Show simple item record

dc.contributor.authorRiba, P
dc.contributor.authorDutta, A
dc.contributor.authorGoldmann, L
dc.contributor.authorFornés, A
dc.contributor.authorRamos, O
dc.contributor.authorLladós, J
dc.date.accessioned2019-10-31T10:10:55Z
dc.date.issued2020-02-03
dc.description.abstractTabular structures in documents offer a complementary dimension to the raw textual data, representing logical or quantitative relationships among pieces of information. In digital mail room applications, where a large amount of administrative documents must be processed with reasonable accuracy, the detection and interpretation of tables is crucial. Table recognition has gained interest in document image analysis, in particular in unconstrained formats (absence of rule lines, unknown information of rows and columns). In this work, we propose a graph-based approach for detecting tables in document images. Instead of using the raw content (recognized text), we make use of the location, context and content type, thus it is purely a structure perception approach, not dependent on the language and the quality of the text reading. Our framework makes use of Graph Neural Networks (GNNs) in order to describe the local repetitive structural information of tables in invoice documents. Our proposed model has been experimentally validated in two invoice datasets and achieved encouraging results. Additionally, due to the scarcity of benchmark datasets for this task, we have contributed to the community a novel dataset derived from the RVL-CDIP invoice data. It will be publicly released to facilitate future research.en_GB
dc.description.sponsorshipEuropean Unionen_GB
dc.identifier.citationICDAR 2019: 15th International Conference on Document Analysis and Recognition, 20-25 September 2019, Sydney, Australia, pp. 122-127.en_GB
dc.identifier.doi10.1109/ICDAR.2019.00028
dc.identifier.grantnumberRTI2018- 095645-B-C21en_GB
dc.identifier.grantnumberFPU15/06264en_GB
dc.identifier.grantnumberRYC-2014-16831en_GB
dc.identifier.grantnumber665919en_GB
dc.identifier.urihttp://hdl.handle.net/10871/39425
dc.language.isoenen_GB
dc.publisherInstitute of Electrical and Electronics Engineers (IEEE)en_GB
dc.rights© 2019 IEEE.
dc.subjectTable Detectionen_GB
dc.subjectAdministrative Documentsen_GB
dc.subjectGraph Representationsen_GB
dc.subjectGeometric Deep Learningen_GB
dc.subjectGraph Neural Networken_GB
dc.titleTable Detection in Invoice Documents by Graph Neural Networksen_GB
dc.typeConference paperen_GB
dc.date.available2019-10-31T10:10:55Z
dc.descriptionThis is the author accepted manuscript. The final version is available from IEEE via the DOI in this record.en_GB
dc.identifier.eissn2379-2140
dc.rights.urihttp://www.rioxx.net/licenses/all-rights-reserveden_GB
dcterms.dateAccepted2019-05-20
rioxxterms.versionAMen_GB
rioxxterms.licenseref.startdate2019-05-20
rioxxterms.typeConference Paper/Proceeding/Abstracten_GB
refterms.dateFCD2019-10-31T10:07:40Z
refterms.versionFCDAM
refterms.dateFOA2020-03-20T15:23:31Z
refterms.panelBen_GB


Files in this item

This item appears in the following Collection(s)

Show simple item record