CRNN.pytorch

Introduction

This is a pytorch implementation of An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognition.
This implementation is similar to https://github.com/meijieru/crnn.pytorch, we rewrite it and make it easy to understand.
A GRCNN implementaion is in https://github.com/jingjing-you/GRCNN_OCR.pytorch.

Requirements

1. Pytorch >= 0.4.0
   
2. opencv
   
3. warp_ctc_pytorch
4. tqdm

Construct dataset

The names of train and val data are constructed as bellow:

img_xxx_$$$.jpg

where 'xxx' represents the number of this image and '$$$' represents the label of this image. For example，‘img_0_WHLU.jpg’.

Other examples are in data_sample directory.

Training the model

You can run python train.py to train your model.

Evaluation

You can run python eval.py to eval your model.

References

1. https://github.com/meijieru/crnn.pytorch
   
2. https://github.com/bgshih/crnn