Description
Extracting content from scanned papers and reports, OCR reads text but can't tell titles, tables and body apart. Layout Parser is a unified deep learning toolkit for document image analysis that detects layout in a few lines.
It ships pretrained models, layout data structures and visualization, and works with OCR engines for structured extraction.
Layout detection:Titles, tables and text.
Pretrained:Many document types.
OCR:Structured output.
Visualization:Check results.
It ships pretrained models, layout data structures and visualization, and works with OCR engines for structured extraction.
Features
Layout detection:Titles, tables and text.
Pretrained:Many document types.
OCR:Structured output.
Visualization:Check results.
