Tesseract does not understand layout. It’s fine for character recognition, but if I still have to pipe the output to a LLM to make sense of the layout and fix common transcription errors, I might as well use a single model. It’s also easier for a visual LLM to extract figures and tables in one pass.

▲

chaps 4 hours ago | parent | next [-]

For my workflows, layout extraction has been so inconsistent that I've stopped attempting to use it. It's simpler to just throw everything into postgis and run intersection checks on size-normalized pages.

▲

kergonath 3 hours ago | parent [-]

Interesting. What kind of layout do you have?

My documents have one or two-column layouts, often inconsistently across pages or even within a page (which tripped older layout detection methods). Most models seem to understand that well enough so they are good enough for my use case.

	▲	chaps 3 hours ago \| parent [-]
		Documents that come from FOIA. So, some scanned, some not. Lots of forms and lots of hand writing to add info that the form format doesn't recognize. Lots of repeated documents, but lots of one-off documents that have high signal.

▲

fudged71 an hour ago | parent | prev [-]

I don't know how, but PyMuPDF4LLM is based on Tessaract and has GNN-based layout detection