bgshih/crnn
A Lua/Torch7 implementation of CRNN (Convolutional Recurrent Neural Network) for image-based sequence recognition, primarily used for OCR and scene text recognition.
Not currently ranked — collecting fresh signals.
star history
This repository implements the Convolutional Recurrent Neural Network architecture that combines CNN feature extraction, RNN sequence modeling, and CTC (Connectionist Temporal Classification) loss for end-to-end sequence recognition from images. The model is built on Torch7 with CUDA GPU support and can be applied to tasks like scene text recognition and general OCR. A pretrained demo model is provided for immediate testing.
Frequently asked
- What is bgshih/crnn?
- A Lua/Torch7 implementation of CRNN (Convolutional Recurrent Neural Network) for image-based sequence recognition, primarily used for OCR and scene text recognition.
- Is crnn open source?
- Yes — bgshih/crnn is open source, released under the MIT license.
- What language is crnn written in?
- bgshih/crnn is primarily written in Lua.
- How popular is crnn?
- bgshih/crnn has 2.1k stars on GitHub.
- Where can I find crnn?
- bgshih/crnn is on GitHub at https://github.com/bgshih/crnn.