li-ming-fan / ocr-detection-ctpn

OCR, CNN+LSTM (CTPN) for image text detection, tensorflow

OCR-DETECTION-CTPN

CNN+LSTM (CTPN) for image text detection

example results

description

To run this repo:

1, python data_base_normalize.py       # to normalize the pre-normalized background images

2, python data_generator.py 0       # to generate validation data

3, python data_generator.py 1       # to generate training data

4, python script_detect.py       # to train and validate

By 1, the pre-normalized images will firstly be rescaled if not of size 800x600, then 800x600 rects will be cropped from the rescaled images. The 800x600 images will be stored in a newly-maked directory, ./images_base.

By 2 and 3, validation data and training data will be generated. These will be store in the newly-maked directories, ./data_valid and ./data_train, respectively.

By 4, the model will be trained and validated. The validation results will be stored in ./data_valid/results. The ckpt files will be stored in a newly-maked directory, ./model_detect.

detection model

The model is mainly based on the method described in the article:

Detecting Text in Natural Image with Connectionist Text Proposal Network

Zhi Tian, Weilin Huang, Tong He, Pan He, Yu Qiao

https://arxiv.org/abs/1609.03605

ocr-detection-ctpn's Issues

The validation results can't be seen in ./data_valid/results.

I can generate generate validation data and training data and they exist in the corresponding position. But The validation results can't be seen in ./data_valid/results. I have checked the path of this file and it's correct. what is the problem, thank you very much.

Can you please provide any of your best pre-trained model and demo.py file and also procedure of how to use it?

Realtime video prediction

Can you please share the code for realtime video prediction?

any descriptions on loss function?

in the loss function, what do u mean by target_ver and target_hor? do u add side-refinement offset in loss function ?

error when i try training ?

Can you help me fix error when i training ?
ResourceExhaustedError (see above for traceback): OOM when allocating tensor with shape[1,128,752,1184] and type float on /job:localhost/replica:0/task:0/device:GPU:0 by allocator GPU_0_bfc
[[Node: conv_comm/conv3/Conv2D = Conv2D[T=DT_FLOAT, data_format="NCHW", dilations=[1, 1, 1, 1], padding="SAME", strides=[1, 1, 1, 1], use_cudnn_on_gpu=true, _device="/job:localhost/replica:0/task:0/device:GPU:0"](conv_comm/res1/last_relu, conv_comm/conv3/kernel/read)]]
Hint: If you want to see a list of allocated tensors when OOM happens, add report_tensor_allocations_upon_oom to RunOptions for current allocation info.

 [[Node: bdrnn2/bw/bw/stack/_947 = _Recv[client_terminated=false, recv_device="/job:localhost/replica:0/task:0/device:CPU:0", send_device="/job:localhost/replica:0/task:0/device:GPU:0", send_device_incarnation=1, tensor_name="edge_2734_bdrnn2/bw/bw/stack", tensor_type=DT_INT32, _device="/job:localhost/replica:0/task:0/device:CPU:0"]()]]

Hint: If you want to see a list of allocated tensors when OOM happens, add report_tensor_allocations_upon_oom to RunOptions for current allocation info.

你这个有什么衡量指标吗？

Usage

Hello - Would it be possible to provide simple usage notes for this? Eg. command to train, command to run demo, etc... Do you know how much faster this implementation is vs tesseract w/ OpenCL?

Thanks