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

Powerful handwritten text recognition. A simple-to-use, unofficial implementation of the paper "TrOCR: Transformer-based Optical Character Recognition with Pre-trained Models".

An image to text model base on transformer which can also be used on OCR task.

AutoRegressive Transformer Model for Optical Character Recognition In Farasi

本项目是一个结合了深度学习与自动化流程的工具，用于处理极验（Geetest）v4 的文字点选验证码。它使用 **YOLOv8** 模型进行目标检测，并集成了 TrOCR 和 PaddleOCR 模型进行文字识别，提供了灵活的配置和强大的混合匹配策略。

OCR-D processor for TrOCR

A service for extracting and indexing archival document images

This is a handwritten-equations to latex-equations converter, built using trocrbasestage1 model + CROHME dataset + custom-built SentencePiece Tokenizer (latex)

TL;DR: captcha image reconition api server

I integrated the TrOCR model with the Django framework

This is a public dataset for training OCR models on using the British Parliament's historic records. It is for digital humanities researchers

A basic Python script to create either a TXT annotation file used to generate a TFRecords dataset, or a CSV file to use it directly as a Filename/label dataset's annotation file.

This script allows to run TrOCR transriptions on any images with a page xml with layout recognized. It follows a 3 step workflow of cropping all the lines, then running trocr on them and then recombining the results of step 2 to generate a new .xml as output of the recognized lines.