GLOBALISE Laypa Region Model - August 2023

GLOBALISE project (opent in nieuw tabblad)

This is a Laypa region detection model that was created to detect and identify regions (such as page number, heading, paragraph, and marginalia) on the scans of the GLOBALISE VOC corpus. It was trained on Ground Truth that is also available in this Dataverse (GLOBALISE Ground Truth for Handwritten Text and Layout Recognition) and applied using the Loghi Handwritten Text Recognition tools to generate VOC transcriptions v2 - GLOBALISE. The model's hyperparameters are available in the file config.yaml.

URI
https://hdl.handle.net/10622/DMAS2T (opent in nieuw tabblad)
Uitgever & maker
GLOBALISE project (opent in nieuw tabblad) Meer van deze uitgever
Onderdeel van
https://datasets.iisg.amsterdam/dataverse/globalise (opent in nieuw tabblad) Doorzoek catalogus
Licentie
CC BY 4.0 Internationaal (opent in nieuw tabblad)
Talen
Engels, Nederlands
Trefwoorden
Artificial Intelligence ModelArts and HumanitiesComputer and Information ScienceHandwriting Recognition
Uitgegeven
15 april 2024
Gewijzigd
15 april 2024

Distributies

Download deze dataset (opent in nieuw tabblad)
Linked Data-samenvatting laden...

Registratie

Geregistreerde URL
https://raw.githubusercontent.com/globalise-huygens/knowledge-graph/main/datasets/model_laypa_region_2023.jsonld (opent in nieuw tabblad) Geldig (opent in nieuw tabblad)
Geregistreerd op
15 april 2024 om 12:39
Laatst gelezen
23 uur geleden
Volledigheid
95%
Ontbrekende eigenschappen: dcterms:spatial