AMSMB HTR Segmentation model for Medieval Notarial Manuscripts

AMSMB HTR Segmentation is a model trained using <a href=https://kraken.re/main/index.html target="_blank" rel="nofollow">Kraken</a> for the task of line segmentation on handwritten historical documents. The model uses the default Kraken segmentation model (<a href=https://github.com/mittagessen/kraken/blob/main/kraken/blla.mlmodel target="_blank" rel="nofollow">blla.mlmodel</a>) as base model and is fine-tuned on the <a href=https://dataverse.bsc.es/dataset.xhtml?persistentId=perma:BSC/0VB0MC target="_blank" rel="nofollow">AMSMB HTR</a> dataset. Details on how the model was trained are reported in the README file.