This is modified to be used in ImageTrans.

Installation:

1. Install Python (version >= 3.8)
2. Download the project: https://github.com/xulihang/manga-image-translator/archive/refs/heads/main.zip
3. Install dependencies: pip install -r requirements.txt
4. Download the model files into the project's folder: detect.ckpt , ocr.ckpt, inpainting.ckpt
5. Download and unzip the OCR CTC model into the same folder: ocr-ctc.zip. This OCR model supports Korean while the OCR model in the previous step only supports English, Chinese and Japanese. Its speed is also higher.
6. Download the OCR 48px model and its dictionary into the same folder. This OCR model has a higher accuracy but requires GPU.
7. Run the server: python server.py

Test pages:

• http://127.0.0.1:8080/ocr.html
• http://127.0.0.1:8080/getmask.html
• http://127.0.0.1:8080/remove.html

How to Enable GPU

To use GPU for inference, you need to do the following:

1. Install a Pytorch version with GPU support according to this.
2. Create a new file named use_cuda in the root of the project to enable CUDA or use_mps to enable the GPU for Mac.

Use with ImageTrans

You can find the additional guide here: https://github.com/xulihang/ImageTrans_plugins/tree/master/mangaTranslatorOCR