paper-image-extractor

You can extract original figures by prioritizing arXiv source packages to locate and copy high-quality image files from directories like pics/ or figures/. This approach preserves the original image quality directly from the author's files.

Yes, you can extract images from a research paper PDF using a reliable fallback mechanism. When arXiv source packages are unavailable, the extraction workflow falls back to parsing the PDF to retrieve embedded figures.

Yes, extracting figures from arXiv sources requires Python 3.8+ and the PyMuPDF library. The deterministic scripts also require network access to arXiv to download source packages and perform PDF fallback extraction.

The best way to ensure traceability is to generate an index.md file containing image metadata and source labels. This output indexing maps each extracted figure to its origin, whether from arXiv source directories or PDF fallback.

The arXiv source extraction checks common image directories within the source package, specifically targeting pics/, figures/, fig, images/, and img. Locating these directories allows the script to copy high-quality original images.