can I modify this so it does both text and equation instead of taking a snip.

lukas-blecher / LaTeX-OCR

pix2tex: Using a ViT to convert images of equations into LaTeX code.

https://lukas-blecher.github.io/LaTeX-OCR/

MIT License

12.43k stars 1.02k forks source link

can I modify this so it does both text and equation instead of taking a snip. #233

Open hayderab opened 1 year ago

hayderab commented 1 year ago

I want to test out the image which includes text and equation, instead of sniping only the equation part can it be modified to read the whole image and text like mathpix snip tool?

lukas-blecher commented 1 year ago

The dataset does not contain these kind of images so its hard to do without an equation detection step first + some other OCR software like tesseract. It is possible, but not the easiest thing to do.