Benjamin Admin 40a77a82f6 fix(ocr-pipeline): use midpoint boundaries for column word assignment ...

Replace containment-with-padding approach with midpoint-based column
ranges. For adjacent columns, the assignment boundary is the midpoint
between them (Voronoi-style). This prevents padding overlap where words
near column borders (e.g. "We" at the start of example sentences) were
assigned to the preceding column. The last column extends generously to
capture all rightmost text.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

2026-03-02 12:53:56 +01:00

168 KiB

Raw Blame History

View Raw