beacandler

EATEN

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

R2CNN

caffe re-implementation of R2CNN: Rotational Region CNN for Orientation Robust Scene Text Detection

awesome-cv-writing

该仓库包含了计算机视觉论文中写作常用的句式、短语、缩写等。

tf-distribute-demo

a demo for tensorflow distribute train

tf-slim-demo

Fine-tunes an InceptionV3 model on the Flowers training set.

awesome-ocr-projects

记录一些有趣的OCR工程：数据、模型等等

chinese-ocr

运用tensorflow实现自然场景文字检测,keras/pytorch实现crnn+ctc实现不定长中文OCR识别

Chinese_OCR_synthetic_data

The progress was used to generate synthetic dataset for Chinese OCR.

ChineseAddress_OCR

Photographing Chinese-Address OCR implemented using CTPN+CTC+Address Correction. 拍照文档中文地址文字识别。

CRNN_Tensorflow

Convolutional Recurrent Neural Networks(CRNN) for Scene Text Recognition

Detectron

FAIR's research platform for object detection research, implementing popular algorithms like Mask R-CNN and RetinaNet.

focal-loss

Focal loss for Dense Object Detection

insightface

Face Recognition Project on MXNet

ocr_attention

PlotNeuralNet

Latex code for making neural networks diagrams

RRPN

Source code of RRPN ---- Arbitrary-Oriented Scene Text Detection via Rotation Proposals

SynthText

Code for generating synthetic text images as described in "Synthetic Data for Text Localisation in Natural Images", Ankush Gupta, Andrea Vedaldi, Andrew Zisserman, CVPR 2016.

SynthText_Chinese_version

Modify from https://github.com/ankush-me/SynthText.git to generate chinese character

Tensorflow-Project-Template

A best practice for tensorflow project template architecture.

text-detection-ctpn

text detection mainly based on ctpn model in tensorflow, id card detect, connectionist text proposal network

TextBoxes_plusplus

TextBoxes++: A Single-Shot Oriented Scene Text Detector

TextProposals

Implementation of the method proposed in the papers " TextProposals: a Text-specific Selective Search Algorithm for Word Spotting in the Wild" and "Object Proposals for Text Extraction in the Wild" (Gomez & Karatzas), 2016 and 2015 respectively.