A tensorflow siamese network implementation. Illustrated using singature recognition/identification.

该脚本根据语料文件生成对应的图像文件，适用于文本识别等CV任务

A repository with anonymized invoices

~1000 book pages + OpenCV + python = page regions identified as paragraphs, lines, images, captions, etc.

Creates synthetic degraded image documents that could be used to train Neural Networks

A synthetic data generator for text recognition

Online-handwritten version of the George Washington Dataset.

Tools necessary to perform a multi-fold pretrained voting approach utlizing OCRopus.

A selection of test lines of several early printed books as well as the corresponding individual OCRopus models and mixed models.

Code and procdures for handwriting object detection and recognition

ScrabbleGAN: Semi-Supervised Varying Length Handwritten Text Generation (CVPR20)

TDF-ICDAR 2019 Dataset for Typeset Math Formula Detection

Dataset for scene text removal

This Web application crawls PDFs from governement websites, performs table detection and displays advanced statistics.

Generate text images for training deep learning ocr model

Total Text Dataset - ICDAR 2017. It consists of 1555 images with more than 3 different text orientations: Horizontal, Multi-Oriented, and Curved, one of a kind.

A tensorflow reproducing of paper “Editing Text in the wild”

CORD: A Consolidated Receipt Dataset for Post-OCR Parsing

Distorted Document Images dataset (DDI-100).