simplify23/cdistnet

Official Pytorch implementations of CDistNet: Perceiving Multi-Domain Character Distance for Robust Text Recognition（IJCV）

cdistnet's Issues

How many layers of MDCDP net did you use in best performing english model?

Any way to train only on language and not images?

I have a relatively small dataset of a different format (license plates) and it often gets license plate format wrong.

I was wondering if there was a way to train the model on just a bunch of text string data without feeding any images at all in order to enforce the format.

Please let me know if it is possible to train the language/semantic model independently, by just feeding string text data of words, without corresponding images.

Is it possible to inference on images

Here i came to know that we are taking predictions on lmdb dataset. So is there any chance to get results out of images ?

about inference

How to set the parameters of input_char，such as predict a new image

CDistNetv2

@simplify23 are you planning to release CDistNetv2 code?
waiting for light weight and faster module

ONNX and tensorRT conversion

did you convert your model into onnx and tensorrt? could you please provide the code?

plan to release code?

How should i prepare dictionary for train on other language?

I don't know how dict_36.txt can train with other characters like upper case.

Open source license?

Are you willing to specify an open-source license such as a MIT License?
The github has no license specified.

Eval request result.csv

Hello,
I don't know what is result.csv and how to to get it?
Thank you

Attention Maps

Could you please release the code to generate the attention maps as published in the paper

Missing transformer

When trying to run test.py I get the following error:

(CDistNet) C:\<path>\CDistNet>python test.py --i_path ..\examples\300_0.jpg 
configs/CDistNet_config.py
<class 'str'>
Traceback (most recent call last):
  File "test.py", line 175, in <module>
    main()
  File "test.py", line 168, in main
    test_one(cfg, args)
  File "test.py", line 126, in test_one
    en = get_parameter_number(model.transformer.encoder)
  File "C:\<path>\miniconda3\envs\CDistNet\lib\site-packages\torch\nn\modules\module.py", line 1178, in __getattr__ 
    type(self).__name__, name))
AttributeError: 'CDistNet' object has no attribute 'transformer'

plan to release code?

waiting for code release

train for other language

hello thanks for your paper and released codes
I want to train your code for other language but I see in lmdbdataset that you use English char and limit the max length to 30 that is true?
I should change line 245 and 246?

`def len(self):
return self.length

def get(self,idx):
    with self.env.begin(write=False) as txn:
        image_key, label_key = f'image-{idx+1:09d}', f'label-{idx+1:09d}'
        label = str(txn.get(label_key.encode()), 'utf-8')  # label
        label = re.sub('[^0-9a-zA-Z]+', '', label)
        label = label[:30]`

Questions on Conv2d of Transformer Layers

Hello, while examining the code,

I noticed that most of the nn.Linear() operations are replaced with nn.Conv2d(kernel_size=(1,1)) operations

when comparing nn.Transformer and the implementation of the code.

Is there a benefit for such replacement?

Has your paper been accepted?

accuracy is lower than other models

I tried to train your model but I got accuracy is lower than other transformer models. could you please let me know how can I got higher accuracy ?

Inference Time

I tried your network and got a good result but I faced the problem of inference speed. could you please let me know I can increase the speed of recognition?

Can it work with phone number ?

I trained the model with billboards but when I inference it, it doesn't work well with sequences of numbers or phone numbers. Can you help me? Thanks very much.