Main Page/Methodology

From HBMa
Revision as of 13:07, 21 December 2023 by Daniel Baránek (talk | contribs) (Created page with "Digitization of scanned registry books uses modern machine learning methods like layout segmentation, handwritten text recognition ([https://kraken.re/main/index.html Kraken], [https://gitlab.com/scripta/escriptorium eScriptorium]) and [https://happytransformer.com/text-to-text/ text-to-text transformers]. Tools")
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

Digitization of scanned registry books uses modern machine learning methods like layout segmentation, handwritten text recognition (Kraken, eScriptorium) and text-to-text transformers.

Tools

Retrieved from "https://matrika.wikibase.cloud/w/index.php?title=Main_Page/Methodology&oldid=4744"