4 ms·
For bitmap based PDFs it would be possible to segment the document into words and images (just bounding boxes, not OCR), then "reflow" them to a different page
by visarga 9y ago
For bitmap based PDFs it would be possible to segment the document into words and images (just bounding boxes, not OCR), then "reflow" them to a different page size, by allocating less words per row.
Does anyone know if this kind of PDF reader exists? Such a PDF reflow reader would work on scanned old books.
- ocrcustomserver 9y agoI'm working on it, email me for early access.