Convert documents to structured data effortlessly. Unstructured is open-source ETL solution for transforming complex documents into clean, structured formats for language models. Visit our website to learn more about our enterprise grade Platform product for production grade workflows, partitioning, enrichments, chunking and embedding.

nlp pdf machine-learning natural-language-processing information-retrieval ocr deep-learning ml docx preprocessing pdf-to-text data-pipelines donut document-image-processing document-parser pdf-to-json document-image-analysis llm document-parsing langchain

• Updated Nov 6, 2026
• HTML

A Repo For Document AI

python nlp ocr tensorflow pytorch document-parser document-layout-analysis table-recognition table-detection document-understanding publaynet layoutlm document-ai document-image-analysis pubtabnet

• Updated Oct 21, 2026
• Python

ExtractThinker is a Document Intelligence library for LLMs, offering ORM-style interaction for flexible and powerful document workflows.

python nlp pdf machine-learning ocr ai openai pdf-to-text document-processing document-image-analysis document-intelligence llm document-parsing langchain

• Updated Sep 16, 2026
• Python

Detectron2 for Document Layout Analysis

python computer-vision deep-learning pytorch neural-networks segmentation object-detection text-detection semantic-segmentation document-image-processing document-layout maskrcnn mask-rcnn dla document-layout-analysis detectron2 publaynet document-image-analysis

• Updated Aug 2, 2024
• Python

image-processing image-segmentation docstrum document-image-analysis

• Updated Apr 3, 2018
• Jupyter Notebook

[Late Submission] Solution for Kuzushiji recognition (Kaggle competition)

kaggle resnet nom unet document-analysis tensorflow-serving viblo tf-serving kuzushiji kuzushiji-recognition nom-document document-image-analysis

• Updated Jun 9, 2021
• Python

Visual Domain Knowledge-based Multimodal Zoning Textual Region Localization in Noisy Historical Document Images

tensorflow image-processing image-segmentation document-image-analysis

• Updated Jan 24, 2022
• C++

SegClarity: An attribution-based XAI workflow for layer-wise interpretability in semantic segmentation

deep-learning segmentation xai document-image-analysis historical-document-analysis

• Updated Nov 29, 2025
• Jupyter Notebook

Open source libraries and APIs to build custom preprocessing pipelines for labeling, training, or production machine learning pipelines.

nlp pdf machine-learning natural-language-processing information-retrieval ocr deep-learning ml docx preprocessing pdf-to-text data-pipelines donut document-image-processing pdf-to-json document-ai document-image-analysis document-parsing langchain

• Updated Mar 3, 2023
• HTML

Extracting structured text from GI Bill index cards for JDoc 2023 paper

administrative-data layout-parser document-image-analysis

• Updated May 17, 2023
• Jupyter Notebook

A fully managed C# implementation of Leptonica 1.87.0: image processing and document image analysis for .NET, no native dependencies

png ocr csharp dotnet gpu jpeg tiff morphology image-processing avalonia leptonica binarization deskew document-image-analysis

• Updated Nov 6, 2026
• C#

Analyze document image complexity based on segmentation results

image-segmentation document-image-analysis

• Updated Jan 26, 2022
• Python

📦 Split buffers and streams into smaller chunks for smooth HTTP uploads and accurate progress tracking.

java nlp pdf minecraft information-retrieval etl analytics xml pipelines sax data-pipelines donut document-image-processing pdf-to-json enterprise-integration document-image-analysis llm langchain

• Updated Oct 14, 2026
• JavaScript

Document image enhancement using the DIBCO dataset. Applies preprocessing and binarization techniques to restore degraded handwritten and printed documents for better readability and analysis.

python opencv computer-vision jupyter-notebook image-processing binarization dibco document-image-analysis

• Updated May 25, 2026
• Jupyter Notebook

Optical m(a)usic recognition without a trained classifier: directional mathematical morphology and graph rewriting transcribe printed scores to MIDI, with a C++20 port 6.1x faster than the Python reference.

python midi mathematical-morphology optical-music-recognition graph-rewriting cpp20 document-image-analysis staff-line-removal deepscores sauvola-binarization

• Updated Oct 29, 2026
• C++

jupyter-notebook voronoi-tessellation document-image-analysis

• Updated Apr 29, 2019
• Jupyter Notebook

Find the rotation angle (−180° to +180°) of scanned or photographed document pages, with a review flag for uncertain pages. OCR pre-processing, Python, ONNX, runs on CPU.

python opencv ocr computer-vision scanned-documents deskew onnx image-rotation skew-correction onnxruntime document-ai document-image-analysis ocr-preprocessing document-orientation page-orientation

• Updated Nov 3, 2026
• Python

Open-source framework for reproducible historical-document research, document analysis, calibration, transcription, translation, indexing, and citation.

python open-source opencv machine-learning ocr computer-vision reproducible-research image-processing genealogy digital-humanities document-analysis handwritten-text-recognition layout-analysis historical-documents archival-research research-software page-segmentation document-image-analysis hidden-texas-history

• Updated Oct 12, 2026
• Python

DocForge-Bench (arXiv:2603.01433): zero-shot benchmark of 14 document-forgery detectors on 8 datasets. Detectors that rank tampered pixels well (high Pixel-AUC) still score near-zero Pixel-F1 at τ=0.5: a calibration gap. Evaluation code, calibration analysis, pinned method setups and results.

benchmark computer-vision pytorch calibration image-forensics zero-shot image-manipulation-detection document-image-analysis document-forgery-detection tampering-localization text-tampering

• Updated Nov 1, 2026
• Python

tensorflow image-segmentation mask-rcnn document-image-analysis

• Updated Apr 29, 2019
• Jupyter Notebook