Bug 397744 - Face recognition not working
Summary: Face recognition not working
Status: RESOLVED WORKSFORME
Alias: None
Product: digikam
Classification: Applications
Component: Faces-Recognition (show other bugs)
Version: 6.0.0
Platform: Appimage Linux
: NOR normal
Target Milestone: ---
Assignee: Digikam Developers
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2018-08-22 13:40 UTC by Rob D
Modified: 2019-03-28 20:49 UTC (History)
1 user (show)

See Also:
Latest Commit:
Version Fixed In: 6.1.0
Sentry Crash Report:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Rob D 2018-08-22 13:40:59 UTC
Faces are detected, but not recognised.  This is the error I get when launched on the command line.  

---

Digikam::DetectionWorker::process: Found 1 faces in "P7123086.ORF" QSize(2400, 3200) QSize(3472, 4640)
OpenCV(3.4.1) Error: Bad argument (This LBPH model is not computed yet. Did you call the train method?) in predict, file /b/dktemp/digikam-master/core/libs/facesengine/recognition-opencv-lbph/facerec_borrowed.cpp, line 376
Digikam::RecognitionDatabase::recognizeFaces: cv::Exception: OpenCV(3.4.1) /b/dktemp/digikam-master/core/libs/facesengine/recognition-opencv-lbph/facerec_borrowed.cpp:376: error: (-5) This LBPH model is not computed yet. Did you call the train method? in function predict
---

Tested with appimage on Fedora 28 and CentOS 7.4
Comment 1 Maik Qualmann 2018-08-22 19:33:06 UTC
This is not a real error message and no bug. There are no entries in the face database. In order for faces to be recognized, they must train the database by entering a few names for the found faces.

Maik
Comment 2 caulier.gilles 2019-03-20 15:16:00 UTC
After 3 weeks of work, i finally completed the compilation of AppImage using Qt
5.11.3 + QWebkit 5.212.

New 6.1.0 pre-release AppImage bundle can be found here (64 bits only for the
moment) :

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

Please check if this bugzilla entry still valid.

Thanks in advance

Gilles Caulier
Comment 3 Maik Qualmann 2019-03-28 20:49:57 UTC
This is not an error, the database has to be trained to get results. I close the bug now.

Maik