Segmentation is the hidden tax in historical HTR

Automated excerpt

It is the step nobody benchmarks: line segmentation, cutting the page into the individual text lines the recogniser expects. A generic line detector trained on modern documents degrades quietly here. It finds 120 text lines, cell by cell, top to bottom: The detector reaches roughly mAP50 ~0. 79 on held-out historical pages.

Selected automatically from source text; not independently written or fact-checked. Read the original for full context.