This project will be used for DART pdf object detection task.

Clone this repo.

git clone https://github.com/snu-dm/DART_pdf_extractor.git
cd DART_pdf_extractor/

This code requires python 3+ and pdfplumber. Please Install dependencies by

pip install -r requirements.txt

Use following parsers in your Linux terminal:

• text extraction (-text)
• table extraction (-table)
• image extraction (-image)
• caption extraction (-caption)
• pdf_dir (-dir)
• save_dir (-save)
• cropped_file_only (-crop)
• total_page_with_segmentation (-page)

Example Usage:

python main.py -text -table -image -caption -crop -page