In CVS revision 1.28 a "pictureFound" callback was added to the text handler.
I'm using wv2 to extract text, so I wonder if it's of any utility to me. Maybe the picture caption is transmitted here?
Do you have any idea how I can exploit it?
I don't think we actually process any picture caption yet, so I wouldn't worry about it. Maybe in the future as we get more features implemented, we'll get that capability.