Konstantin Baierer

g-ocr. GOCR - State-of-the-Art OCR foundation for German documents. Fast, CPU, no GPU.

Mongoku. 🔥The Web-scale GUI for MongoDB

chipotlai-max. The AI coding agent that runs on stolen Chipotle compute 🌯 Fork of OpenCode with Pepper AI as default model. Community project to add providers from Home Depot, Lowes, Target, Starbucks & more.

textbite-dataset. TextBite: A Historical Czech Document Dataset for Logical Page Segmentation

dfine_kraken. D-FINE for document region segmentation

sbb_images. Image Annotation Tool and Image Search

ubma-segmentation-ocr-model. This repository contains a segmentation model for historical and modern prints.

chr2025-works-on-my-machine. Research artefact for the paper ‘“Works on My Machine”: A Case Study of Replicability Challenges in Computational Humanities Research’ at CHR 2025

UVDoc. Code for the paper "UVDoc: Neural Grid-based Document Unwarping"

datasette. Tutorial für die Nutzung der von IDM 4 bereitgestellten Datasette-Instanz

pagexml-mets-viewer. Web app to upload and display multiple PageXML files

awesome-ocr.

RustPython. A Python Interpreter written in Rust

ocrd_paddleocr. OCR-D integration for PaddleOCR

tch-rs. Rust bindings for the C++ api of PyTorch.

DeepSeek-OCR. Contexts Optical Compression

ocrd_party. OCR-D processor for the party text recognizer

ocrd_yolo. OCR-D wrapper for yolo based on the ocrd_detectron2 wrapper

sbb_binarizer_pytorch_converter. Python

twisted. Event-driven networking engine written in Python.

Ground-Truth-CHR2024. Coordinates of manually annotated job ads with a link to ANNO Corpus.

nicegui. Create web-based user interfaces with Python. The nice way.

inventory-card-reader. This repository contains code to read, process, and integrate data from inventory cards.

Ge-ez-HWR. Python

scribeocr. Web interface for recognizing text, proofreading OCR, and creating fully-digitized documents.

HTRMoPo. Schemas for repositories of HTR/OCR models

party. Page-wise text recognition with lower-supervision line data models

ocapy. OCR Confidence Analysis script written in python

tooltuesday. Jupyter Notebook

elementpath. XPath 1.0/2.0/3.0/3.1 parsers and selectors for ElementTree and lxml

templating-kubernetes. Templating Kubernetes resources with *real* code

flynt. A tool to automatically convert old string literal formatting to f-strings

sbb_page_extractor. Python

HookFormer. Contextual HookFormer for Glacier Calving Front Segmentation (DOI: 10.1109/TGRS.2024.3368215)

loghi. Shell

EasyOCR. Ready-to-use OCR with 80+ supported languages and all popular writing scripts including Latin, Chinese, Arabic, Devanagari, Cyrillic and etc.

PagePlus. This script processes PAGE XML files, a format widely used in document layout analysis, to perform various operations like validating, repairing, extending, and modifying text regions and lines.

loghi. Loghi is a comprehensive toolkit designed for Handwritten Text Recognition (HTR) and Optical Character Recognition (OCR), offering an accessible approach to transcribing historical documents and training models for specialized needs.

laypa. Layout analysis to find layout elements in documents (similar to P2PaLA)

alto2txt. Convert ALTO XML to plain text + minimal metadata

sbb_pixelwise_segmentation. Obsolete repo, merged into eynollah

types-lxml. Complete lxml external type annotation

dta-lexdb-applications. formatting and integrating the Deutches Textarchiv dictionary into various applications

ulb-groundtruth-eval-odem-other. OCR Groundtruth ULB VD18 - OCR-D Phase III

ulb-groundtruth-eval-odem-lat. OCR Groundtruth ULB VD18 Latin - OCR-D Phase III

ulb-groundtruth-eval-odem-ger. OCR Grountruth ULB VD18 German Fraktur - OCR-D Phase III

asciimatics. A cross platform package to do curses-like operations, plus higher level APIs and widgets to create text UIs and ASCII art animations

textual-web. Run TUIs and terminals in your browser

notcurses. blingful character graphics/TUI library. definitely not curses.

flet. Build realtime web, mobile and desktop apps in Python only. No frontend experience required.

pytest-docker. Docker-based integration tests

madoc-platform. A platform for the display, enrichment, and curation of IIIF-based digital objects

metha. Command line OAI-PMH harvester and client with built-in cache.

ocrd_network_tests. Shell

terminal7. A Multi Platform Terminal Emulator & Multiplexer Running over WebRTC with Touch First UI

operandi_benchmarking. Benchmarking data for OCR-D - Operandi

gt-fraktur. Shell

tessdata_contrib. User contributed (non Google) OCR models for Tesseract

infocom-zcode-terps. Historical source code for Infocom's Z-machine interpreters

website_next_generation. The new website

ocropus4extract. Jupyter Notebook

ocropus4inf. Jupyter Notebook

ocropus4train. Jupyter Notebook

page2tei. A repository for illustrating the transformation of a PAGE XML file into XML-TEI format, resulting from experimentations made for the LECTAUREP project.

audiveris. Latest generation of Audiveris OMR engine

odem-ocr. Digitalization Workflows with OCR Backend

pylib. Personal python library

HTRVX. HTRVX : HTR Validation with XSD

citlab-article-separation-new. Modules used for separating articles in (historical) newspapers and similar documents. This repository is part of the European Union's Horizon 2020 project NewsEye. For more information about the project see https://www.newseye.eu/.

NodeFlow. An Editor for creating simple or complex OCR workflows

delft. a Deep Learning Framework for Text https://delft.readthedocs.io/

ocrd-network-setup. Deploy ocrd-processing-server in a VM

GT-commentaries-OCR. Ground truth data for the Optical Character Recognition of Historical Classical Commentaries.

gt_corpus_benchmark. This repo provides a collection of ground truth data. The collection was compiled under different aspects (complexity of the layouts and use of the fonts). The individual data are also characterized by metadata. The metadata is based on the labeling scheme of OCR-D/PrimaLab.

wikimedia-ocr. This repository is now at https://gitlab.wikimedia.org/toolforge-repos/ocr

gesetze. Bundesgesetze und -verordnungen

ocrd_monitor. Web frontend for ocrd_manager

quiver-benchmarks. Benchmarking OCR-D workflows in Docker

gt_structure_text. The OCR-D Ground Truth text and structure corpus was created between 2015 -2017. In the years since 2017, this corpus has been further curated and supplemented with metadata where appropriate. The corpus includes page XML files within annotations of the text and structure include.

ocrd_froc. Python

stabi-berlin-gt. Ground truth for digitized publications of Staatsbibliothek zu Berlin

image.textlinedetector. Segment Images in Text Lines and Words

requests-unixsocket. Use requests to talk HTTP via a UNIX domain socket

google-ocr-testbed. A repository containing scripts and output for Google OCR of sample files.

mean_average_precision. Mean Average Precision for Object Detection

policy. OCR-D Empfehlungen Volltextdigitalisierung

quiver-data. Providing the Ground Truth workspaces for the OCR-D QUIVER application

choco-mufin. Tools for normalizing the use of some characters and checking file consistencies

choco-mufin. Tools for normalizing the use of some characters and checking file consistencies

pageattrlib. XSLT Library to work with PAGE XML custom attributes

docstruct. Document structure detection from PAGE-XML to METS-XML

awk. One true awk

Mienai.ttf. Mienai (見えない) is a "last-resort" font that does the same job as Adobe Blank in a smaller package, at the cost of some compatibility.

operandi. Project repository for the OCR-D Implementation Project OPERANDI.

digi-gt. Ground truth for the digitized historic collections of UB Mannheim

fix-perspective. C++

OtoN_Converter. Converter from basic OCRD process workflow to Nextflow workflow script

ocr-util. Evaluate data from mass digitalization workflows

digi-gt. Ground truth for the digitized historic collections of UB Mannheim

DIVA-DAF. Repository for the deep-learning framework DIVA-DAF which is build with historical document image analysis in mind.

curt. Python

nmalign. forced alignment of lists of string by fuzzy string matching

ocrd_kitodo. Docker integration of Kitodo.Production and OCR-D

frat. Fast Rectangle Annotation Tool

sbb_binarize_flow_from_directory. Python

ocrd-webapi-implementation. Python

ddblabs-iiimets. IIIF to METS/MODS conversion script

ocr-pipeline. OCR Pipeline Module for Digitalization Workflows

ocrd_manager. frontend for ocrd_controller and adapter towards ocrd_kitodo

vim-outlaw. The wanted outliner!

diffgram. The AI Datastore for Schemas, BLOBs, and Predictions. Use with your apps or integrate built-in Human Supervision, Data Workflow, and UI Catalog to get the most value out of your AI Data.

ocrd_controller. Path to network implementation of OCR-D

ocrd_page2tei. OCR-D wrapper for page2tei

kitodo-dev-docker. Shell

DeslantImg. The deslanting algorithm sets text upright in images. Python, C++ and OpenCL implementations provided.

OPERANDI_TestRepo. This is the OPERANDI project's test repository.

sbb_column_classifier. Get the number of columns for a document image

ocrd_detectron2. OCR-D wrapper for detectron2 based segmentation models

OCR-Metrics-CER-WER. Sample implementation of OCR metrics (CER, WER) calculation with TesseractOCR and fastwer

OCR17plus. Data for layout analysis and HTR.

foliautils. Command-line utilities for working with the Format for Linguistic Annotation (FoLiA), powered by libfolia (C++), written by Ko van der Sloot (CLST, Radboud University)

ticcltools. Tools for TICCL

PICCL. A set of workflows for corpus building through OCR, post-correction and normalisation

qt-agi-studio. AGI (Adventure Game Interpreter) is the adventure game engine used by Sierra On-Line(tm) to create some of their early games. QT AGI Studio (formerly known as Linux AGI Studio) is a program which allows you to view, create and edit AGI games. Basically, it is an enhanced port of the Windows AGI Studio developed by Peter Kelly.

nagi. New Adventure Game Interpreter

LibreLingo. 🐢 🌎 📚 a community-owned language-learning platform

qutebrowser. A keyboard-driven, vim-like browser based on Python and Qt.

mets-model. Simple METS Api

click-repl. Subcommand REPL for click apps

pymux. A terminal multiplexer (like tmux) in Python

ptpython. A better Python REPL

libtmux. ⚙️ Python API / wrapper for tmux

tmuxp-config. Configs for tmuxp (https://github.com/tony/tmuxp)

tmuxp. 🖥️ Session manager for tmux, built on libtmux.

flusswerk. Easily create AMQP/RabbitMQ based workflows.

parcel. 📦🚀 Blazing fast, zero configuration web application bundler

mocri. Small gRPC microservice for OCR based on kraken

teihub. Automated listing of repos in GitHub with XML files containing teiHeader. Find a project using TEI today!

unilm. Large-scale Self-supervised Pre-training Across Tasks, Languages, and Modalities

mets2iiif. An implementation of the IIIF Presentation API v2 based on XSLT

table-transformer. Table Transformer (TATR) is a deep learning model for extracting tables from unstructured documents (PDFs and images). This is also the official repository for the PubTables-1M dataset and GriTS evaluation metric.

ulb-zeitungsprojekt-hp1. Training data from "Hauptphase I" of project "Digitalisierung historischer deutscher Zeitungen"

metaindex. Official Site: www.metaindex.fr // Cataloger application: HTML5 client // J2EE server // integrated interface with Kibana for statistics.

im2alto. ImageWare MyBib created OCR Output to ALTO XML XSL Transformation

archive-hocr-tools. Efficient hOCR tooling

HandwritingRecognitionSystem. Handwriting Recognition System based on a deep Convolutional Recurrent Neural Network architecture

tesseract-ocr-for-php. A wrapper to work with Tesseract OCR inside PHP.

modern-unix. A collection of modern/faster/saner alternatives to common unix commands.

dhh-text-2021. Slides for the workshop Digital Herrnhut in textwissenschaftlichen Kontexten

ocrd_typegroups_classifier. Font family detection in historical documents.

fargv. fast_argv

browse-ocrd-physical-import. A plugin for browse-ocrd to scan book pages with an android phone camera

hip21_ocrevaluation. A Survey of OCR Evaluation Tools and Metrics (HIP'21)

nautilusocr. METS/ALTO OCR enhancing tool by the National Library of Luxembourg (BnL)

DOOM-FX. Doom/FX for Super Nintendo with SuperFX GSU2A

Junicode-font. A new version of Junicode font

WA-map. HTML

htr-united. Ground Truth Resources for the HTR of patrimonial documents

dahncorpus. Ground Truth dataset for French 20th typewritten OCR produced by the DAHN project

Issues. For interactive problem-solving, discussion, help, etc

fantasy. A curated list of available fantasy consoles/computers.

PAGETools. Small collection of PAGE XML related scripts used at the ZPD Würzburg

ExpReal. An expressive realizer for interactive narratives

ocrd_contrib_ubma. Helper scripts for OCR-D

List-of-Dirty-Naughty-Obscene-and-Otherwise-Bad-Words. List of Dirty, Naughty, Obscene, and Otherwise Bad Words

ocr-eng-bio-testfiles. OCR English (Bio, Natur) ground truth and testfiles

Mask_RCNN. Mask R-CNN for object detection and instance segmentation on Keras and TensorFlow

sfst. Stuttgart Finite State Transducer system

doctr. docTR (Document Text Recognition) - a seamless, high-performing & accessible library for OCR-related tasks powered by Deep Learning. Ongoing development and maintenance by t2k.

acid-banger. The Endless Acid Banger

cibuildwheel. 🎡 Build Python wheels for all the platforms with minimal configuration.

splitflap. DIY split-flap display

rhasspy. Offline private voice assistant for many human languages

leptseg. Python

tesseract_jpn-vert. Python

page2tsv. PAGE-XML to TSV

Assemblies-of-putative-SARS-CoV2-spike-encoding-mRNA-sequences-for-vaccines-BNT-162b2-and-mRNA-1273. RNA vaccines have become a key tool in moving forward through the challenges raised both in the current pandemic and in numerous other public health and medical challenges. With the rollout of vaccines for COVID-19, these synthetic mRNAs have become broadly distributed RNA species in numerous human populations. Despite their ubiquity, sequences are not always available for such RNAs. Standard methods facilitate such sequencing. In this note, we provide experimental sequence information for the RNA components of the initial Moderna (https://pubmed.ncbi.nlm.nih.gov/32756549/) and Pfizer/BioNTech (https://pubmed.ncbi.nlm.nih.gov/33301246/) COVID-19 vaccines, allowing a working assembly of the former and a confirmation of previously reported sequence information for the latter RNA. Sharing of sequence information for broadly used therapeutics has the benefit of allowing any researchers or clinicians using sequencing approaches to rapidly identify such sequences as therapeutic-derived rather than host or infectious in origin. For this work, RNAs were obtained as discards from the small portions of vaccine doses that remained in vials after immunization; such portions would have been required to be otherwise discarded and were analyzed under FDA authorization for research use. To obtain the small amounts of RNA needed for characterization, vaccine remnants were phenol-chloroform extracted using TRIzol Reagent (Invitrogen), with intactness assessed by Agilent 2100 Bioanalyzer before and after extraction. Although our analysis mainly focused on RNAs obtained as soon as possible following discard, we also analyzed samples which had been refrigerated (~4 ℃) for up to 42 days with and without the addition of EDTA. Interestingly a substantial fraction of the RNA remained intact in these preparations. We note that the formulation of the vaccines includes numerous key chemical components which are quite possibly unstable under these conditions-- so these data certainly do not suggest that the vaccine as a biological agent is stable. But it is of interest that chemical stability of RNA itself is not sufficient to preclude eventual development of vaccines with a much less involved cold-chain storage and transportation. For further analysis, the initial RNAs were fragmented by heating to 94℃, primed with a random hexamer-tailed adaptor, amplified through a template-switch protocol (Takara SMARTerer Stranded RNA-seq kit), and sequenced using a MiSeq instrument (Illumina) with paired end 78-per end sequencing. As a reference material in specific assays, we included RNA of known concentration and sequence (from bacteriophage MS2). From these data, we obtained partial information on strandedness and a set of segments that could be used for assembly. This was particularly useful for the Moderna vaccine, for which the original vaccine RNA sequence was not available at the time our study was carried out. Contigs encoding full-length spikes were assembled from the Moderna and Pfizer datasets. The Pfizer/BioNTech data [Figure 1] verified the reported sequence for that vaccine (https://berthub.eu/articles/posts/reverse-engineering-source-code-of-the-biontech-pfizer-vaccine/), while the Moderna sequence [Figure 2] could not be checked against a published reference. RNA preparations lacking dsRNA are desirable in generating vaccine formulations as these will minimize an otherwise dramatic biological (and nonspecific) response that vertebrates have to double stranded character in RNA (https://www.nature.com/articles/nrd.2017.243). In the sequence data that we analyzed, we found that the vast majority of reads were from the expected sense strand. In addition, the minority of antisense reads appeared different from sense reads in lacking the characteristic extensions expected from the template switching protocol. Examining only the reads with an evident template switch (as an indicator for strand-of-origin), we observed that both vaccines overwhelmingly yielded sense reads (>99.99%). Independent sequencing assays and other experimental measurements are ongoing and will be needed to determine whether this template-switched sense read fraction in the SmarterSeq protocol indeed represents the actual dsRNA content in the original material. This work provides an initial assessment of two RNAs that are now a part of the human ecosystem and that are likely to appear in numerous other high throughput RNA-seq studies in which a fraction of the individuals may have previously been vaccinated. ProtoAcknowledgements: Thanks to our colleagues for help and suggestions (Nimit Jain, Emily Greenwald, Lamia Wahba, William Wang, Amisha Kumar, Sameer Sundrani, David Lipman, Bijoyita Roy). Figure 1: Spike-encoding contig assembled from BioNTech/Pfizer BNT-162b2 vaccine. Although the full coding region is included, the nature of the methodology used for sequencing and assembly is such that the assembled contig could lack some sequence from the ends of the RNA. Within the assembled sequence, this hypothetical sequence shows a perfect match to the corresponding sequence from documents available online derived from manufacturer communications with the World Health Organization [as reported by https://berthub.eu/articles/posts/reverse-engineering-source-code-of-the-biontech-pfizer-vaccine/]. The 5’ end for the assembly matches the start site noted in these documents, while the read-based assembly lacks an interrupted polyA tail (A30(GCATATGACT)A70) that is expected to be present in the mRNA.

sbb_page_extractor. Python

police-settlements. A FiveThirtyEight/The Marshall Project effort to collect comprehensive data on police misconduct settlements from 2010-19.

memory_profiler. Monitor Memory usage of Python code

blitz. Blitz++ Multi-Dimensional Array Library for C++

GTCheck. Check your modified Ground Truth files with visual support!

kill-the-newsletter. Convert email newsletters into Atom feeds

8VIM. A Text Editor inside a keyboard, drawing it's inspiration from 8pen and Vim.

glyphcollector. Elm

BitmapFonts. My collection of bitmap fonts pulled from various demoscene archives over the years

Ocrd_anybaseocr_block_segmentation. Jupyter Notebook

page-xml-to-image. Python

blitzDrt. Tool to correct perspective distortion. Does not correct verticals (column separators, table lines) yet. Uses Blitz++ library.

pawls. Software that makes labeling PDFs easy.

Pipeline. pipeline for xslt and python plugins

docExtractor. (ICFHR 2020 oral) Code for "docExtractor: An off-the-shelf historical document element extraction" paper

ocr-data. Jupyter Notebook

pytesstrain. Python tools for Tesseract OCR training

kraken_generated-data. Python

OCR_GS_Data. Double-checked Gold Standard Data for Training and Testing OCR Engines

tesstrainsh-win. Train Tesseract LSTM with tesstrain.sh on Windows

vim-visual-multi. Multiple cursors plugin for vim/neovim

sbb_ocr_postcorrection. Two-Step Approach to OCR Post-Correction

page2tei. Python snippets that might be useful for exporting transcribed pages from PAGE XML to TEI XML