manuscript-ocr Documentation
Manuscript OCR is an open-source Python framework for full OCR/HTR pipelines targeting pre-reform Russian manuscripts of the 18th-19th centuries as well as modern texts. The project focuses on digitization and analysis of historical textual heritage, with methods tailored to legacy orthography, complex page layouts, handwriting variation, and high computational efficiency under limited resources.