I have used digiKam 5 & 6 now (on Windows) , wanting to move away from Picassa and keep my large Name Tagging efforts in tact. The Tags are written in the photos.
When I import and review the name tags digiKam seems to do a pretty good job of keeping the names , however I have noticed that the regions can be regularly in the wrong place. So I end up with a tree called Uncle Bob.
I have also noticed that the problem seems to be with photos that have a portrait orientation. I don’t rotate images, I rely on software reading the tags to rotate for viewing.It looks like the regions for the box are pre-rotation. so for example, if I have 3 faces all tagged inPpicassa on a portrait photo, and look at them in Digikam the boxes are where the faces would be if the photo had not been rotated.
I am reasonably new to the world of digiKam and how photo tagging actually works, it looks like from the raw data added to the photo, which I viewed with Adobe Bridge XML window, Picassa and digiKam write the tags differently.
What I want to know is are the misplaced portrait regions a bug with digiKam or Picassa , and therefore is my tagging data wrong and need deleting , or is it right and I should report a bug , which when fixed, show all my data correctly.
Does anyone else have any experience of this issue ?