Li Peng [Tue, 29 Nov 2016 07:28:53 +0000 (15:28 +0800)]
more optimization for warpAffine and warpPerspective
Add new OpenCL kernels for bicubic interploation, it is 20% faster
than current warp image kernel with bicubic interploation.
Signed-off-by: Li Peng <peng.li@intel.com>
Vadim Pisarevsky [Tue, 29 Nov 2016 16:42:04 +0000 (16:42 +0000)]
Merge pull request #7538 from Tetragramm:CLAHEfix
Vadim Pisarevsky [Tue, 29 Nov 2016 16:40:10 +0000 (16:40 +0000)]
Merge pull request #7610 from terfendail:ovx_sobel
Vitaly Tuzov [Tue, 29 Nov 2016 16:34:46 +0000 (19:34 +0300)]
Added wrappers for OpenVX context attributes and setter for convolution scale attribute (#7732)
Vitaly Tuzov [Thu, 3 Nov 2016 13:54:06 +0000 (16:54 +0300)]
Added OpenVX based processing to Sobel
Alexander Alekhin [Tue, 29 Nov 2016 12:19:18 +0000 (12:19 +0000)]
Merge pull request #7726 from pengli:warp_image
Alexander Alekhin [Tue, 29 Nov 2016 12:15:08 +0000 (12:15 +0000)]
Merge pull request #7747 from alalek:fix_cuda_test
Li Peng [Wed, 23 Nov 2016 04:57:33 +0000 (12:57 +0800)]
optimization for warpAffine and warpPerspective
Add new ocl kernels for warpAffine and warpPerspective,
The average performance improvemnt is about 30%. The new
ocl kernels require CV_8UC1 format and support nearest
neighbor and bilinear interpolation.
Signed-off-by: Li Peng <peng.li@intel.com>
Alexander Alekhin [Mon, 28 Nov 2016 20:56:58 +0000 (23:56 +0300)]
test: fix cuda build
Vadim Pisarevsky [Mon, 28 Nov 2016 21:04:03 +0000 (21:04 +0000)]
Merge pull request #7548 from tomoaki0705:fixRngAvoidFmadd
Alexander Alekhin [Mon, 28 Nov 2016 16:10:12 +0000 (19:10 +0300)]
Update googletests 1.8.0 (
20160822) (#7738)
* restore Google Test 1.7.0 (get patch)
* ts: update Google Test to 1.8.0 release
https://github.com/google/googletest
* ts: re-apply OpenCV patch for gtest
* ts: fixes for gtest 1.8.0
* ts: workaround MSVS2015 problem in gtest
Vadim Pisarevsky [Mon, 28 Nov 2016 16:09:42 +0000 (16:09 +0000)]
Merge pull request #7745 from ColinFinck:master
Vitaly Tuzov [Fri, 25 Nov 2016 16:13:27 +0000 (19:13 +0300)]
Added wrappers for OpenVX context attributes and setter for convolution scale attribute
Colin Finck [Mon, 28 Nov 2016 09:15:16 +0000 (10:15 +0100)]
objdetect: Properly check for C++11 as it's done in the matching .cpp file. Fixes #7623
Alexander Alekhin [Sat, 26 Nov 2016 08:15:56 +0000 (08:15 +0000)]
Merge pull request #7735 from alalek:ts_cv_enum_static_fiasco
Alexander Alekhin [Sat, 26 Nov 2016 08:14:45 +0000 (08:14 +0000)]
Merge pull request #7734 from alalek:fix_gtk_deprecated
Alexander Alekhin [Sat, 26 Nov 2016 08:11:59 +0000 (08:11 +0000)]
Merge pull request #7730 from alalek:fix_contrib_872
Alexander Alekhin [Fri, 25 Nov 2016 11:56:00 +0000 (14:56 +0300)]
ts: CV_ENUM fix static initialization order fiasco
Alexander Alekhin [Fri, 25 Nov 2016 13:16:25 +0000 (16:16 +0300)]
highgui: g_thread_create deprecated warning
Alexander Alekhin [Fri, 25 Nov 2016 15:19:43 +0000 (18:19 +0300)]
imgproc: fix findContours (submatrix input)
Vadim Pisarevsky [Fri, 25 Nov 2016 14:24:23 +0000 (14:24 +0000)]
Merge pull request #7710 from alalek:debug_check_matrelease
Alexander Alekhin [Fri, 25 Nov 2016 10:02:36 +0000 (10:02 +0000)]
Merge pull request #7721 from apavlenko:vx-fix-build-1.0.1
Alexander Alekhin [Fri, 25 Nov 2016 09:22:07 +0000 (09:22 +0000)]
Merge pull request #7719 from paroj:calib_ret
Alexander Alekhin [Thu, 24 Nov 2016 16:15:56 +0000 (16:15 +0000)]
Merge pull request #7709 from alalek:fix_java_tests
apavlenko [Thu, 24 Nov 2016 15:48:51 +0000 (18:48 +0300)]
eliminating conflict with ivx.hpp
Pavel Rojtberg [Thu, 24 Nov 2016 12:56:08 +0000 (13:56 +0100)]
calibrateCamera: specifiy what kind of re-projection error is returned
Vitaly Tuzov [Thu, 24 Nov 2016 12:02:12 +0000 (15:02 +0300)]
Added wrappers for vx_matrix and vx_convolution (#7651)
Alexander Alekhin [Thu, 24 Nov 2016 10:34:17 +0000 (10:34 +0000)]
Merge pull request #7718 from guoxuedong:updateExifReader
guoxuedong [Thu, 24 Nov 2016 03:08:11 +0000 (11:08 +0800)]
Update exif.cpp
ExifReader::getExif may enter infinite loop with jpeg image which have no EOI.
For example, bytesToSkip may be set to 0 and fseek seems like fseek(f, -2 , SEEK_CUR) for image that end with RST7(FF D7) instead of EOI.
Alexander Alekhin [Wed, 23 Nov 2016 14:44:05 +0000 (17:44 +0300)]
Merge pull request #7714 from onixie:master
Fix docs explaining the returnPoints argument of convexHull.
Rostislav Vasilikhin [Wed, 23 Nov 2016 14:39:18 +0000 (18:39 +0400)]
fixed wrong equivalence in YUV conversion (#7481)
* fixed wrong equivalence in YUV conversion
* fixed channel order from YVU to YUV
Vadim Pisarevsky [Wed, 23 Nov 2016 14:15:02 +0000 (14:15 +0000)]
Merge pull request #7707 from K-Shinotsuka:issue37
monoid911 [Wed, 23 Nov 2016 11:43:52 +0000 (20:43 +0900)]
Fix docs explaining the returnPoints argument of convexHull.
Alexander Alekhin [Wed, 23 Nov 2016 10:43:08 +0000 (13:43 +0300)]
core: fix absdiff (non-optimized, fp) to prevent "-0" results
Alexander Alekhin [Tue, 22 Nov 2016 17:08:27 +0000 (20:08 +0300)]
java tests fixes
Alexander Alekhin [Tue, 22 Nov 2016 16:49:55 +0000 (19:49 +0300)]
java: skip test in case of missed classes from opencv_contrib
Alexander Alekhin [Tue, 15 Nov 2016 16:24:03 +0000 (19:24 +0300)]
java: fix testIsSubmatrix test
Alexander Alekhin [Tue, 15 Nov 2016 19:25:42 +0000 (22:25 +0300)]
calib3d: fix fisheye for test fisheyeTest.stereoCalibrateFixIntrinsic
Alexander Alekhin [Tue, 15 Nov 2016 19:40:55 +0000 (22:40 +0300)]
core: drop type/dims/rows/cols information in Mat::release()
Alexander Alekhin [Wed, 23 Nov 2016 10:50:30 +0000 (10:50 +0000)]
Merge pull request #7640 from alalek:fix_test_tvl1
k-shinotsuka [Wed, 23 Nov 2016 10:18:41 +0000 (19:18 +0900)]
add NEON code for RGB2Luv_f.
Alexander Alekhin [Tue, 22 Nov 2016 19:10:48 +0000 (19:10 +0000)]
Merge pull request #7163 from savuor:openvx_sample
Vadim Pisarevsky [Tue, 22 Nov 2016 13:57:13 +0000 (13:57 +0000)]
Merge pull request #7662 from K-Shinotsuka:issue34
Rostislav Vasilikhin [Tue, 22 Nov 2016 13:27:06 +0000 (16:27 +0300)]
hide info warnings
apavlenko [Mon, 21 Nov 2016 09:31:48 +0000 (12:31 +0300)]
hiding silly warnings
Rostislav Vasilikhin [Mon, 21 Nov 2016 07:50:29 +0000 (10:50 +0300)]
compilation fixed for C++98
Andrey Pavlenko [Fri, 18 Nov 2016 13:22:11 +0000 (16:22 +0300)]
warnings in ivx.hpp fixed
* fixing 1.0.1
* fixing build without c++11
* hiding warnings
Rostislav Vasilikhin [Fri, 18 Nov 2016 13:11:32 +0000 (16:11 +0300)]
"no previous declaration" warnings fixed in samples
Andrey Pavlenko [Fri, 18 Nov 2016 12:11:17 +0000 (15:11 +0300)]
build fixes for openvx-1.0.1 and c++03
* fixing 1.0.1
* fixing build without c++11
Rostislav Vasilikhin [Thu, 17 Nov 2016 16:42:58 +0000 (19:42 +0300)]
changed image creation from handle
apavlenko [Thu, 17 Nov 2016 16:27:14 +0000 (19:27 +0300)]
sugar for single plane images
Rostislav Vasilikhin [Thu, 17 Nov 2016 12:58:23 +0000 (15:58 +0300)]
minor fixes in samples and ivx.hpp
Rostislav Vasilikhin [Thu, 1 Sep 2016 11:34:34 +0000 (14:34 +0300)]
interoperability OpenVX samples added
Vadim Pisarevsky [Tue, 22 Nov 2016 13:49:14 +0000 (13:49 +0000)]
Merge pull request #7664 from K-Shinotsuka:issue35
Vadim Pisarevsky [Tue, 22 Nov 2016 13:44:42 +0000 (13:44 +0000)]
Merge pull request #7665 from K-Shinotsuka:issue36
Vadim Pisarevsky [Tue, 22 Nov 2016 13:39:20 +0000 (13:39 +0000)]
Merge pull request #7621 from terfendail:ovxhal_resizefix
Vadim Pisarevsky [Tue, 22 Nov 2016 13:35:43 +0000 (13:35 +0000)]
Merge pull request #7646 from alalek:cmake_ccache
Vadim Pisarevsky [Tue, 22 Nov 2016 13:31:26 +0000 (13:31 +0000)]
Merge pull request #7641 from alalek:warnings
Vadim Pisarevsky [Tue, 22 Nov 2016 12:40:22 +0000 (12:40 +0000)]
Merge pull request #7627 from sovrasov:fs_vec_of_vec
Vadim Pisarevsky [Tue, 22 Nov 2016 11:35:07 +0000 (11:35 +0000)]
Merge pull request #7688 from alalek:fix_orb_test
Vadim Pisarevsky [Tue, 22 Nov 2016 10:19:52 +0000 (10:19 +0000)]
Merge pull request #7699 from alalek:test_hog_eps
Alexander Alekhin [Tue, 22 Nov 2016 09:49:17 +0000 (09:49 +0000)]
Merge pull request #7702 from lupustr3:pvlasov/ipp2017_filter_fix
Pavel Vlasov [Mon, 21 Nov 2016 14:18:03 +0000 (17:18 +0300)]
IPP 2017 filter2D fix;
filter2D IPP runtime check;
Alexander Alekhin [Mon, 21 Nov 2016 11:20:25 +0000 (14:20 +0300)]
test: change EPS in Objdetect_HOGDetector tests
LaurentBerger [Mon, 21 Nov 2016 01:28:59 +0000 (02:28 +0100)]
Merge pull request #7691 from LaurentBerger:I7676
* Solve issue 7676
* I7676 bis
Arek [Mon, 21 Nov 2016 01:22:43 +0000 (02:22 +0100)]
Merge pull request #7695 from ArkadiuszRaj:aravis-release-bug-fix
* Aravis: fixing releasing object when no communication with camera is possible
* Removing unnecessary include
Alexander Alekhin [Thu, 17 Nov 2016 13:58:12 +0000 (16:58 +0300)]
perf: disabled checks for magic numbers
Results are not bit-exact
Alexander Alekhin [Thu, 17 Nov 2016 13:36:39 +0000 (16:36 +0300)]
features2d: update rotation invariance test
Don't process keypoints near border, because after rotation border points
have contrast gradients with rotation border fill.
Alexander Alekhin [Fri, 18 Nov 2016 12:19:00 +0000 (12:19 +0000)]
Merge pull request #7644 from alalek:cmake_python_install
Alexander Alekhin [Tue, 8 Nov 2016 15:00:03 +0000 (18:00 +0300)]
cmake: ccache
Alexander Alekhin [Thu, 17 Nov 2016 15:41:39 +0000 (18:41 +0300)]
warnings
Alexander Alekhin [Thu, 10 Nov 2016 12:55:47 +0000 (15:55 +0300)]
ffmpeg: check return value
Alexander Alekhin [Thu, 17 Nov 2016 17:54:25 +0000 (17:54 +0000)]
Merge pull request #7625 from souch55:Fixapi
Alexander Alekhin [Thu, 17 Nov 2016 09:40:27 +0000 (09:40 +0000)]
Merge pull request #7673 from pengli:deriv
Li Peng [Thu, 27 Oct 2016 06:52:26 +0000 (14:52 +0800)]
laplacian ocl kernel optimization
This ocl kernel is 46%~171% faster than current laplacian 3x3
ocl kernel in the perf test, with image format "CV_8UC1".
Signed-off-by: Li Peng <peng.li@intel.com>
Vitaly Tuzov [Wed, 16 Nov 2016 10:36:37 +0000 (13:36 +0300)]
Replaced magic constant in large image handling check
Andrey Pavlenko [Wed, 16 Nov 2016 14:05:08 +0000 (17:05 +0300)]
Merge pull request #7681 from apavlenko:ivx-docs
OpenVX C++ wrappers docs (#7681)
* adding wrapper headers to the project to see them in IDE project explorer
* readme
* doxygen: toplevel brief-s
* more doxygen
* more doxygen
Vladislav Sovrasov [Tue, 8 Nov 2016 11:04:17 +0000 (14:04 +0300)]
Add support of vector of vectors serialization to FileStorage
Alexander Alekhin [Tue, 15 Nov 2016 19:35:38 +0000 (19:35 +0000)]
Merge pull request #7678 from dec1:tag_3.0.0
Alexander Alekhin [Tue, 15 Nov 2016 15:59:27 +0000 (15:59 +0000)]
Merge pull request #7675 from filnet:msys2
k-shinotsuka [Sun, 13 Nov 2016 14:13:02 +0000 (23:13 +0900)]
add SSE code for RGB2HLS_f.
Declan Moran [Tue, 15 Nov 2016 14:02:10 +0000 (15:02 +0100)]
fix error when building android sample:
https://github.com/opencv/opencv/issues/7590
Philippe Renon [Mon, 14 Nov 2016 23:33:11 +0000 (00:33 +0100)]
calib3d: fix missing cv::redirectError symbol error
happens on msys2 with gcc 6.2.0
see also http://stackoverflow.com/questions/
38552221/undefined-reference-to-cvredirecterror-while-creating-shared-build-of-opencv-3
sourin [Wed, 9 Nov 2016 11:19:57 +0000 (16:49 +0530)]
Fixed with review comments.
Alexander Alekhin [Mon, 14 Nov 2016 13:40:04 +0000 (13:40 +0000)]
Merge pull request #7653 from pengli:deriv
Alexander Alekhin [Mon, 14 Nov 2016 13:38:44 +0000 (13:38 +0000)]
Merge pull request #7643 from alalek:libpng_aarch64
Alexander Alekhin [Mon, 14 Nov 2016 13:37:41 +0000 (13:37 +0000)]
Merge pull request #7638 from ArkadiuszRaj:move-exif-from-jpeg-to-loadsave
Juha Reunanen [Mon, 14 Nov 2016 13:31:29 +0000 (15:31 +0200)]
Merge pull request #7464 from reunanen:test-7409-7458
Change contour test images to be very wide (#7464)
* Change contour test images to be very wide (#7409, #7458)
Unfortunately, slows down the tests.
* Decrease the number of contour test cases, in order to (at least partially) offset the test run duration increase caused by making the test images wider
* Don't test with very wide images on 32-bit architectures
Li Peng [Thu, 27 Oct 2016 04:41:40 +0000 (12:41 +0800)]
sobel and scharr ocl kernel optimization
It improves 108%~230% performance in the perf test
with image format "CV_8UC1" and kernel size 3.
Signed-off-by: Li Peng <peng.li@intel.com>
k-shinotsuka [Sun, 13 Nov 2016 06:24:00 +0000 (15:24 +0900)]
add SSE code for HLS2RGB_f.
k-shinotsuka [Sat, 12 Nov 2016 10:02:43 +0000 (19:02 +0900)]
add SSE code for HSV2RGB_f.
Vadim Pisarevsky [Fri, 11 Nov 2016 20:37:38 +0000 (20:37 +0000)]
Merge pull request #7656 from alalek:python_warning
Vadim Pisarevsky [Fri, 11 Nov 2016 20:35:24 +0000 (20:35 +0000)]
Merge pull request #7654 from souch55:Fixgtk
Vadim Pisarevsky [Fri, 11 Nov 2016 20:30:24 +0000 (20:30 +0000)]
Merge pull request #7539 from Tetragramm:90DegreeRotations
Vadim Pisarevsky [Fri, 11 Nov 2016 20:27:41 +0000 (20:27 +0000)]
Merge pull request #7645 from alalek:cmake_arm_toolchain
Vadim Pisarevsky [Fri, 11 Nov 2016 15:06:56 +0000 (15:06 +0000)]
Merge pull request #7642 from alalek:plantuml
Alexander Alekhin [Fri, 11 Nov 2016 14:45:12 +0000 (17:45 +0300)]
build: fix warning in python parser
/usr/lib/python3.5/re.py:203: FutureWarning: split() requires a non-empty pattern match.
return _compile(pattern, flags).split(string, maxsplit)
Vadim Pisarevsky [Fri, 11 Nov 2016 14:13:18 +0000 (14:13 +0000)]
Merge pull request #7634 from paroj:undistpt
Vadim Pisarevsky [Fri, 11 Nov 2016 14:09:11 +0000 (14:09 +0000)]
Merge pull request #7648 from alalek:cmake_warnings
sourin [Fri, 11 Nov 2016 05:25:38 +0000 (10:55 +0530)]
Fix error gtk