eScriptorium Workshop

2025-11-13

• hpc
• openOnDemand

• zenodo - models
• party OCR - future OCR
• htr united catalog
• catmus transcribing guides
• catmus paper
• Hand classification with DINO + K-Means + UMAP
• Kraken + YOLO for segmentation - this is the recommended approach
• RTK - Pipeline for Segmentation with YALTAi, Clean Up, Kraken
• A Controlled Vocabulary to Describe and Process Digital Facsimiles
• Annotation Guidelines
• SegmOnto - Layout analysis model trained with YALTAi, relying on YOLO models, and Kraken
• ALTO to TEI
• Fine-tuning a BERT Model to Explore a Stylistic Phenomenon
• Apertus EPFL : latest open swiss model