My project is actually about OCRing the official documents of Nepal that is present in Devnagari and English combined scripts. Generated few datasets of 81 images with their ground truths.* I am confused about the clear roadmap for training them as different sources followed different approaches. If anyone here have done training in Devnagari scripts, It would be of great help to show the roadmap to me. *Plus is using the tesseract ocr the best approach to go for Devnagari scripts as I have also seen that other deep learning approaches are also used posing significant accuracy over Tesseract.
-- You received this message because you are subscribed to the Google Groups "tesseract-ocr" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion visit https://groups.google.com/d/msgid/tesseract-ocr/35d360e3-1726-417a-996e-38ff7e18ad31n%40googlegroups.com.
