Kornia VS tesserocr

Kornia: Differential Computer Vision\ (3 comments)

Kornia - Kornia is a differentiable computer vision library for PyTorch. It consists of a set of routines and differentiable modules to solve generic computer vision problems. At its core, the package uses PyTorch as its main backend both for efficiency and to take advantage of the reverse-mode auto-differentiation to define and compute the gradient of complex functions.

Use kornia.augmentation where this problem is solved doing the augmentations in batch outside the dataloader. https://github.com/kornia/kornia

I haven't had a situation to use it, but I think Kornia looks cool: https://github.com/kornia/kornia

If you’re after speed I’d recommend mss for screenshots/recording, and tesserocr instead of pytesseract (note in particular the OpenCV option.

tesserocr is an actual binding to the tesseract library, and is better in practically every way than pytesseract (more efficient, more options for usage, doesn’t require saving images to disk before they can be processed, and more).

If you have an electronically created pdf (not scanned) and you’re just wanting to run OCR on embedded images then you’ll want a pdf library that can extract the figure images for you, and then you can use tesserocr to run OCR on those images.