You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
This repository has been archived by the owner on Jul 13, 2024. It is now read-only.
We planned to add OCR, so that players can highlight words.
The OCR feature needs to return text, and bounding boxes of words. Once we have bounding boxes of words, we can detect where on a page the player is looking or pointing or touching with their controller, and then highlight words based on the intersection of the bounding boxes and the player's interaction.
We were considering using a Google API for this.
However, I did some more research about open source and offline alternatives. Here is a list of Python wrappers for tesseract: https://pypi.python.org/pypi?%3Aaction=search&term=tesseract&submit=search
Tesseract OCR returns bounding text and bounding boxes. The API is advanced, though, and written in C++(?)
tesseract-sip is a set of Python wrappers for Tesseract OCR. It depends on Tesseract.
https://pypi.python.org/pypi/tesseract_sip/0.1.1
Examples of interacting with external Python scripts can be seen in our Unity project.