Login  Register

[digikam] [Bug 331912] New: face detection and recognition dos not work, tags and factags counts for one person are different, memory goes up and up without any result

Posted by G.Rass on Mar 09, 2014; 1:19pm
URL: http://digikam.185.s1.nabble.com/digikam-Bug-331912-New-face-detection-and-recognition-dos-not-work-tags-and-factags-counts-for-one-pt-tp4668660.html

https://bugs.kde.org/show_bug.cgi?id=331912

            Bug ID: 331912
           Summary: face detection and recognition dos not work, tags and
                    factags counts for one person are different, memory
                    goes up and up without any result
    Classification: Unclassified
           Product: digikam
           Version: 3.5.0
          Platform: openSUSE RPMs
                OS: Linux
            Status: UNCONFIRMED
          Severity: major
          Priority: NOR
         Component: Face Management
          Assignee: [hidden email]
          Reporter: [hidden email]

First I cleared for one person the difference of counts in face tags (100) and
tags (104) to 104 face tags. Doing this cased to fill the memory to no limit,
so i killed digiKam.
Then I restarted digikam and tried to find more faces, but the memory filled up
again. (I did it all with valgrind this time)

Reproducible: Always

Steps to Reproduce:
1. ad a face tag
 or start detecting faces
Actual Results:  
It not possible to us digikam without killing it and restarting the program.
sometimes digikam remains in the process table although the window closed with
killing.

Expected Results:  
Adding new face tags without filling up the memory, also with the automatic
face detection.

I think the database can not deal with the difference of the tag of a person
and the face tag of a person.
Sometimes there is even the same face taged twice in one picture.

Valgrind:


==4621==
==4621== 1,712 bytes in 2 blocks are possibly lost in loss record 26,227 of
26,933
==4621==    at 0x4C277AB: malloc (vg_replace_malloc.c:270)
==4621==    by 0x1E6388B6: ??? (in /usr/lib64/libsqlite3.so.0.8.6)
==4621==    by 0x1E614D29: ??? (in /usr/lib64/libsqlite3.so.0.8.6)
==4621==    by 0x1E61C9B7: ??? (in /usr/lib64/libsqlite3.so.0.8.6)
==4621==    by 0x1E61C9DC: ??? (in /usr/lib64/libsqlite3.so.0.8.6)
==4621==    by 0x1E65124D: ??? (in /usr/lib64/libsqlite3.so.0.8.6)
==4621==    by 0x1E6948D6: ??? (in /usr/lib64/libsqlite3.so.0.8.6)
==4621==    by 0x2B7B216E: ??? (in
/usr/lib64/qt4/plugins/sqldrivers/libqsqlite.so)
==4621==    by 0x4E468A0: QSqlDatabase::open() (in
/usr/lib64/libQtSql.so.4.8.5)
==4621==    by 0x783C3B0:
Digikam::DatabaseCoreBackendPrivate::open(QSqlDatabase&)
(databasecorebackend.cpp:213)
==4621==    by 0x783C97C:
Digikam::DatabaseCoreBackendPrivate::databaseForThread()
(databasecorebackend.cpp:119)
==4621==    by 0x783EE82:
Digikam::DatabaseCoreBackend::open(Digikam::DatabaseParameters const&)
(databasecorebackend.cpp:768)
==4621==
==4621== 1,728 bytes in 18 blocks are possibly lost in loss record 26,230 of
26,933
==4621==    at 0x4C277AB: malloc (vg_replace_malloc.c:270)
==4621==    by 0x1E6388B6: ??? (in /usr/lib64/libsqlite3.so.0.8.6)
==4621==    by 0x1E614D29: ??? (in /usr/lib64/libsqlite3.so.0.8.6)
==4621==    by 0x1E61C9B7: ??? (in /usr/lib64/libsqlite3.so.0.8.6)
==4621==    by 0x1E61CAF0: ??? (in /usr/lib64/libsqlite3.so.0.8.6)
==4621==    by 0x1E61CB6C: ??? (in /usr/lib64/libsqlite3.so.0.8.6)
==4621==    by 0x1E6219DA: ??? (in /usr/lib64/libsqlite3.so.0.8.6)
==4621==    by 0x1E641B50: ??? (in /usr/lib64/libsqlite3.so.0.8.6)
==4621==    by 0x1E641C6B: ??? (in /usr/lib64/libsqlite3.so.0.8.6)
==4621==    by 0x1E641E1C: sqlite3_create_function_v2 (in
/usr/lib64/libsqlite3.so.0.8.6)
==4621==    by 0x1E641E64: sqlite3_create_function (in
/usr/lib64/libsqlite3.so.0.8.6)
==4621==    by 0x1E694BCB: ??? (in /usr/lib64/libsqlite3.so.0.8.6)
==4621==
==4621== 1,728 bytes in 18 blocks are possibly lost in loss record 26,231 of
26,933

--
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