aniketdata

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

Ground truth line annotations for the Berliner Börsen-Zeitung

A synthetic data generator for text recognition

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

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

Online-handwritten version of the George Washington Dataset.

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

Distorted Document Images dataset (DDI-100).

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

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

Code and procdures for handwriting object detection and recognition

A repository with anonymized invoices

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.

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

Generate text images for training deep learning ocr model

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

CORD: A Consolidated Receipt Dataset for Post-OCR Parsing

EATEN: Entity-aware Attention for Single Shot Visual Text Extraction