzBGW2LBPeZ5GSKla59FYzyzuwYJwz59DySZM4Fop5YWO-D0O0_90IiE_zyOWa-QxXlNwraYRHlsZjxFabmgMYBZGXb11-imWrjWXQxuH_Q0ew2VZOD-5U7BiGTf3uThBfG30alPwNiRF5ig |
link.rule.ids | 228,230,786,891 |
linkProvider | Cornell University |
openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=A+Few-Shot+Attention+Recurrent+Residual+U-Net+for+Crack+Segmentation&rft.au=Katsamenis%2C+Iason&rft.au=Protopapadakis%2C+Eftychios&rft.au=Bakalos%2C+Nikolaos&rft.au=Doulamis%2C+Anastasios&rft.date=2023-03-02&rft_id=info:doi/10.48550%2Farxiv.2303.01582&rft.externalDocID=2303_01582 |