ekane3 / ocr-card-app

Projet Transverse de M2

OCR Part of the YACC project, made by Team YACC

➊ To run OCR make sure to downlad detection weights:

Download detection model weights , rename it as db_resnet50_latest.pt if it is not already the case and put it in this folder

➋ Install dependencies by running the following command:

$pip install -r requirements.txt

If you face dependencies issues try to use virtual python environment.

➌ After that, to perform OCR on YuGiOh Cards, run the following command:

$python main.py <path to your image file>

we provide a test image, you could run the following command to test it:

$python main.py test.jpg

➍ You can explore the notebooks folder to understand how we've scrapped data and made our ~5000 YugiOh card images dataset, and how we've made our custom card detector:

https://www.cards-capital.com/ © all rights reserved for the card images