A-bone1/Attention-ocr-Chinese-Version
A TensorFlow implementation of attention-based OCR for Chinese text recognition using CRNN architecture.
Not currently ranked — collecting fresh signals.
star history
This project modifies Google’s Attention OCR model for Chinese character recognition. It uses a CRNN architecture combining convolutional layers with attention mechanisms to extract text from street view imagery. The model is trained on FSNS-format datasets containing 5,400 Chinese characters, and supports training custom datasets by following the provided dataset pipeline.
Frequently asked
- What is A-bone1/Attention-ocr-Chinese-Version?
- A TensorFlow implementation of attention-based OCR for Chinese text recognition using CRNN architecture.
- Is Attention-ocr-Chinese-Version open source?
- Yes — A-bone1/Attention-ocr-Chinese-Version is an open-source project tracked on heatdrop.
- What language is Attention-ocr-Chinese-Version written in?
- A-bone1/Attention-ocr-Chinese-Version is primarily written in Python.
- How popular is Attention-ocr-Chinese-Version?
- A-bone1/Attention-ocr-Chinese-Version has 429 stars on GitHub.
- Where can I find Attention-ocr-Chinese-Version?
- A-bone1/Attention-ocr-Chinese-Version is on GitHub at https://github.com/A-bone1/Attention-ocr-Chinese-Version.