Towards Accurate Scene Text Recognition with Semantic Reasoning Networks

Unofficial PyTorch implementation of the paper, which integrates not only global semantic reasoning module but also parallel visual attention module and visual-semantic fusion decoder.the semanti reasoning network(SRN) can be trained end-to-end.

At present, the accuracy of the paper cannot be achieved. And i borrowed code from deep-text-recognition-benchmark

model

result

total_accuracy: 80.597

Feature

• predict the character at once time
• DistributedDataParallel training

Requirements

Pytorch >= 1.1.0

Test

1. download the evaluation data from deep-text-recognition-benchmark

2. download the pretrained model from Baidu, Password: d2qn

3. test on the evaluation data

python test.py --eval_data path-to-data --saved_model path-to-model

Train

1. download the training data from deep-text-recognition-benchmark

2. training from scratch

python train.py --train_data path-to-train-data --valid-data path-to-valid-data

Reference

1. bert_ocr.pytorch
2. deep-text-recognition-benchmark
3. 2D Attentional Irregular Scene Text Recognizer
4. Towards Accurate Scene Text Recognition with Semantic Reasoning Networks

difference with the origin paper

• use resnet for 1D feature not resnetFpn 2D feature
• use add not gated unit for visual-semanti fusion decoder

other

It is difficult to achieve the accuracy of the paper, hope more people to try and share