This is the official implementation of "Multi-Oriented Scene Text Detection via Corner Localization and Region Segmentation".

For more details, please refer to our paper.

Please cite the paper in your publications if it helps your research:

@inproceedings{lyu2018multi,
      title={Multi-oriented scene text detection via corner localization and region segmentation},
      author={Lyu, Pengyuan and Yao, Cong and Wu, Wenhao and Yan, Shuicheng and Bai, Xiang},
      booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition},
      pages={7553--7563},
      year={2018}
}

1. Requirements
2. Installation
3. Models
4. Test
5. Train
6. License

• NVIDIA GPU, Ubuntu 14.04, Python2.7, CUDA8/9
• PyTorch 0.2.0_3

git clone https://github.com/lvpengyuan/corner.git
sh ./make.sh   or  cd rpsroi_pooling && python build.py

Download the model and place it in weights/

Our trained model: Google Drive;

You can test a model in a single scale:

python eval_all.py

or in multi-scale:

python eval_multiscale.py

Note that, you should modify the model path and the test dataset before testing.

python train.py

To train a new model, you should modify the training settings before training.

This code is only for academic purpose.