* 'master' of github.com:itseez/opencv: (82 commits)
moved part of video to contrib/{outflow, bgsegm}; moved matlab to contrib
added some basic functionality needed by the new face module (moved from the old "contrib")
moved to the new opencv_contrib/face module
fixed various warnings and obvious errors reported by clang compiler and the coverity tool.
Fixed review comment from Vadim Pisarevsky
modified farneback sample to use T-API
ECC patch by the author (G. Evangelidis); fixed some OCL Farneback optical flow test failures on Mac
small fix for GaussianBlur ocl test
fix binary package build
small fix for ocl_resize
fix IOS framework
fixed test ocl_MatchTemplate for sparse matrix
Fixed typos
fixing error, wrong template method param.
fixing Mac build
some formal changes (generally adding constness)
Fixed choice of kercn and rowsPerWI for non-Intel device.
fixed nDiffs for CalcBackProject
fixed tests for ocl_filter2d, ocl_matchTemplate, ocl_histogram.cpp
Fixed issue: Mat::copyTo(UMat) if device copy is obsolete. Added test.
...
Conflicts:
modules/core/include/opencv2/core/mat.inl.hpp
According to opencl 1.2 spec 5.4.2:
enqueues a command to unmap a previously mapped region of a memory object.
...
CL_INVALID_VALUE if mapped_ptr is not a valid pointer returned by
clEnqueueMapBuffer, or clEnqueueMapImage for memobj.
So if the u->data is not from a clEnqueueMapBuffer call, we should not
call clEnqueueUnmapMemObject() unmap it. With this patch, the cases
./opencv_test_video --gtest_filter=OCL_Video/FarnebackOpticalFlow.Mat/*
could work well with beignet 0.9.1, Otherwise, it will get a
CL_INVALID_VALUE at the clEnqueueUnmapMemObject().
Signed-off-by: Zhigang Gong <zhigang.gong@intel.com>
I propose forEach method for cv::Mat and cv::Mat_.
This is solution for the overhead of MatIterator_<_Tp>.
I runs a test that micro opecode runs all over the pixel of cv::Mat_<cv::Point3_<uint8_t>>.
And this implementation 40% faster than the simple pointer, 80% faster than iterator.
With OpenMP, 70% faster than simple pointer, 95% faster than iterator (Core i7 920).
Above all, code is more readable.
My test code is here.
https://gist.github.com/kazuki-ma/8285876
Thanks.
