Bug 484743

Summary:	Face recognition, new run, new assignment
Product:	[Applications] digikam	Reporter:	Andy <zunar>
Component:	Faces-Recognition	Assignee:	Digikam Developers <digikam-bugs-null>
Status:	CONFIRMED ---
Severity:	normal	CC:	caulier.gilles, michael_miller
Priority:	NOR
Version First Reported In:	8.4.0
Target Milestone:	---
Platform:	Microsoft Windows
OS:	Microsoft Windows
Latest Commit:		Version Fixed/Implemented In:
Sentry Crash Report:

Description Andy 2024-03-30 10:17:32 UTC

Hello,
The question is just for my understanding.
With face recognition, an extremely large number of faces are incorrectly assigned to a person. Let's say a total of 19000 unconfirmed faces. If I assign some of the faces of the 19,000 photos to the correct person and start face recognition again, will the remaining 18,900 faces be analyzed again and possibly moved to other people? Maybe even ignored if I previously ignored faces that looked the same?
I couldn't find anything about it in the manual.

Regards
Andy

Comment 1 caulier.gilles 2024-10-07 20:48:31 UTC

Michael,

Another entry where we can respond on the right way when SFace DNN model will be ready for production.

Best

GIlles

Comment 2 Michael Miller 2024-10-20 15:38:10 UTC

I think this is a documentation issue.  I'll address it in the updated 8.5.0 face management documentation.

Cheers,
Mike

Comment 3 caulier.gilles 2024-12-01 09:04:02 UTC

digiKam 8.5.0 is out with many improvements in face detection and recognition. Please update these entry accordingly with this version. Thanks in advance...

https://www.digikam.org/news/2024-11-16-8.5.0_release_announcement/

Comment 4 caulier.gilles 2025-04-11 18:13:49 UTC

Hi,

The 8.7.0 pre-release Windows installer from today have been rebuilt from
scratch with Qt 6.8.3, KDE 6.12, OpenCV 4.11 + CUDA support, Exiv2 0.28.5, ExifTool 13.27, ffmpeg 7, all image codecs updated to last version (jxl, avif, heif, aom, etc.).

Please try with this version to see if your problem still reproducible...

https://files.kde.org/digikam/

Thanks in advance
Best regards

Gilles Caulier