/research-charnet

CharNet: Convolutional Character Networks

Primary LanguagePythonOtherNOASSERTION

Convolutional Character Networks

This project hosts the testing code for CharNet, described in our paper:

Convolutional Character Networks
Linjie Xing, Zhi Tian, Weilin Huang, and Matthew R. Scott;
In: Proceedings of the IEEE International Conference on Computer Vision (ICCV), 2019.

Installation

pip install torch torchvision
python setup.py build develop

Run

  1. Please run bash download_weights.sh to download our trained weights.

  2. For ICDAR 2015, please run the following command line. Please replace images_dir with the directory containing ICDAR 2015 testing images. The results will be in results_dir.

    python tools/test_net.py configs/icdar2015_hourglass88.yaml <images_dir> <results_dir>
    

Citation

If you find this work useful for your research, please cite as:

@inproceedings{xing2019charnet,
title={Convolutional Character Networks},
author={Xing, Linjie and Tian, Zhi and Huang, Weilin and Scott, Matthew R},
booktitle={Proceedings of the IEEE International Conference on Computer Vision (ICCV)},
year={2019}
}

Contact

For any questions, please feel free to reach:

github@malong.com

License

CharNet is CC-BY-NC 4.0 licensed, as found in the LICENSE file. It is released for academic research / non-commercial use only. If you wish to use for commercial purposes, please contact bd@malong.com.