some memo for the OCR technology
- 未加文字方向检测模块的模型 主要检测中文和英文以及一些符号https://github.com/YCG09/chinese_ocr
- 具有文字方向检测模块的模型 都是大神啊https://github.com/jiangxiluning/chinese-ocr
- C++的https://github.com/jiangxiluning/FOTS.PyTorch
- python的https://github.com/xieyufei1993/InceptText-Tensorflow
- Caffe的https://github.com/lvpengyuan/masktextspotter.caffe2#datasets
- https://github.com/AKSHAYUBHAT/DeepVideoAnalytics/tree/master/docs/experiments/ocr
- https://github.com/Belval/CRNN/tree/master/CRNN
- https://github.com/bear63/sceneReco
- 一个将densenet和crnn做了对比的项目
- https://github.com/diaomin/crnn-mxnet-chinese-text-recognition
- https://github.com/Bartzi/see
- https://github.com/MaybeShewill-CV/CRNN_Tensorflow
- https://github.com/oyxhust/CNN-LSTM-CTC-text-recognition
- https://github.com/xieyufei1993/InceptText-Tensorflow
- https://github.com/jiangxiluning/FOTS.PyTorch
- https://github.com/lvpengyuan/masktextspotter.caffe2
https://github.com/tangzhenyu/Scene-Text-Understanding
- 清华的自然场景数据集
- 某网络数据集, 300w条字符, https://pan.baidu.com/s/1QkI7kjah8SPHwOQ40rS1Pw密码:lu7m
- 利用图片和文字, 合成自然场景下的文本, 40G, 800w个单词