Hello everyone,
I am currently developing a business card management system and I am looking for an OCR model that can extract text from business card images and then structure the extracted text.
The project requirements are as follows:
High accuracy in extracting text from printed business card images.
Ability to handle various layouts and structures of business cards.
Multilingual support, including Japanese, is preferred.
Capability to structure the extracted text (classify into name, job title, company name, phone number, email address, etc.).
If you know of any models available on Hugging Face that meet these requirements, either for OCR or text structuring, I would greatly appreciate your recommendations.
Also, if there are any caveats or best practices when using the selected models, please do share them.
I am open to using two separate models for OCR and text structuring, as it might provide better results. However, if there is a single model that can fulfill all the requirements, that would be even better.
I look forward to hearing your opinions and advice. Thank you in advance for your help.
Best regards,
Yoshida
