Overview
I created a web application for NDL Classical Book OCR using Hugging Face Space. You can try it at the link below. Upload an image, and after about 1 minute, the OCR result text and JSON data will be displayed.
https://huggingface.co/spaces/nakamura196/ndl_kotenseki_ocr
The following article was used as a reference for creating this application.
https://qiita.com/relu/items/e882e23a9bd07243211b
Choosing the Right Tool
I have separately prepared a Google Colab tutorial as another environment for trying NDL Classical Book OCR.
The above allows you to use a GPU for free, enabling high-speed OCR processing. However, the initial startup of the notebook requires downloading related libraries and models, which takes about 4 minutes of setup time.
On the other hand, the application created using Hugging Face Space allows OCR processing to be executed immediately, as long as it has been used within the past 72 hours (as of February 2023). It also supports usage via Web API (described later). However, since the free tier does not provide GPU access, processing takes about 1 minute per image. (You can also use a paid tier to access GPU.)
API Usage
I have prepared a notebook where you can try using the API. You may occasionally encounter 504 Gateway Time-out errors, but I hope it serves as a useful reference.
Summary
I hope this is helpful for those who want to try out the accuracy (not speed) of NDL Classical Book OCR.