/MNTSR

Primary LanguagePython

Requirements

  • easydict==1.9
  • editdistance==0.5.3
  • lmdb==1.2.1
  • matplotlib==3.3.4
  • opencv-python==4.5.2.52
  • Pillow==8.2.0
  • tensorboard==2.5.0
  • tensorboard-data-server==0.6.1
  • tensorboard-plugin-wit==1.8.0
  • torch==1.2.0
  • torchvision==0.2.1
  • tqdm==4.61.0
  • ipython

Dataset

Download all resources at Dataset with password: 6dbe

  • TextZoom dataset
  • Pretrained weights of CRNN
  • Pretrained weights of Transformer-based recognizer

All the resources shoulded be placed under ./dataset/mydata, for example

./dataset/mydata/train1
./dataset/mydata/train2
./dataset/mydata/test
./dataset/mydata/pretrain_transformer.pth
...

Pre-trained Model

[Model] with password v4zq

./checkpoint/zxytrain/epoch_best.pth

Testing

CUDA_VISIBLE_DEVICES=GPU_NUM python main.py --batch_size=16 --STN --exp_name EXP_NAME --text_focus --resume YOUR_MODEL --test --test_data_dir ./dataset/mydata/test

Demo for Testing

CUDA_VISIBLE_DEVICES=0 python main.py --batch_size=16 --STN --exp_name zxytest --text_focus --test --test_data_dir ./dataset/mydata/test