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
Hi! I'm using partition_pdf on a PDF file that includes tables that have been rotated 90° to the left (ie. the top left cell is on the bottom left part of the page). The OCR step does not take into account this and the output text is incorrect. For example: the output text starts with n o i t a t r o p s n a r T (transportation is the header of the column on the right of the table).
Do I understand correctly that there is no possibility with partition_pdf to detect text orientation before performing OCR?
Thanks!
The text was updated successfully, but these errors were encountered:
Hi @vivien000, We don't officially support detecting text orientation before performing OCR on rotated text at the moment, though it's something we could add support for in the future.
Hi! I'm using
partition_pdf
on a PDF file that includes tables that have been rotated 90° to the left (ie. the top left cell is on the bottom left part of the page). The OCR step does not take into account this and the output text is incorrect. For example: the output text starts withn o i t a t r o p s n a r T
(transportation
is the header of the column on the right of the table).Do I understand correctly that there is no possibility with
partition_pdf
to detect text orientation before performing OCR?Thanks!
The text was updated successfully, but these errors were encountered: