Paper reference for text-recognition-0012

[FVian]

Hi,

Looking at  text-recognition-0012 pre-trained model description page here, the following architecture information was provided:

This is a network for text recognition scenario. It consists of a VGG16-like backbone and bidirectional LSTM encoder-decoder. The network is able to recognize case-insensitive alpha-numeric text (36 unique symbols).

But do you have a paper reference for this architecture?

Best regards,

Felipe Ximenes

[David_C_Intel]

Hello Felipe.

Thank you for reaching out. The text-recognition-0012 is an Intel pre-trained model, as it is not open-sourced, we cannot give more information than the provided here.

If you have more questions, feel free to ask us.

Best regards,

David

[FVian]

Hi David, Thanks for your explanation. I hope in the future Intel open-sourced this model. Best regards, Felipe Ximenes

[Vladimir_Dudnik]

Although this particular model is available in OpenVINO IR form only, you may find TensorFlow text recognition model in OpenVINO training extension

Regards,
  Vladimir

[FVian]

Hi vladimir-dudnik,

Thanks for your reply. I didn't know this repository.

Best regards,

Felipe Ximenes