Volltext-Downloads (blau) und Frontdoor-Views (grau)

Fine-Grained Product Classification on Leaflet Advertisements

  • In this paper, we describe a first publicly available fine-grained product recognition dataset based on leaflet images. Using advertisement leaflets, collected over several years from different European retailers, we provide a total of 41.6k manually annotated product images in 832 classes. Further, we investigate three different approaches for this fine-grained product classification task,In this paper, we describe a first publicly available fine-grained product recognition dataset based on leaflet images. Using advertisement leaflets, collected over several years from different European retailers, we provide a total of 41.6k manually annotated product images in 832 classes. Further, we investigate three different approaches for this fine-grained product classification task, Classification by Image, Classification by Text, as well as Classification by Image and Text. The last both approaches use the text extracted directly from the leaflet product images. We show, that the combination of image and text as input improves the classification of visual difficult to distinguish products. The final model leads to an accuracy of 96.4% with a Top-3 score of 99.2%. https://github.com/ladwigd/Leaflet-Product-Classification…zeige mehrzeige weniger

Metadaten exportieren

Weitere Dienste

Suche bei Google Scholar

Statistik

frontdoor_oas
Metadaten
Dokumentart:Konferenzveröffentlichung
Art der Konferenzveröffentlichung:Konferenzartikel
Zitierlink: https://opus.hs-offenburg.de/10147
Bibliografische Angaben
Titel (Englisch):Fine-Grained Product Classification on Leaflet Advertisements
Konferenzangaben:The Upper Rhine Artificial Intelligence Symposium (5. : 16-17 November 2023 : Mulhouse, France)
Verfasserangaben:Daniel LadwigStaff Member, Bianca LammStaff MemberORCiD, Janis KeuperStaff MemberORCiDGND
Erscheinungsjahr:2024
Urhebende Körperschaft:ENSISA-IRIMAS
Erste Seite:144
Letzte Seite:151
Titel des übergeordneten Werkes (Englisch):UR-AI2023 : The Upper-Rhine Artificial Intelligence Symposium : Artificial Intelligence for Time Series, Robotics and Beyond
Herausgeber*in:Jean-Philippe Lauffenburger, Jonathan Weber
URL:https://urai2023.sciencesconf.org/data/pages/book_urai2023_en_2024.pdf
Sprache:Englisch
Inhaltliche Informationen
Fakultäten / Einrichtungen:Fakultät Elektrotechnik, Medizintechnik und Informatik (EMI) (ab 04/2019)
Forschung:IMLA - Institute for Machine Learning and Analytics
Sammlungen der Hochschule Offenburg:Bibliografie
Freies Schlagwort / Tag:fine-grained; image classification; leaflets; products; retail; text extraction
Formale Angaben
Relevanz für "Jahresbericht über Forschungsleistungen":1-fach | Konferenzbeitrag
Open-Access-Status: Open Access 
 Bronze 
Lizenz (Deutsch):License LogoUrheberrechtlich geschützt