m2man / ctpn.pytorch

Pytorch implementation of CTPN (Detecting Text in Natural Image with Connectionist Text Proposal Network)

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

ctpn.pytorch

Pytorch implementation of CTPN (Detecting Text in Natural Image with Connectionist Text Proposal Network)

Paper

https://arxiv.org/pdf/1609.03605.pdf

train

training dataset: ICDAR2013 and ICDAR2017.
If you want to train your own dataset, you need to change the 'img_dir' and 'label_dir' in file ctpn/config.py, then run

python train.py

training loss

predict

Download pretrained model from './weights/', change the test image path in file predict.py, then run:

python predict.py

result

result_1
result_2

references

https://github.com/opconty/pytorch_ctpn

https://github.com/courao/ocr.pytorch

About

Pytorch implementation of CTPN (Detecting Text in Natural Image with Connectionist Text Proposal Network)

License:MIT License


Languages

Language:Python 100.0%