Team,
I’m trying to use digikam but id does a mess.
Before I start opening bugs can you be so kind to help me?
I’ve mapped more or less 100 people I’m interested in with lightroom, then I wrote all those info in the images in the XMP data so that I have the name/hierarchy and the region.
I’ve imported the pictures in a new database in digiKam and the result is a total mess.
-
it deadlocked three times in scanning the pictures (21K pictures) - I had to exit and enter again for the scan to restart
-
the contact structure is messed up: I have tags outside of the hierarchy:
-
it screwed up all faces putting tens or hundreds of faces under a name as Confirmed when they are not, this should be all the same face 
this is a MESS no face identical
It continues to loose the OpenCL status, after every time I open it I have to Test GPU… and after it becomes:
this are my settings:
I’ve changed the file to point to my exiftool that is a higher version but it does not change here
I enabled the debug log and I get this error:
but it makes no sense as I’ve selected the Exiftool executable…
@msavazzi:
First, welcome to the KDE Discuss forums.
Given that, the face recognition has been executed with Adobe Photoshop Lightroom and, you’ve imported the files (.JPEG?) into digiKam, a question which needs to be asked is:
- Which platform are you using to execute digiKam?
And, which hardware is being used for the Operating System?
Is there enough main memory?
Is there enough free disk space?
Is the hardware reasonably modern and, does it have enough performance?
An issue with the Face Recognition could be, that digiKam really needs to scan the faces for itself –
- The Face Recognition algorithm isn’t the same Face Recognition algorithm as that used by Adobe.
OK, you’ve written Face Recognition tags to the XMP area in the files but, AFAIK there ain’t no method for digiKam to extract that data into it’s own Face Recognition database.
- Which platform are you using to execute digiKam?
Windows 11
And, which hardware is being used for the Operating System?
Windows 11 on intel
Is there enough main memory?
yes 64+GB
Is there enough free disk space?
yes 1Tb
Is the hardware reasonably modern and, does it have enough performance?
yes
An issue with the Face Recognition could be, that digiKam really needs to scan the faces for itself
–> this is bad, it should start from all the data already in the images
- The Face Recognition algorithm isn’t the same Face Recognition algorithm as that used by Adobe.
→ yes that’s why I want to use it
OK, you’ve written Face Recognition tags to the XMP area in the files but, AFAIK there ain’t no method for digiKam to extract that data into it’s own Face Recognition database.
→ this is bad, it should read the images and train the algorithm on the already tagged images
It makes no sense to start from scratch when you have good an certified data already available
@msavazzi:
We need to wait for a digiKam developer to chime in.
Given that, possibly, you’re using the default SQLite digiKam database, have you enabled the “WAL” mode for the databases?
<The SQLite Database>
@msavazzi:
BTW, I’m using the following system for digiKam with about 18 thousand photos in the (SQLite) database –
Operating System: openSUSE Leap 15.6
KDE Plasma Version: 5.27.11
KDE Frameworks Version: 5.115.0
Qt Version: 5.15.12
Kernel Version: 6.4.0-150600.23.50-default (64-bit)
Graphics Platform: X11
Processors: 8 × AMD Ryzen 5 3400G with Radeon Vega Graphics
Memory: 29.3 GiB of RAM
Graphics Processor: AMD Radeon Vega 11 Graphics
Manufacturer: ASUS