The code and trained models of:
TextField: Learning A Deep Direction Field for Irregular Scene Text Detection, TIP 2019 [Paper]
Please cite the related works in your publications if it helps your research:
@article{xu2018textfield,
title={TextField: Learning A Deep Direction Field for Irregular Scene Text Detection},
author={Xu, Yongchao and Wang, Yukang and Zhou, Wei and Wang, Yongpan and Yang, Zhibo and Bai, Xiang},
journal={arXiv preprint arXiv:1812.01393},
year={2018}
}
-
Caffe and SynthText pretrained model [Link]
-
Datasets: [Total-Text], [ICDAR2015]
-
OpenCV 3.4.3
-
MATLAB
cp Makefile.config.example Makefile.config
# adjust Makefile.config (for example, enable python layer)
make all -j16
# make sure to include $CAFFE_ROOT/python to your PYTHONPATH.
make pycaffe
Please refer to Caffe Installation to ensure other dependencies.
# download datasets and pretrained model then
mkdir data && mv [your_dataset_folder] data/
mkdir models && mv [your_pretrained_model] models/
# an example on Total-Text dataset
cd examples/TextField/
python train.py --gpu [your_gpu_id] --dataset total --initmodel ../../models/synth_iter_800000.caffemodel
# an example on Total-Text dataset
cd evaluation/total/
./eval.sh
| Recall | Precision | F-measure | Link |
|---|---|---|---|
| 0.816 | 0.824 | 0.820 | [Google drive] |
*lambda=0.50 for post-processing
| Recall | Precision | F-measure | Link |
|---|---|---|---|
| 0.811 | 0.846 | 0.828 | [Google drive] |
*lambda=0.75 for post-processing