Högskolan i Skövde

his.sePublications
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • apa-cv
  • 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
AdAPT: Advertisement detector adaptation under newspaper domain shift with null-based pseudo-labeling
Media Research AB, Östersund, Sweden ; Mid Sweden University, Sundsvall, Sweden.ORCID iD: 0000-0001-6418-2876
University of Skövde, School of Engineering Science. University of Skövde, Virtual Engineering Research Environment. ITAB Shop Products AB, Jönköping, Sweden ; Jönköping University, Sweden. (Virtual Production Development (VPD))ORCID iD: 0000-0001-8880-7965
Mid Sweden University, Sundsvall, Sweden.ORCID iD: 0000-0001-8661-7578
Mid Sweden University, Sundsvall, Sweden.ORCID iD: 0000-0002-3774-4850
Show others and affiliations
2026 (English)In: Machine Learning with Applications, E-ISSN 2666-8270, Vol. 23, article id 100806Article in journal (Refereed) Published
Abstract [en]

Detecting advertisements in digitized newspapers is a key step in large-scale media analytics and digital archiving. However, variations in layout, typography, and advertisement design across publishers and time periods cause significant domain shifts that reduce the generalization ability of supervised detectors. This paper presents AdAPT, a confidence-guided pseudo-labeling pipeline for unsupervised domain adaptation in advertisement detection. The proposed method leverages both advertisement-free (Null) and advertisement-containing pages from unlabeled target domains to generate reliable pseudo-labels. By retraining a YOLO-based detector using labeled source data combined with filtered pseudo-labeled target samples, AdAPT achieves robust adaptation without requiring manual annotation. Experiments conducted on two unseen newspapers (Adresseavisen and iTromsø) demonstrate that Null-based pseudo-labeling provides the most stable and accurate adaptation, yielding up to 38% error reduction compared to the baseline. The results highlight AdAPT as a simple, scalable, and annotation-efficient solution for maintaining high-performance advertisement detection across diverse newspaper collections.

Place, publisher, year, edition, pages
Elsevier, 2026. Vol. 23, article id 100806
Keywords [en]
Cross-domain advertisement detection, Deep learning, Object detection, Domain adaptation, Pseudo labeling
National Category
Computer Sciences Computer graphics and computer vision
Research subject
Virtual Production Development (VPD)
Identifiers
URN: urn:nbn:se:his:diva-26067DOI: 10.1016/j.mlwa.2025.100806ISI: 001635811600001OAI: oai:DiVA.org:his-26067DiVA, id: diva2:2021634
Funder
Knowledge FoundationMid Sweden University
Note

CC BY 4.0

Corresponding author: Faeze Zakaryapour Sayyad

Received 16 October 2025, Revised 21 November 2025, Accepted 30 November 2025, Available online 1 December 2025, Version of Record 3 December 2025.

Declaration of competing interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: Faeze Zakaryapour Sayyad reports financial support was provided by Mid Sweden University. Faeze Zakaryapour Sayyad reports a relationship with Knowledge Foundation (kks.se) within the Industrial graduate school Smart Industry Sweden, Media Research company that includes: employment and funding grants. If there are other authors, they declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgment

This work was supported in part by The Knowledge Foundation(kks.se) within the Industrial graduate school Smart Industry Sweden and Media Research AB. The authors would also like to thank Oscar Berg for his valuable comments on this work.

Available from: 2025-12-15 Created: 2025-12-15 Last updated: 2025-12-30Bibliographically approved

Open Access in DiVA

fulltext(2085 kB)49 downloads
File information
File name FULLTEXT01.pdfFile size 2085 kBChecksum SHA-512
268264ce26d93585645e8d05591a9a445ef1e4210d3ab61c2d1351e2698b07879ebebba0ad7e1aa364176728ab3b2e7c98eb960e281f07eb87520e2d89ab8f54
Type fulltextMimetype application/pdf

Other links

Publisher's full text

Authority records

Pettersson, Tobias

Search in DiVA

By author/editor
Zakaryapour Sayyad, FaezePettersson, TobiasSeyed Jalaleddin, MousaviradShallari, IridaO’Nils, Mattias
By organisation
School of Engineering ScienceVirtual Engineering Research Environment
In the same journal
Machine Learning with Applications
Computer SciencesComputer 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

doi
urn-nbn

Altmetric score

doi
urn-nbn
Total: 277 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • apa-cv
  • 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