| Method | Backbone | Fine-tuning | Config | Precision (%) | Recall (%) | F-measure (%) | Model | Log |
|---|---|---|---|---|---|---|---|---|
| PaddlePaddle_PANet | ResNet18 | N | panet_r18_ctw.py | 84.51 | 78.62 | 81.46 | Model | Log |
| mmocr_PANet | Resnet18 | N | -- | 77.6 | 83.8 | 80.6 | -- | -- |
Python 3.6+
paddlepaddle-gpu 2.0.2
nccl 2.0+
mmcv 0.2.12
editdistance
Polygon3
pyclipper
opencv-python 3.4.2.17
Cython
Install paddle following the official tutorial.
pip install -r requirement.txt
./compile.shPlease refer to dataset/README.md for dataset preparation.
download resent18 pre-train model in
pretrain/resnet18.pdparams
pretrain_resnet18 password: j5g3
CUDA_VISIBLE_DEVICES=0,1,2,3 python dist_train.py ${CONFIG_FILE}
For example:
CUDA_VISIBLE_DEVICES=0,1,2,3 python dist_train.py config/pan/pan_r18_ctw.py
#checkpoint continue
python3.7 dist_train.py config/pan/pan_r18_ctw_train.py --nprocs 1 --resume checkpoints/pan_r18_ctw_trainThe evaluation scripts of CTW 1500 dataset. CTW
Text detection
./start_test.shThis project is developed and maintained by IMAGINE Lab@National Key Laboratory for Novel Software Technology, Nanjing University.
This project is released under the Apache 2.0 license.