Malmö University Publications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
RetinaGate: A Gated Feature Pyramid Network for Improved Object Detection with SE-based Attention
Malmö University, Faculty of Technology and Society (TS), Department of Computer Science and Media Technology (DVMT). Malmö University, Sustainable Digitalisation Research Centre (SDRC).ORCID iD: 0000-0002-9464-7010
Malmö University, Faculty of Technology and Society (TS), Department of Computer Science and Media Technology (DVMT). Malmö University, Sustainable Digitalisation Research Centre (SDRC).ORCID iD: 0000-0003-0998-6585
Malmö University, Faculty of Technology and Society (TS), Department of Computer Science and Media Technology (DVMT). Malmö University, Sustainable Digitalisation Research Centre (SDRC).ORCID iD: 0000-0002-3797-4605
Axis Communications AB, Lund, Sweden.
Show others and affiliations
2025 (English)In: Proceedings of Swedish AI Society Workshop 2025 (SAIS 2025) / [ed] Sławomir Nowaczyk; Anna Vettoruzzo, CEUR , 2025, p. 1-11Conference paper, Published paper (Refereed)
Abstract [en]

Object detection is a critical task in computer vision with wide-ranging applications, from autonomous driving tosurveillance systems. Despite notable progress, challenges such as detecting small objects, managing occlusions,and effectively integrating multiscale features persist. We propose RetinaGate, a novel object detection architec-ture that introduces a Gated Feature Pyramid Network (G-FPN) to adaptively fuse multi-scale features, enhancedby Squeeze-and-Excitation-based channel attention for improved accuracy. As a plug-and-play module, G-FPNcan be seamlessly integrated into existing detection models to enhance their accuracy. These enhancementsstrengthen the model’s capacity to capture fine-grained details and leverage contextual information more effec-tively. Experimental results on three benchmark datasets demonstrate that RetinaGate outperforms the baselineRetinaNet in terms of detection accuracy, particularly in challenging detection scenarios such as underwater.

Place, publisher, year, edition, pages
CEUR , 2025. p. 1-11
Series
CEUR Workshop Proceedings, E-ISSN 1613-0073 ; 4037
Keywords [en]
Object Detection, RetinaNet, FPN, Gated Fusion, RetinaGate, SEBlock
National Category
Computer graphics and computer vision
Identifiers
URN: urn:nbn:se:mau:diva-79975Scopus ID: 2-s2.0-105017747184OAI: oai:DiVA.org:mau-79975DiVA, id: diva2:2005678
Conference
Swedish AI Society Workshop 2025 (SAIS 2025) Halmstad, Sweden, 16-17 June 2025.
Available from: 2025-10-10 Created: 2025-10-10 Last updated: 2025-10-14Bibliographically approved

Open Access in DiVA

fulltext(613 kB)93 downloads
File information
File name FULLTEXT01.pdfFile size 613 kBChecksum SHA-512
52d33428d7d4cc56066e8f16cf60c77c26d4299c7018c840d252bc9a04dba0a94aa532237f93288275bf567a0764e7cce34ee10d345c1bdb854f4eb9e6ff17c5
Type fulltextMimetype application/pdf

Other links

ScopusFulltext

Authority records

Jamali, MahtabDavidsson, PaulKhoshkangini, RezaMihailescu, Radu-Casian

Search in DiVA

By author/editor
Jamali, MahtabDavidsson, PaulKhoshkangini, RezaMihailescu, Radu-Casian
By organisation
Department of Computer Science and Media Technology (DVMT)Sustainable Digitalisation Research Centre (SDRC)
Computer graphics and computer vision

Search outside of DiVA

GoogleGoogle Scholar
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 1878 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf