Alexander Alekhin [Mon, 15 Oct 2018 12:07:40 +0000 (12:07 +0000)]
Merge pull request #12516 from seiko2plus:changeUnvMultiply16
tellowkrinkle [Sun, 14 Oct 2018 18:18:05 +0000 (13:18 -0500)]
Merge pull request #12697 from tellowkrinkle:FasterCocoaWindows
* Make cocoa windows draw faster
* Use a CALayer for rendering when possible Uses GPU to scale images, which is important because retina macs will want window sizes much larger (in pixels) than the image
* Fix mouse logic for cocoa windows
* Only halve resolution on retina if image is larger than display
Alexander Alekhin [Sun, 14 Oct 2018 13:33:23 +0000 (13:33 +0000)]
Merge pull request #12823 from tompollok:3.4
tompollok [Sat, 13 Oct 2018 19:35:10 +0000 (21:35 +0200)]
change area() emptiness checks to empty()
Alexander Alekhin [Sat, 13 Oct 2018 13:20:33 +0000 (13:20 +0000)]
Merge pull request #12805 from drkoller:imwritedoc
Alexander Alekhin [Sat, 13 Oct 2018 13:12:55 +0000 (13:12 +0000)]
Merge pull request #12815 from alalek:issue_12812
Alexander Alekhin [Sat, 13 Oct 2018 12:28:08 +0000 (12:28 +0000)]
Merge pull request #12641 from dkurt:dnn_samples_args_autofill
Alexander Alekhin [Sat, 13 Oct 2018 12:25:35 +0000 (12:25 +0000)]
Merge pull request #12802 from alalek:cmake_pdb
drkoller [Thu, 11 Oct 2018 20:27:13 +0000 (16:27 -0400)]
Clean up documentation for imread and imwrite
Alexander Alekhin [Fri, 12 Oct 2018 19:10:06 +0000 (19:10 +0000)]
core: re-throw allocation exception if there is no fallback
Alexander Alekhin [Fri, 12 Oct 2018 16:59:25 +0000 (16:59 +0000)]
Merge pull request #12811 from take1014:resize_large_image
Alexander Alekhin [Fri, 12 Oct 2018 16:59:01 +0000 (16:59 +0000)]
Merge pull request #12806 from seiko2plus:imgprocPyramidsVSX
take1014 [Fri, 12 Oct 2018 14:08:03 +0000 (23:08 +0900)]
resolves 11283
Sayed Adel [Thu, 11 Oct 2018 23:03:57 +0000 (23:03 +0000)]
imgproc: Enable VSX on pyrDown & pyrUp
Alexander Alekhin [Thu, 11 Oct 2018 19:30:49 +0000 (19:30 +0000)]
Merge pull request #12803 from alalek:samples_use_cmake_from_msvs
Alexander Alekhin [Thu, 11 Oct 2018 19:29:04 +0000 (19:29 +0000)]
Merge pull request #12796 from alalek:fix_openmp_performance
Alexander Alekhin [Thu, 11 Oct 2018 19:28:29 +0000 (19:28 +0000)]
Merge pull request #12791 from alalek:win32_dllmain_detect_termination
Alexander Alekhin [Thu, 11 Oct 2018 19:26:36 +0000 (19:26 +0000)]
Merge pull request #12784 from terfendail:pyramids_wintr
Alexander Alekhin [Thu, 11 Oct 2018 19:26:17 +0000 (19:26 +0000)]
Merge pull request #12748 from terfendail:resize_wintr2
Alexander Alekhin [Thu, 11 Oct 2018 16:32:11 +0000 (16:32 +0000)]
Merge pull request #12792 from tomoaki0705:fixCudaImgprocColorGray
Alexander Alekhin [Thu, 11 Oct 2018 16:15:52 +0000 (19:15 +0300)]
samples: prefer CMake from MSVS
MSVS is shipped with CMake compoment.
Alexander Alekhin [Wed, 10 Oct 2018 16:16:32 +0000 (19:16 +0300)]
cmake: improve PDB support
Tomoaki Teshima [Thu, 11 Oct 2018 09:45:20 +0000 (18:45 +0900)]
update the threshold
* fix test failure of cudaimgproc convert color bgr2lab series
Alexander Alekhin [Thu, 11 Oct 2018 09:42:05 +0000 (12:42 +0300)]
Merge pull request #12521 from allnes:detect_qr_code
Objdetect: QRCode detect + decode
Sayed Adel [Tue, 2 Oct 2018 18:44:03 +0000 (20:44 +0200)]
imgproc:simd Enable VSX and wide universal intrinsics for accumulate operations
- improve cpu dispatching calls to allow more SIMD extentions
(SSE4.1, AVX2, VSX)
- wide universal intrinsics
- replace dummy v_expand with v_expand_low
- replace v_expand + v_mul_wrap with v_mul_expand for product accumulate operations
- use FMA for accumulate operations
- add mask and more types to accumulate's performance tests
Sayed Adel [Wed, 12 Sep 2018 19:32:34 +0000 (21:32 +0200)]
Change behaviour of 16-bit multiply operator
- redefine 16-bit multiply operator to perform saturating multiply
instead of non-saturating multiply
- implement 8-bit multiply operator to perform saturating multiply
- implement v_mul_wrap() for 8-bit, 16-bit non-saturating multiply
- improve performance of v_mul_hi() for VSX
- update intrin tests with new changes
- replace unv 16-bit multiplication operator with v_mul_wrap due behavior changes
- Several improvements depend on vpisarev review
* initial forward declarations for universal intrinsics
* move emulating SSE intrinsics into separate file
* implement v_mul_expand for 8-bit
* reimplement saturating multiply using v_mul_expand + v_pack
* map v_expand, v_load_expand, v_load_expand_q to sse4.1
* fix overflow avx2::v_pack(uint32)
* implement two universal intrinsics v_expand_low and v_expand_high
Vitaly Tuzov [Mon, 10 Sep 2018 17:05:45 +0000 (20:05 +0300)]
Fixed meanStdDev() implementation for the case input matrix has more than 4 channels
Alexander Alekhin [Thu, 11 Oct 2018 00:49:45 +0000 (00:49 +0000)]
js: enable calib3d code (no bindings)
Alexander Alekhin [Wed, 10 Oct 2018 20:22:07 +0000 (20:22 +0000)]
Merge pull request #12794 from alalek:fix_warnings
Alexander Alekhin [Wed, 10 Oct 2018 20:03:02 +0000 (20:03 +0000)]
Merge pull request #12793 from alalek:cmake_fix_python_standalone
Alexander Alekhin [Wed, 10 Oct 2018 19:57:43 +0000 (19:57 +0000)]
core: call omp_set_dynamic() for better CPU usage
Similar to 'OMP_DYNAMIC=TRUE'.
Alexander Alekhin [Wed, 10 Oct 2018 19:26:19 +0000 (19:26 +0000)]
fix build warnings
Vitaly Tuzov [Tue, 9 Oct 2018 17:37:11 +0000 (20:37 +0300)]
pyrDown and pyrUp SSE2 implementations replaced with wide universal intrinsics implementations
Alexander Alekhin [Wed, 10 Oct 2018 18:04:39 +0000 (18:04 +0000)]
ocl(win32): bypass deallocate() during process termination
Alexander Alekhin [Wed, 10 Oct 2018 13:03:57 +0000 (16:03 +0300)]
cmake: fix Python detection for standalone bindings
Alexander Nesterov [Fri, 17 Aug 2018 16:01:02 +0000 (13:01 -0300)]
Added QR code decoding.
Alexander Alekhin [Wed, 10 Oct 2018 10:53:46 +0000 (13:53 +0300)]
cmake: add DllMain() into each OpenCV DLL
to detect process termination after ExitProcess() call
Alexander Alekhin [Wed, 10 Oct 2018 09:18:32 +0000 (09:18 +0000)]
Merge pull request #12788 from alalek:samples_winpack_support_buildtools
Alexander Alekhin [Wed, 10 Oct 2018 00:34:34 +0000 (00:34 +0000)]
samples: add search for MSVS 2017 BuildTools
Alexander Alekhin [Tue, 9 Oct 2018 16:01:29 +0000 (19:01 +0300)]
Merge pull request #12636 from alalek:winpack_samples_cpp_build_script
* samples: add winpack "drap & drop" build script
* samples: add search for MSVS 2017 Enterprise
Hamdi Sahloul [Tue, 9 Oct 2018 14:39:39 +0000 (23:39 +0900)]
CUDA/BgSegm: fix the threshold of MOG2.Update test when detectShadow=true (#12762)
Apoorv Goel [Tue, 9 Oct 2018 12:49:17 +0000 (18:19 +0530)]
Merge pull request #12777 from UnderscoreAsterisk:document-cvtColorTwoPlane
* Add documentation for cvtColorTwoPlane
* Change brief and add links
Lubov Batanina [Mon, 8 Oct 2018 19:18:41 +0000 (22:18 +0300)]
Merge pull request #12596 from l-bat:l-bat/shufflenet_onnx
* Add Shufflenet support in ONNX
* Add test for transpose layer
Alexander Alekhin [Mon, 8 Oct 2018 19:15:54 +0000 (19:15 +0000)]
Merge pull request #12765 from cv3d:ts/cuda/device_name
Alexander Alekhin [Mon, 8 Oct 2018 15:30:16 +0000 (15:30 +0000)]
Merge pull request #12755 from alalek:fix_allocSingleton
Alexander Alekhin [Mon, 8 Oct 2018 15:22:23 +0000 (15:22 +0000)]
Merge pull request #12754 from alalek:dnn_ocl4dnn_async_expressions
Alexander Alekhin [Mon, 8 Oct 2018 15:21:50 +0000 (15:21 +0000)]
Merge pull request #12752 from alalek:dnn_cleanup_getUMat
Alexander Alekhin [Mon, 8 Oct 2018 14:45:50 +0000 (17:45 +0300)]
Merge pull request #12725 from alalek:cmake_python_win32
* cmake: don't ignore Python from PATH environment variable
- this breaks selection between 32/64-bit Python
- this breaks Anaconda/Conda environments
- it is not the CMake default behavior, expected by many projects
* cmake: add Python version check, fallback path on CMake 3.12+
* cmake: drop Python 2.6, allow version selection for Python 3.x
Alexander Alekhin [Mon, 8 Oct 2018 14:36:32 +0000 (14:36 +0000)]
Merge pull request #12767 from dkurt:fix_js_features2d
Vitaly Tuzov [Tue, 25 Sep 2018 11:14:21 +0000 (14:14 +0300)]
Replaced SSE2 area resize implementation with wide universal intrinsic implementation
Dmitry Kurtaev [Sun, 7 Oct 2018 19:55:38 +0000 (22:55 +0300)]
Fix JavaScript bindings for features2d module
Hamdi Sahloul [Sun, 7 Oct 2018 17:51:45 +0000 (02:51 +0900)]
TS/CUDA: Show device ID in case of multiple devices
Alexander Alekhin [Fri, 5 Oct 2018 16:03:34 +0000 (16:03 +0000)]
Merge pull request #12753 from mshabunin:fix-ffmpeg-dl
Alexander Alekhin [Fri, 5 Oct 2018 15:23:05 +0000 (18:23 +0300)]
core: update allocSingleton implementation, valgrind suppression
Maksim Shabunin [Fri, 5 Oct 2018 13:38:44 +0000 (16:38 +0300)]
FFmpeg: correct licenses install location
Alexander Alekhin [Fri, 5 Oct 2018 14:06:50 +0000 (17:06 +0300)]
dnn: don't use Mat expressions with async UMat functions
Alexander Alekhin [Fri, 5 Oct 2018 12:10:58 +0000 (15:10 +0300)]
dnn(ocl4dnn): don't use getUMat()
especially in CPU only processing
Alexander Alekhin [Fri, 5 Oct 2018 10:20:12 +0000 (10:20 +0000)]
Merge pull request #12744 from alalek:issue_12736
Alexander Alekhin [Fri, 5 Oct 2018 09:28:07 +0000 (09:28 +0000)]
Merge pull request #12749 from powderluv:fix-clang-cl-tzcnt
Anush Elangovan [Thu, 4 Oct 2018 17:48:14 +0000 (10:48 -0700)]
_tzcnt_u32() is undefined in clang-cl so use alternate impl
_tzcnt_u32() is not exported by clang-cl intrin.h so check for
clang-cl and enable an alterate for _tzcnt_u32()
Some discussions:
http://lists.llvm.org/pipermail/cfe-dev/2016-October/051329.html
https://bugs.llvm.org/show_bug.cgi?id=30506
TEST=Build with clang-cl
Alexander Alekhin [Thu, 4 Oct 2018 16:42:16 +0000 (16:42 +0000)]
Merge pull request #12745 from savuor:fix/ocl_arg_local_size
Rostislav Vasilikhin [Thu, 4 Oct 2018 11:26:32 +0000 (14:26 +0300)]
ocl::KernelArg::Local(): added size argument
Alexander Alekhin [Thu, 4 Oct 2018 14:17:22 +0000 (14:17 +0000)]
Merge pull request #12741 from GeorgeMironov:fix-cmake-protobuf
Alexander Alekhin [Thu, 4 Oct 2018 11:43:43 +0000 (14:43 +0300)]
cmake: define CV_ErrorNoReturn under CV_STATIC_ANALYSIS
to avoid build break without `__OPENCV_BUILD`
Alexander Alekhin [Thu, 4 Oct 2018 11:33:34 +0000 (11:33 +0000)]
Merge pull request #12735 from tomoaki0705:fixCudaStereoTestFailure
Alexander Alekhin [Thu, 4 Oct 2018 11:31:47 +0000 (11:31 +0000)]
Merge pull request #12726 from alalek:dnn_ocl_conv_avoid_multiple_expr_evaluation
Georgy Mironov [Thu, 4 Oct 2018 10:48:40 +0000 (13:48 +0300)]
Fix install with external protobuf
Tomoaki Teshima [Wed, 3 Oct 2018 23:35:34 +0000 (08:35 +0900)]
fix test failure of StereoBeliefPropagation
* if the src has odd number of height, access error happens
* it could happen on width, too
* check both dst and src range in both width and height
Apoorv Goel [Wed, 3 Oct 2018 12:50:04 +0000 (18:20 +0530)]
Merge pull request #12731 from UnderscoreAsterisk:fix-12012
* fix #12012
* Replace CV_StsBadFlag with Error::StsBadFlag
Alexander Alekhin [Wed, 3 Oct 2018 07:21:06 +0000 (07:21 +0000)]
Merge pull request #12727 from rnk:fix-clang-cl-cmake
Alexander Alekhin [Wed, 3 Oct 2018 06:14:43 +0000 (06:14 +0000)]
Merge pull request #12693 from peters:patch-1
Reid Kleckner [Tue, 2 Oct 2018 21:30:52 +0000 (14:30 -0700)]
Fix CMake for clang-cl
Avoids passing -Wall, which will enable literally every warning in
Clang. Avoids passing -ffunction-sections etc to clang-cl, which each
generate warnings.
Alexander Alekhin [Tue, 2 Oct 2018 21:10:52 +0000 (21:10 +0000)]
dnn(ocl4dnn): calculate activation expression once
- to avoid multiple conditional calls via sub_group() functions
Alexander Alekhin [Tue, 2 Oct 2018 20:52:34 +0000 (20:52 +0000)]
Merge pull request #12722 from tomoaki0705:fixCudaFiltersFailure
Peter Rekdal Sunde [Sat, 29 Sep 2018 17:43:01 +0000 (19:43 +0200)]
Obtain fourcc when `AVStream` is network stream
The `codec_tag` is only available when opening a file from disk. If `AVStream` is a network stream then `fourcc` must be obtained using `codec_id`. I have tested the following scenarios:
1) Open a `.mp4` file and verify that `codec_tag` is returned (old behavior)
2) Open a `rtsp` stream and verify that `codec_fourcc` is returned (Tested with a MJPEG, H264 and H265 stream)
Alexander Alekhin [Tue, 2 Oct 2018 16:40:13 +0000 (16:40 +0000)]
Merge pull request #12605 from terfendail:integral_wintr
Alexander Alekhin [Tue, 2 Oct 2018 16:39:50 +0000 (16:39 +0000)]
Merge pull request #12581 from terfendail:lapack_intr
Tomoaki Teshima [Tue, 2 Oct 2018 15:59:08 +0000 (00:59 +0900)]
fix test failure of cudafilters Median_Accuracy
* avoid race condition
Vitaly Tuzov [Tue, 18 Sep 2018 09:22:34 +0000 (12:22 +0300)]
SSE2 code in invert() replaced with universal intrinsics
Suleyman TURKMEN [Mon, 1 Oct 2018 19:22:37 +0000 (22:22 +0300)]
Merge pull request #12649 from sturkmen72:patch-9
Update hog.cpp (#12649)
Alexander Alekhin [Mon, 1 Oct 2018 15:57:57 +0000 (15:57 +0000)]
Merge pull request #12669 from mshabunin:add-paths-config
Alexander Alekhin [Mon, 1 Oct 2018 15:57:31 +0000 (15:57 +0000)]
Merge pull request #12620 from alalek:ocl_extra_build_options
Alexander Alekhin [Mon, 1 Oct 2018 15:36:49 +0000 (15:36 +0000)]
Merge pull request #12711 from alalek:dnn_ocl_idlf_drop_weights_buf
Alexander Alekhin [Fri, 21 Sep 2018 15:04:18 +0000 (15:04 +0000)]
ocl: OPENCV_OPENCL_BUILD_EXTRA_OPTIONS parameter
Maksim Shabunin [Thu, 27 Sep 2018 12:52:42 +0000 (15:52 +0300)]
Added support for multi-path configuration parameter (env)
Vitaly Tuzov [Fri, 21 Sep 2018 13:21:40 +0000 (16:21 +0300)]
integral() implementation updated to utilize wide universal intrinsics
Alexander Alekhin [Mon, 1 Oct 2018 13:35:03 +0000 (13:35 +0000)]
Merge pull request #12714 from antonioborondo:fix_documentation_cuda_compare
Alexander Alekhin [Mon, 1 Oct 2018 13:32:51 +0000 (13:32 +0000)]
Merge pull request #12709 from WydD:3.4
Alexander Alekhin [Mon, 1 Oct 2018 13:32:29 +0000 (13:32 +0000)]
Merge pull request #12708 from ldevulder:protobuf-fix-compilation-issue-on-s390
Antonio Borondo [Mon, 1 Oct 2018 12:23:13 +0000 (13:23 +0100)]
Fix documentation of cv::cuda::compare
Alexander Alekhin [Sun, 30 Sep 2018 20:59:48 +0000 (23:59 +0300)]
Merge pull request #12705 from berak:imread_document_grayscale
Alexander Alekhin [Sun, 30 Sep 2018 20:22:39 +0000 (20:22 +0000)]
dnn(ocl4dnn): drop weights_buf
- avoid memory access violation during "prefetch" stage
Loic Petit [Sun, 30 Sep 2018 20:24:02 +0000 (22:24 +0200)]
Fix frame rate rounding in ffmpeg wrapper (#9023)
Loic Devulder [Fri, 28 Sep 2018 13:33:18 +0000 (15:33 +0200)]
3rdparty/protobuf: fix compilation issue on s390
This commit fixes an issue while trying to compile
on s390x architecture.
This is simply a backport of a fixe already applied
in official protobuf code:
- https://github.com/protocolbuffers/protobuf/pull/3955
berak [Sun, 30 Sep 2018 13:22:26 +0000 (15:22 +0200)]
document imread grayscale behaviour
Hamdi Sahloul [Sat, 29 Sep 2018 17:03:54 +0000 (02:03 +0900)]
Merge pull request #12667 from cv3d:fix/ts_report
TS: fix Python v2/v3 compatibility (#12667)
* TS: fix Python2.7 compatibility
* TS: fix Python3 compatibility
* py3: use integer division '/' => '//' instead of cast
Alexander Alekhin [Sat, 29 Sep 2018 16:39:04 +0000 (16:39 +0000)]
Merge pull request #12690 from cv3d:fix/ts_python_exe
Alexander Alekhin [Sat, 29 Sep 2018 16:36:25 +0000 (16:36 +0000)]
Merge pull request #12666 from tomoaki0705:cudevTestFailureLuvRGB
Hamdi Sahloul [Sat, 29 Sep 2018 10:34:44 +0000 (19:34 +0900)]
Utilize the currently running Python executable
especially if it matches the module being tested
Tomoaki Teshima [Sat, 29 Sep 2018 14:13:12 +0000 (23:13 +0900)]
fix test failure of cudev
* follow the implementation of Luv2RGBfloat in imgproc/src/color_lab.cpp
* loosen threshold in cudaimgproc