manuscript-ocr

Contents:

  • Getting Started
    • Installation
      • Basic installation (inference only)
      • GPU acceleration (NVIDIA CUDA)
    • Quick Start
    • Example Notebooks
    • Additional installation options
      • Apple Silicon acceleration (CoreML)
      • Installation with training support
    • Minimum System Requirements
  • Pipeline Usage Guide
    • Quick start
    • Choosing stages
    • Combining processors
      • Stack
      • Recognizer voting
      • Detector consensus
    • Model memory
    • TrOCR CUDA decoding
    • Recognition crops and batch memory
    • Custom stages
    • Recognition and correction levels
    • Results, visualization and timing
  • Model Zoo
    • Detection
    • Layout
    • Recognition
    • Correction
  • Library Structure
    • Modules
    • Public entry points
  • API Reference
    • Data Structures
      • Document tree
      • Elements and fields
      • Tables
      • Copying and serialization
        • COCO files
      • Text output
      • Custom elements
        • Element
        • Page
        • TextRegion
        • Table
        • Cell
        • Image
        • to_coco()
        • from_coco()
    • Pipeline and Stack
      • Pipeline
        • Pipeline
      • Stack
        • Stack
    • Detectors
      • ConsensusWBFDetector
        • ConsensusWBFDetector
      • YOLO
        • YOLO
      • EAST
        • EAST
      • PPOCR
        • PPOCR
      • RFDETR
        • RFDETR
      • Mask2Former
        • Mask2Former
    • Layouts
      • SimpleSorting
        • SimpleSorting
      • ReadingSorting
        • ReadingSorting
    • Recognizers
      • TRBA
        • TRBA
        • TRBA.predict()
      • PPOCRRec
        • PPOCRRec
        • PPOCRRec.predict()
      • TrOCR
        • TrOCR
        • TrOCR.predict()
      • MajorityVotingRecognizer
        • MajorityVotingRecognizer
      • MedoidRecognizer
        • MedoidRecognizer
        • MedoidRecognizer.predict()
    • Correctors
      • CharLM
        • CharLM.predict()
        • CharLM.train()
        • CharLM.export()
    • Utilities
      • Images and test pages
        • read_image()
        • create_page_from_image()
        • create_page_from_text()
      • Grouping and ordering
        • group_regions()
        • organize_page()
      • Collapsing text regions
        • merge_text_spans()
        • collapse_line_text_spans()
        • collapse_block_text_spans()
        • collapse_page_text_spans()
      • Geometry
        • polygon_to_bbox()
        • crop_axis_aligned()
        • crop_polygon_mask()
        • order_quad_points()
        • warp_quad()
        • merge_polygons()
      • Visualization
        • visualize_page()
      • Training helper
        • set_seed()
  • Related Work
  • Migrating from 0.1.12 to 0.1.13
    • Elements and reading order
    • Saved results
  • Release Notes
    • 0.1.13
    • 0.1.12
    • 0.1.11
    • 0.1.10
  • Authors and Acknowledgements
    • Authors
    • Acknowledgements
manuscript-ocr
  • Overview: module code

All modules for which code is available

  • manuscript._pipeline
  • manuscript._stack
  • manuscript.correctors._charlm
  • manuscript.data.tree
  • manuscript.detectors._consensus
  • manuscript.detectors._east
  • manuscript.detectors._mask2former
  • manuscript.detectors._ppocr
  • manuscript.detectors._rfdetr
  • manuscript.detectors._yolo
  • manuscript.layouts._readingsorting
  • manuscript.layouts._simplesorting
  • manuscript.recognizers._majority_voting
  • manuscript.recognizers._medoid
  • manuscript.recognizers._ppocr_rec
  • manuscript.recognizers._trba
  • manuscript.recognizers._trocr
  • manuscript.utils.geometry
  • manuscript.utils.grouping
  • manuscript.utils.io
  • manuscript.utils.page_transforms
  • manuscript.utils.sorting
  • manuscript.utils.training
  • manuscript.utils.visualization

© Copyright 2026, Konstantin Kozhin.

Built with Sphinx using a theme provided by Read the Docs.