teresasun Goto Github PK

aggregation-cross-entropy

Aggregation Cross-Entropy for Sequence Recognition. CVPR 2019.

aster

Recognizing cropped text in natural images.

attentionocr

Scene text recognition

awesome-ocr

A curated list of promising OCR resources

awesome-synthtext

A curated list of awesome synthetic data for text location and recognition

bezier_curve_text_spotting

A PyTorch implementation of "ABCNet: Real-time Scene Text Spotting with Adaptive Bezier-Curve Network" (CVPR 2020 oral)

binarynet-caffe2

Quantized Neural Network on Caffe2

caffe-mobilenet-v3

Caffe Implementation of MobileNets V3

caffe-mobilenetv2-relu6

Caffe implementation of ReLU6 Layer

caffe-ocr-linux

caffe-ocr-linux

caffe-quantization

Added quantization layer into caffe (support a coarse level fixed point simulation)

caffe_ocr

主流ocr算法研究实验性的项目，目前实现了CNN+BLSTM+CTC架构

caffe_ocr_ctpn_ctc

端对端文字识别项目

caffe_ocr_for_linux

caffe_ssd_lstm_lpr

基于SSD+Resnet+CTC的中文车牌检测识别

caffelstm-ocr

基于caffe的LSTM OCR案例，能够利用该案例完成序列的识别，包括验证码、车牌、身份证号码、地址等长序列动长的内容识别

cascaded_mobilenet-v2

cascaded convolutional neural network for facial point detection

ccpd

[ECCV 2018] CCPD: a diverse and well-annotated dataset for license plate detection and recognition

centerloss_caffe_mnist

It's the script of Center loss on mnist dataset running on Caffe.

chinese-poetry

最全中华古诗词数据库, 唐宋两朝近一万四千古诗人, 接近5.5万首唐诗加26万宋诗. 两宋时期1564位词人，21050首词。

chinese_ocr

中文ocr识别

chineseocr

yolo3+ocr

chineseocr_lite

超轻量级中文ocr，支持竖排文字识别, 支持ncnn推理 , psenet(8.5M) + crnn(6.3M) + anglenet(1.5M) 总模型仅17M

classic-convolution-network

里面会保存许多优秀的卷积神经网络结构，这些结构可以帮助我们更好的设计网络。

crnn

A TensorFlow implementation of https://github.com/bgshih/crnn

crnn-1

Convolutional Recurrent Neural Network (CRNN) for image-based sequence recognition.

crnn-attention

crnn feature net + attention 机制

crnn.caffe

crnn.caffe

crnn_chinese_characters_rec

chinese characters recognition

ctcdecoder

This version of CTC decoder is similar to that of Deep Speech implementation of CTC decoding with added feature of capturing timestamps for every word