mittagessen

kraken. OCR engine for all the languages

party. Page-wise text recognition with lower-supervision line data models

kraken-models. Recognition Models for Kraken and CLSTM

curt. Python

dfine_kraken. D-FINE for document region segmentation

orli. Orli is ordered regression of lines

page2darknet. conversion script for PageXML to darknet bbox format

seg. neural layout analysis tests.

conformer_ocr. text recognizer with a conformer

pangoline. Synthetic data generation for ATR training

patchseg. patchwise semantic segmentation

bytellama. A llama with octet tokenization

char_lm. LMs for OCR postcorrection

dmml_notebooks. IPython notebooks for the 2023 image analysis course

segfblla. Python

ntcn_ocr. Python

alto_ro_examples. Examples on how to use the proposed reading order mechanism for ALTO 4.3

dmml_notes. DMML 2023 lecture notes

cutout. classification of potential cutouts in manuscripts