Observed arrival · 2026-09-03
Image to Text: offline OCR that keeps tables intact
A Chrome extension and browser demo that extracts text from screen regions locally, with table output that can be pasted into spreadsheets.
Field notes
The page describes a screen-capture workflow built around Ctrl+Shift+S, with alternate capture paths through right-click, paste, and file drop. Its table mode uses the spatial position of recognized text to reconstruct spreadsheet structure, and the example invoice exposes item, quantity, unit price, amount, and total fields. The homepage also states that the approximately 6 MB model loads only when the demo starts and that extra language packs are downloaded separately.
Observed signals
Read the marks
Editorial observations of this landing page, not a rating.
One card from the complete issue