Closed siphyshu closed 4 months ago
I think this discussion is not needed anymore. At the time of opening this issue, I was not aware that it's relatively simple to use tesseract-ocr on linux. As in, it's just a sudo apt install tesseract-ocr
. Incidentally, it also becomes easy to set up on cloud platforms. Hence, tesseract works for now.
Research and test alternatives to
tesseract-ocr
The setup needed for tesseract is a bit too tedious, as the steps involve downloading the binary and setting it in the PATH, to be deployed easily as a web-app like streamlit for example.
If there was a way to simply import a library and get 98% accurate results, it would be a perfect fit.