Commit Graph

19991 Commits

Author SHA1 Message Date
Alexander Alekhin
74490399db Merge pull request #6268 from robbertkl:patch-1 2016-03-16 09:59:36 +00:00
Roman Donchenko
9a5d7f1a0b Backport PR #3813 to 2.4
It has already been partially backported by #6199; this commit completes
the backport.
2016-03-16 12:34:41 +03:00
Robbert Klarenbeek
dd510e1de9 Fix incompatibility with some C++11 implementations 2016-03-15 17:21:05 +01:00
Dan
1d21493302 Merge branch 'issue_6251' of http://github.com/dtmoodie/opencv into thrust_allocator_usage_pr 2016-03-15 11:18:41 -04:00
Dan
a21ede9452 Thrust allocator usage. 2016-03-15 11:14:47 -04:00
Dan
7a934f9ee0 Compatibility and bug fixes. 2016-03-15 11:14:42 -04:00
Dan
a8ca560684 Fixed more missing async calls. 2016-03-15 11:14:42 -04:00
Dan
79ecefb51f Implemented async calls. 2016-03-15 11:14:41 -04:00
Alexander Alekhin
97d8ca3464 Merge pull request #6266 from gongzg:master 2016-03-15 12:51:20 +00:00
Zhigang Gong
0b08d2559e fix potential race condition in canny.cl.
See the below code snippet:

while(l_counter != 0)
{
    int mod = l_counter % LOCAL_TOTAL;
    int pix_per_thr = l_counter / LOCAL_TOTAL + ((lid < mod) ? 1 : 0);

    for (int i = 0; i < pix_per_thr; ++i)
    {
        int index = atomic_dec(&l_counter) - 1;
        ....
    }
    ....
    barrier(CLK_LOCAL_MEM_FENCE);
}

If we don't put a barrier before the for loop, then there is a possiblity
that some work item enter this loop but the others are not, the the l_counter
will be reduced in the for loop and may be changed to zero, and the other
work items may can't enter the while loop. If this happens, it breaks the
barrier's rule which requires all the work items reach the same barrier.
And it may hang the GPU depends on the implementation of opencl platform.

This issue is raised at:
https://github.com/Itseez/opencv/issues/5175

Signed-off-by: Zhigang Gong <zhigang.gong@linux.intel.com>
2016-03-15 19:11:15 +08:00
Alexander Alekhin
947307baae Merge pull request #6261 from dtmoodie:issue_6251 2016-03-15 07:51:24 +00:00
Dan
42c36c0c80 Linux bug fix for ThrustAllocator. 2016-03-14 11:09:52 -04:00
Alexander Alekhin
7783206934 Merge pull request #6259 from polevoy-d-v:move_sematics_fix 2016-03-14 14:46:21 +00:00
Alexander Alekhin
5517ea0adb Merge pull request #6255 from delftswa2016:fix-5002 2016-03-14 14:46:06 +00:00
Прун Виктор
a90a576d76 fixed incorrect behaviour of move semantics for cv::Ptr, cv::Mat, cv::UMat in case when rvalue-reference references to *this. 2016-03-14 16:10:23 +03:00
Alexander Alekhin
695a430007 Merge pull request #6254 from berak:patch-2 2016-03-14 11:27:55 +00:00
Alexander Alekhin
708761b219 Merge pull request #6253 from berak:patch-1 2016-03-14 11:27:11 +00:00
Alexander Alekhin
4f9e1a9a58 Merge pull request #6249 from sturkmen72:patch-15 2016-03-14 11:26:51 +00:00
Alexander Alekhin
9d49cef29e Merge pull request #6245 from delftswa2016:fix-for-issue-4375 2016-03-14 11:26:33 +00:00
Alexander Alekhin
cc3433f349 Merge pull request #6045 from matthieu-ft:master 2016-03-14 11:25:23 +00:00
renu285
45f9d32866 Removed older libraries pointing to 2.4 in eclipse tutorial docs 2016-03-14 12:09:56 +01:00
berak
79ac072e6c train_svmsgd.cpp : float response
responses must be float, else we trigger an assertion in line 150
2016-03-14 08:09:36 +01:00
berak
4555f9ac25 svmsgd.cpp, fix #6248
add a cast to double, to make vs compilers happy
2016-03-14 08:02:09 +01:00
Suleyman TURKMEN
826b5a2ef1 Update lda.cpp
typo correction
2016-03-13 21:11:34 +02:00
shruthikashyap
1fbc6ab05d Adding the syntax and usage for cv2.undistortPoints(). 2016-03-13 13:32:24 +01:00
Vadim Pisarevsky
fbc221d334 Merge pull request #6096 from mnoskova:mn/SVMSGD_to_opencv3_0 2016-03-12 17:16:35 +00:00
Vadim Pisarevsky
cd1426ba8d Merge pull request #6025 from sovrasov:new_python_tests 2016-03-12 17:11:19 +00:00
Vadim Pisarevsky
c186f424c5 Merge pull request #6119 from mshabunin:hal_filter2d 2016-03-12 17:08:10 +00:00
Vadim Pisarevsky
c9db05de22 Merge pull request #6137 from dtmoodie:thrust_allocator 2016-03-12 17:07:03 +00:00
Vadim Pisarevsky
bad09e5f58 Merge pull request #6206 from alalek:polar_fix 2016-03-12 16:40:35 +00:00
Vadim Pisarevsky
93c4dfb16d Merge pull request #6236 from alalek:backport_6232 2016-03-12 16:38:56 +00:00
Vadim Pisarevsky
da82acda6b Merge pull request #6241 from cbalint13:agast 2016-03-12 16:38:39 +00:00
Balint Cristian
f819323296 Fix AGAST keypoint declared size. 2016-03-12 16:49:14 +02:00
Alexander Alekhin
52b9207105 Merge pull request #6237 from mshabunin:fix-vtk-on-2.4 2016-03-11 14:20:16 +00:00
Maksim Shabunin
d2e451e877 Ported VTK cmake detection from master branch 2016-03-11 16:25:47 +03:00
Alexander Alekhin
c41dc915fd Merge pull request #6232 from piotr-semenov:semenovp_fixes 2016-03-11 12:23:41 +00:00
Piotr Semenov
00e91fd3d0 Fix that corrects the OpenCV's random access iterator distance: d(x,y) = -d(y,x) 2016-03-11 15:20:36 +03:00
Alexander Alekhin
afc62f076a Merge pull request #6234 from asmorkalov:as/deb_version_check 2016-03-11 10:06:25 +00:00
Alexander Smorkalov
9c4204b89e Added check if debian package version matches version from version.hpp 2016-03-11 11:59:15 +03:00
Alexander Alekhin
bcca213b88 Merge pull request #6224 from alalek:fix_java_binary 2016-03-10 17:03:43 +00:00
Maksim Shabunin
508c382673 Merge pull request #6208 from alalek:templates 2016-03-10 11:35:30 +00:00
Maksim Shabunin
3afb5a6006 Merge pull request #6171 from terfendail:vt/new_python_tests_backport 2016-03-10 09:09:32 +00:00
Piotr Semenov
108ddc199a Fix that corrects the OpenCV's random access iterator distance: d(x,y) = -d(y,x) 2016-03-10 11:58:32 +03:00
Maksim Shabunin
7e44faf635 Merge pull request #6228 from wangguoqin1001:2.4 2016-03-10 08:10:19 +00:00
WANG Guoqin
8cfc87cf01 MacOSX / QTKit capture: trying to fix invalid timer call 2016-03-10 02:49:10 +08:00
Alexander Alekhin
a7d28f3b1f Merge pull request #6217 from krrishnarraj:master 2016-03-09 09:42:17 +00:00
Alexander Alekhin
cbd0ee21d5 revert: java bindings become CMake modules (loadable plugins)
This reverts part of commit 0db10a3052
2016-03-09 12:13:46 +03:00
Vladislav Sovrasov
e2434ab7c3 Use AKAZE detector instead of ORB in feature_homography test 2016-03-09 11:37:25 +03:00
Matthieu FT
78475a47a5 fix: bg substraction for float images with OpenCL 2016-03-07 16:23:05 +01:00
Krishnaraj Bhat
9b8013d193 gcc6: fix misleading indentation warning
cosmetic changes to fix this warning.
real bugs not found
2016-03-07 17:06:16 +05:30