extract-paper-images

To extract figures from an arXiv paper, provide the arXiv ID to fetch the source package and pull true figures from directories like pics, figures, or images. This method avoids PDF logos and decorative images by prioritizing original source files.

Yes, you can extract figures from a local PDF file. The tool uses PyMuPDF for PDF-based extraction as a fallback when arXiv source packages are unavailable, ensuring you still get the images needed for your research notes.

Extracted paper images are organized into a structured directory under 20_Research/Papers/[field]/[paper title]/images. An index.md file is also generated to provide notes-friendly links and preserve image sources like arxiv-source or pdf-figure.

Extracting images from arXiv source packages yields higher quality figures because it retrieves the original true figures directly. This approach bypasses PDF compression artifacts and filters out decorative images or logos typically embedded in compiled PDFs.

The image extraction process requires Python with the PyMuPDF and requests libraries. PyMuPDF handles PDF-based image extraction, while requests manages fetching arXiv source packages when using an arXiv ID as input.