|
https://bugs.kde.org/show_bug.cgi?id=321296
Bug ID: 321296 Summary: opencv error message during face scan Classification: Unclassified Product: digikam Version: 3.3.0 Platform: Compiled Sources OS: Linux Status: UNCONFIRMED Severity: normal Priority: NOR Component: Face Management Assignee: [hidden email] Reporter: [hidden email] hi, during scan i get this: digikam(25201)/KFACE KFaceIface::OpenCVFaceDetector::cascadeResult: detectMultiScale: image size 800 532 searchIncrement 1.1 grouping 3 flags 1 min size 26 26 OpenCV Error: Bad argument (This LBPH model is not computed yet. Did you call the train method?) in predict, file /var/tmp/portage/media-libs/opencv-2.4.5/work/opencv-2.4.5/modules/contrib/src/facerec.cpp, line 822 digikam(25201)/KFACE: cv::Exception: /var/tmp/portage/media-libs/opencv-2.4.5/work/opencv-2.4.5/modules/contrib/src/facerec.cpp:822: error: (-5) This LBPH model is not computed yet. Did you call the train method? in function predic Reproducible: Sometimes -- You are receiving this mail because: You are the assignee for the bug. _______________________________________________ Digikam-devel mailing list [hidden email] https://mail.kde.org/mailman/listinfo/digikam-devel |
|
https://bugs.kde.org/show_bug.cgi?id=321296
Gilles Caulier <[hidden email]> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |[hidden email] Component|Face Management |libkface -- You are receiving this mail because: You are the assignee for the bug. _______________________________________________ Digikam-devel mailing list [hidden email] https://mail.kde.org/mailman/listinfo/digikam-devel |
|
In reply to this post by Simon Bühler
https://bugs.kde.org/show_bug.cgi?id=321296
--- Comment #1 from Gilles Caulier <[hidden email]> --- Which OpenCV release do you use ? Go to Help/Components Info for details. Gilles Caulier -- You are receiving this mail because: You are the assignee for the bug. _______________________________________________ Digikam-devel mailing list [hidden email] https://mail.kde.org/mailman/listinfo/digikam-devel |
|
In reply to this post by Simon Bühler
https://bugs.kde.org/show_bug.cgi?id=321296
--- Comment #2 from Gilles Caulier <[hidden email]> --- Oups, i just see : OpenCV 2.4.5... -- You are receiving this mail because: You are the assignee for the bug. _______________________________________________ Digikam-devel mailing list [hidden email] https://mail.kde.org/mailman/listinfo/digikam-devel |
|
In reply to this post by Simon Bühler
https://bugs.kde.org/show_bug.cgi?id=321296
--- Comment #3 from Marcel Wiesweg <[hidden email]> --- Could it simply be that you really did not train anything yet? All is new with libkface, so chances are you did not rebuild the training data, nor confirm new faces. -- You are receiving this mail because: You are the assignee for the bug. _______________________________________________ Digikam-devel mailing list [hidden email] https://mail.kde.org/mailman/listinfo/digikam-devel |
|
In reply to this post by Simon Bühler
https://bugs.kde.org/show_bug.cgi?id=321296
Tito <[hidden email]> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |[hidden email] --- Comment #4 from Tito <[hidden email]> --- I get the exact same errors. Same opencv version. I scan for faces and confirm some by adding a tag to them. Although some others photos are nearly the same (shot with high frame rate) the recognition status is given as "completed" after 0 or 1 seconds and none of the faces are identified. Deleting the libkface database and starting all over does not change anything. digikam 3.3.0-beta2 on Slackware-current, all compiled from source. -- You are receiving this mail because: You are the assignee for the bug. _______________________________________________ Digikam-devel mailing list [hidden email] https://mail.kde.org/mailman/listinfo/digikam-devel |
|
In reply to this post by Simon Bühler
https://bugs.kde.org/show_bug.cgi?id=321296
--- Comment #5 from Tito <[hidden email]> --- With digikam-3.3.0rc3 the error does not occur no more for me. digikam also recognizes some faces after training now. :) -- You are receiving this mail because: You are the assignee for the bug. _______________________________________________ Digikam-devel mailing list [hidden email] https://mail.kde.org/mailman/listinfo/digikam-devel |
|
In reply to this post by Simon Bühler
https://bugs.kde.org/show_bug.cgi?id=321296
Gilles Caulier <[hidden email]> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|UNCONFIRMED |RESOLVED Version Fixed In| |3.3.0 Resolution|--- |FIXED -- You are receiving this mail because: You are the assignee for the bug. _______________________________________________ Digikam-devel mailing list [hidden email] https://mail.kde.org/mailman/listinfo/digikam-devel |
|
In reply to this post by Simon Bühler
https://bugs.kde.org/show_bug.cgi?id=321296
Slobodan Simic <[hidden email]> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |[hidden email] --- Comment #6 from Slobodan Simic <[hidden email]> --- I have same errors with openCV even with digikam 3.3.0 taken from kubuntu-ppa repository here is console output OpenCV Error: Bad argument (This LBPH model is not computed yet. Did you call the train method?) in predict, file /build/buildd/opencv-2.4.5+dfsg/modules/contrib/src/facerec.cpp, line 822 digikam(2793)/KFACE: cv::Exception: /build/buildd/opencv-2.4.5+dfsg/modules/contrib/src/facerec.cpp:822: error: (-5) This LBPH model is not computed yet. Did you call the train method? in function predict and it goes over and over again... percentage of process done is 14% and stays there and I can continue working in digikam. Is it digikam (openCV) or package error? -- You are receiving this mail because: You are the assignee for the bug. _______________________________________________ Digikam-devel mailing list [hidden email] https://mail.kde.org/mailman/listinfo/digikam-devel |
| Free forum by Nabble | Edit this page |
