oomol-lab/pdf-craft

PDF craft can convert PDF files into various other formats. This project will focus on processing PDF files of scanned books.

What it solves

pdf-craft is a conversion library designed to turn PDFs—including scanned documents—into searchable, editable formats like Markdown or EPUB. It specifically targets books, academic papers, and technical documents, handling complex elements like tables of contents, footnotes, tables, formulas, and images.

How it works

The library uses an OCR (Optical Character Recognition) pipeline to convert page images into text. It supports multiple OCR backends, including DeepSeek OCR and Baidu's Unlimited OCR, which can be run either locally on an NVIDIA GPU (via CUDA) or through remote vendor services. For translation, it integrates with separate text LLMs to translate content during conversion or translate existing EPUBs. It can also perform "PDF to PDF" translation, where translated text is written back into the original PDF layout.

Who it’s for

Researchers, students, and developers who need to extract text from scanned PDFs or convert them into structured, editable formats for further processing or translation into other languages.

Highlights

  • Flexible OCR Backends: Supports both local GPU execution and remote API-based OCR.
  • Multi-format Output: Converts PDFs to Markdown, EPUB, or translated PDFs.
  • LLM-powered Translation: Integrates with text LLMs to translate content during the conversion process.
  • Complex Document Handling: Specifically optimized for academic and technical documents with formulas and tables.
  • EPUB Translation: Ability to translate existing EPUB files with options to replace original text or append translations.

Related

  • Project
    overcuriousity/pdf2epub

    A tool that converts PDF files into structured Markdown and EPUB formats using intelligent layout detection and AI-powered text extraction.

  • Project
    datalab-to/marker

    A high-accuracy document conversion tool that transforms PDFs and other files into structured Markdown, JSON, or HTML, specializing in tables and equations.

  • Project
    magicrew/doc7

    A document conversion tool that turns PDFs, Office files, and images into AI-ready Markdown using multimodal vision models, supporting local inference for privacy and cost-efficiency.

  • Project
    PaddlePaddle/PaddleOCR

    PaddleOCR is an open‑source OCR and document‑AI toolkit that converts images/PDFs into structured, LLM‑ready JSON or Markdown. It offers multilingual scene‑text models (PP‑OCRv6), a lightweight vision‑language parser (PaddleOCR‑VL‑1.6) for tables, formulas, and ancient scripts, and a structure‑aware pipeline (PP‑StructureV3). Models run on CPU, GPU, XPU, NPU or via ONNX/TensorRT, and integrate with RAG/agent platforms like Dify and RAGFlow.

  • Project
    scribeocr/scribeocr

    A browser-based web application for recognizing text from images, proofreading OCR data with high precision, and creating fully digitized, ebook-style documents.