deep-text-recognition-benchmark
deep-text-recognition-benchmark copied to clipboard
clovaai
Text recognition (optical character recognition) with deep learning methods, ICCV 2019
that is best repository that i see since 3 years ago.......... it is better than paddle ocr,easy ocr,.... thanks for your best work
--------------------------------------------------------------------------- RuntimeError Traceback (most recent call last) in ----> 1 model.load_state_dict(new_state_dict) ~/text/lib/python3.6/site-packages/torch/nn/modules/module.py in load_state_dict(self, state_dict, strict) 828 if len(error_msgs) > 0: 829 raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format( -->...
So my command is : CUDA_VISIBLE_DEVICES=0 python3.8 train.py --train_data data_lmdb_release/training --valid_data data_lmdb_release/validation --select_data MJ-ST --batch_ratio 0.5-0.5 --Transformation TPS --FeatureExtraction ResNet --SequenceModeling BiLSTM --Prediction Attn --rgb --imgW 130 --imgH 50 --sensitive...
Check whether the splitting ratio adds up to one, e.g. `opt.batch_ratio="0.1-0.2"` would raise a error while "1"or "0.2-0.8" won't.
I have a pre-trained model from EasyOCR(None-VGG-BiLSTM-CTC), and I want to retrain it on my own data. Question: Should I freeze FeatureExtraction and SequenceModeling part and just fine-tune CTC?
I want to use this project to recongnize space,but in fact I train dataset,I found the word image deep text recognition benchmark predict result is: deeptextrecognitionbenchmark so,space char can be...
hello everyone, i use TPS-ResNet-BiLSTM-Attn-case-sensitive.pth pretrained model, is this pretrained model include space character?
Hello, Does this model can handle multiple line text? If we can use Attention mechanism can it give accurate prediction?
Filtering the images containing characters which are not in opt.character Filtering the images whose label is longer than opt.batch_max_length -------------------------------------------------------------------------------- dataset_root: data_lmdb/training/ opt.select_data: ['/'] opt.batch_ratio: ['1'] -------------------------------------------------------------------------------- dataset_root: data_lmdb/training/ dataset:...
