extract-info-from-pdf-paper

This Python script uses pdfminer.six, PyPDF2, pdf2image to extract information (text, image) from pdf paper.

Install dependencies

pip3 install -r requirement.txt

How to Run

python3 extract-text-image.py

Check out output folder for the result of extraction of this script.

Note

Change the input and output file path in the Python script.

pdf_path = "2305.02301.pdf"
out_text_path = "output/2305.02301.txt" 

TODO: extract latex

• 1. extract latex formula image
• 2. use LaTeX-OCR for convert latex formula image to latex
• 3. replace the plaintext formula in result with $$ latex $$

TODO: extract table

Windows Usage

1. Download the Windows version of poppler:https://github.com/oschwartz10612/poppler-windows/releases
2. After decompression, move the "poppler-23.11.0" subfolder to C:\Program Files.
3. Add environment variable: "C:\Program Files\poppler-23.11.0\Library\bin", save and exit
4. run