Alexander Alekhin [Mon, 18 Feb 2019 14:50:03 +0000 (14:50 +0000)]
Merge pull request #13850 from nglee:dev_FixCudaStereoBMHangRace
Alexander Alekhin [Mon, 18 Feb 2019 14:39:10 +0000 (14:39 +0000)]
Merge pull request #13835 from catree:real_time_pose_tutorial_keypoints_matching
Alexander Alekhin [Mon, 18 Feb 2019 13:38:44 +0000 (13:38 +0000)]
Merge pull request #13833 from alalek:core_dispatch_matmul
catree [Thu, 14 Feb 2019 17:50:01 +0000 (18:50 +0100)]
Fix arguments parsing. Add possibility to choose between different features type. Add keypoints matching visualization. Auto format code.
Alexander Alekhin [Thu, 14 Feb 2019 13:00:17 +0000 (16:00 +0300)]
core: dispatch matmul
- gemm: keep baseline only (lapack is 10x+ faster, lets reduce binary size)
- transform / distTransform
- scaleAdd (32f/64f only)
- Mahalanobis: keep baseline only (no perf tests)
- mulTransposed: keep baseline only (no perf tests)
- dot
Namgoo Lee [Mon, 18 Feb 2019 03:10:05 +0000 (03:10 +0000)]
cuda::StereoBM - fix hanging and racing issue
- Fix hanging issue on 2080 Ti
- Fix racing issue
Alexander Alekhin [Fri, 15 Feb 2019 13:54:24 +0000 (13:54 +0000)]
Merge pull request #13774 from l-bat:fix-IE-tests
Alexander Alekhin [Thu, 14 Feb 2019 16:07:41 +0000 (19:07 +0300)]
core: keep history of matmul.cpp
Alexander Alekhin [Thu, 14 Feb 2019 12:10:11 +0000 (15:10 +0300)]
core: clone matmul.dispatch.cpp
Alexander Alekhin [Thu, 14 Feb 2019 12:09:28 +0000 (15:09 +0300)]
core: clone matmul.simd.hpp
Alexander Alekhin [Thu, 14 Feb 2019 15:52:20 +0000 (15:52 +0000)]
Merge pull request #13823 from alalek:core_dispatch_countNonZero
Alexander Alekhin [Thu, 14 Feb 2019 11:40:27 +0000 (11:40 +0000)]
Merge pull request #13726 from nicolaerosia:android-ffmpeg
Nicolae Rosia [Wed, 30 Jan 2019 15:43:23 +0000 (17:43 +0200)]
cmake: allow FFmpeg linking on Android
details: https://github.com/opencv/opencv/pull/13726
In order to help CMake and PkgConfig find FFmpeg you need to:
1) export `PKG_CONFIG_LIBDIR=${SYSROOT}`
where `SYSROOT=${ANDROID_NDK}/sysroot`
2) Add `-DCMAKE_PREFIX_PATH=${FFMPEG_INSTALL_PREFIX}`
3) Append `-DCMAKE_FIND_ROOT_PATH=${FFMPEG_INSTALL_PREFIX}`
Alexander Alekhin [Wed, 13 Feb 2019 15:27:55 +0000 (18:27 +0300)]
core: dispatch count_non_zero
Alexander Alekhin [Thu, 14 Feb 2019 10:15:43 +0000 (13:15 +0300)]
core: keep history of count_non_zero.cpp
Alexander Alekhin [Wed, 13 Feb 2019 12:55:42 +0000 (15:55 +0300)]
core: clone count_non_zero.dispatch.cpp
Alexander Alekhin [Wed, 13 Feb 2019 12:55:02 +0000 (15:55 +0300)]
core: clone count_non_zero.simd.hpp
Liubov Batanina [Wed, 13 Feb 2019 12:59:51 +0000 (15:59 +0300)]
Changed condition for resize and lrn layers
Alexander Alekhin [Thu, 14 Feb 2019 09:36:10 +0000 (09:36 +0000)]
Merge pull request #13808 from tailsu:sd/libtiff-4.0.10
Stefan Dragnev [Tue, 12 Feb 2019 13:43:58 +0000 (14:43 +0100)]
update to libtiff-4.0.10
Alexander Alekhin [Thu, 14 Feb 2019 08:16:42 +0000 (08:16 +0000)]
Merge pull request #13796 from alalek:core_dispatch_sum
Alexander Alekhin [Wed, 13 Feb 2019 16:19:16 +0000 (16:19 +0000)]
Merge pull request #13807 from tailsu:sd/libpng-1.6.36
Alexander Alekhin [Mon, 11 Feb 2019 13:21:49 +0000 (16:21 +0300)]
core: dispatch sum
Stefan Dragnev [Tue, 12 Feb 2019 12:22:13 +0000 (13:22 +0100)]
update libpng to 1.6.36
* re-enabled NEON routines for arm64
Alexander Alekhin [Wed, 13 Feb 2019 10:49:36 +0000 (13:49 +0300)]
core: keep history of sum.cpp
Alexander Alekhin [Mon, 11 Feb 2019 13:14:43 +0000 (16:14 +0300)]
core: clone sum.dispatch.cpp
Alexander Alekhin [Mon, 11 Feb 2019 13:13:55 +0000 (16:13 +0300)]
core: clone sum.simd.hpp
Alexander Alekhin [Tue, 12 Feb 2019 21:50:29 +0000 (21:50 +0000)]
Merge pull request #13810 from nglee:dev_CudaSafeBlockScan
Namgoo Lee [Tue, 12 Feb 2019 08:24:57 +0000 (08:24 +0000)]
cudalegacy: Use safe block scan function
Alexander Alekhin [Tue, 12 Feb 2019 14:25:21 +0000 (14:25 +0000)]
Merge pull request #13803 from thangktran:thangktran/fix-misleading-runtime-assertion-calibrateCamera
Alexander Alekhin [Tue, 12 Feb 2019 14:07:42 +0000 (14:07 +0000)]
Merge pull request #13799 from dkurt:dnn_ie_future_2
Thang Tran [Mon, 11 Feb 2019 21:07:30 +0000 (22:07 +0100)]
calib3d: Fix calibrateCamera() misleading error
objectPoints and imagePoints are not checked whether they're empty and
cause checkVector() to fail, thus result in a wrong error message.
Fixes: https://github.com/opencv/opencv/issues/6002
Alexander Alekhin [Tue, 12 Feb 2019 13:02:22 +0000 (13:02 +0000)]
Merge pull request #13800 from dkurt:fix_nvidia_caffe
Alexander Alekhin [Tue, 12 Feb 2019 12:08:30 +0000 (12:08 +0000)]
Merge pull request #13780 from alalek:core_dispatch_convertTo
Dmitry Kurtaev [Mon, 11 Feb 2019 17:08:17 +0000 (20:08 +0300)]
Read raw floats data from Caffe models
Dmitry Kurtaev [Mon, 11 Feb 2019 14:13:39 +0000 (17:13 +0300)]
Fix Intel's Inference Engine backend from future. Second try.
Alexander Alekhin [Mon, 11 Feb 2019 14:05:43 +0000 (14:05 +0000)]
Merge pull request #13794 from alalek:backport_13786
Alexander Alekhin [Mon, 11 Feb 2019 12:39:35 +0000 (12:39 +0000)]
Merge pull request #13793 from vertextau:tutorial-fix
klemens [Sat, 9 Feb 2019 21:29:54 +0000 (22:29 +0100)]
Liubov Batanina [Mon, 11 Feb 2019 09:39:28 +0000 (12:39 +0300)]
Enabled tests on IE backend
Oleh [Mon, 11 Feb 2019 08:45:16 +0000 (10:45 +0200)]
Update video_input_psnr_ssim.markdown
Alexander Alekhin [Sat, 9 Feb 2019 15:26:11 +0000 (15:26 +0000)]
Merge pull request #13782 from paroj:pypath
Alexander Alekhin [Sat, 9 Feb 2019 15:25:48 +0000 (15:25 +0000)]
Merge pull request #13783 from alalek:fix_13741
Pavel Rojtberg [Fri, 8 Feb 2019 16:24:31 +0000 (17:24 +0100)]
python: sys.path[0] is reserved for the main script
Alexander Alekhin [Fri, 8 Feb 2019 14:56:25 +0000 (17:56 +0300)]
core: dispatch convert_scale
Alexander Alekhin [Fri, 8 Feb 2019 14:37:59 +0000 (17:37 +0300)]
core: dispatch convert
Alexander Alekhin [Fri, 8 Feb 2019 15:29:34 +0000 (15:29 +0000)]
Merge pull request #13769 from savuor:cvtColor_tests_16u_32f
Alexander Alekhin [Fri, 8 Feb 2019 15:23:46 +0000 (15:23 +0000)]
Merge pull request #13766 from dkurt:dnn_mask_rcnn_atrous
Alexander Alekhin [Fri, 8 Feb 2019 15:22:17 +0000 (15:22 +0000)]
Merge pull request #13760 from AnastasiaaSenina:fix-bug
Alexander Alekhin [Fri, 8 Feb 2019 13:39:13 +0000 (16:39 +0300)]
core: keep history of convert_scale.cpp
Alexander Alekhin [Fri, 8 Feb 2019 13:38:50 +0000 (16:38 +0300)]
core: clone convert_scale.dispatch.cpp
Alexander Alekhin [Fri, 8 Feb 2019 13:35:52 +0000 (16:35 +0300)]
core: clone convert_scale.simd.hpp
Alexander Alekhin [Fri, 8 Feb 2019 13:29:54 +0000 (16:29 +0300)]
core: keep history of convert.cpp
Alexander Alekhin [Fri, 8 Feb 2019 13:29:16 +0000 (16:29 +0300)]
core: clone convert.dispatch.cpp
Alexander Alekhin [Fri, 8 Feb 2019 13:27:59 +0000 (16:27 +0300)]
core: clone convert.simd.hpp
Alexander Alekhin [Fri, 8 Feb 2019 10:18:43 +0000 (10:18 +0000)]
Merge pull request #13746 from allnes:fix_step_dnn
Alexander Alekhin [Fri, 8 Feb 2019 10:13:11 +0000 (10:13 +0000)]
Merge pull request #13764 from nglee:dev_CudaCLAHE16bitSupport
AnastasiaaSenina [Thu, 7 Feb 2019 14:50:55 +0000 (17:50 +0300)]
fixed bug: added threshold for variables 'rotate_a', ' rotate_c'
Rostislav Vasilikhin [Thu, 7 Feb 2019 16:26:36 +0000 (19:26 +0300)]
disabled 16u and 32f perf tests
Alexander Alekhin [Thu, 7 Feb 2019 15:53:58 +0000 (15:53 +0000)]
Merge pull request #13775 from terfendail:morph_fix
Rostislav Vasilikhin [Thu, 7 Feb 2019 15:20:29 +0000 (18:20 +0300)]
disabled sanity check for 32f
Vitaly Tuzov [Thu, 7 Feb 2019 12:58:34 +0000 (15:58 +0300)]
Fixed out of bound reading issue in erode() and dilate()
Alexander Alekhin [Thu, 7 Feb 2019 13:26:17 +0000 (16:26 +0300)]
core(ocl): fix log messages
Alexander Nesterov [Fri, 1 Feb 2019 17:23:51 +0000 (16:23 -0100)]
Fix change step
Liubov Batanina [Thu, 7 Feb 2019 08:52:27 +0000 (11:52 +0300)]
Using IE backend for normalize layer tests
Namgoo Lee [Tue, 5 Feb 2019 16:37:33 +0000 (16:37 +0000)]
Add CV_16UC1 support for cuda::CLAHE
Due to size limit of shared memory, histogram is built on
the global memory for CV_16UC1 case.
The amount of memory needed for building histogram is:
65536 * 4byte = 256KB
and shared memory limit is 48KB typically.
Added test cases for CV_16UC1 and various clip limits.
Added perf tests for CV_16UC1 on both CPU and CUDA code.
There was also a bug in CV_8UC1 case when redistributing
"residual" clipped pixels. Adding the test case where clip
limit is 5.0 exposes this bug.
Rostislav Vasilikhin [Wed, 6 Feb 2019 14:56:44 +0000 (17:56 +0300)]
perf tests for cvtColor for 16U and 32f added
Alexander Alekhin [Wed, 6 Feb 2019 12:09:58 +0000 (12:09 +0000)]
Merge pull request #13738 from dkurt:dnn_ie_lock_shared_plugins
Dmitry Kurtaev [Wed, 6 Feb 2019 10:05:11 +0000 (13:05 +0300)]
Enable ResNet-based Mask-RCNN models from TensorFlow Object Detection API
Dmitry Kurtaev [Thu, 31 Jan 2019 13:10:59 +0000 (16:10 +0300)]
Add a mutex for shared Inference Engine plugins
Alexander Alekhin [Tue, 5 Feb 2019 14:58:27 +0000 (14:58 +0000)]
Merge pull request #13762 from alalek:ocl_dnn_mvn_local_size
Alexander Alekhin [Tue, 5 Feb 2019 12:27:29 +0000 (15:27 +0300)]
dnn(ocl): use compile-time LOCAL_SIZE parameter
instead of get_local_size(0) and dynamic local memory allocation
Alexander Alekhin [Tue, 5 Feb 2019 12:37:53 +0000 (12:37 +0000)]
Merge pull request #13670 from allnes:dnn_fix_obj_detect_sample
winice [Mon, 4 Feb 2019 15:59:00 +0000 (23:59 +0800)]
Merge pull request #13737 from winice-test:master
* Optical Flow rework to use wide universal intrinsics
* remove if (useSIMD) check as review requested
Alexander Alekhin [Sat, 2 Feb 2019 14:22:02 +0000 (14:22 +0000)]
Merge pull request #12855 from dkurt:js_features2d_tests
Dmitry Kurtaev [Tue, 16 Oct 2018 09:23:43 +0000 (12:23 +0300)]
Add tests for features2d JavaScript bindings
Rostislav Vasilikhin [Fri, 1 Feb 2019 16:09:31 +0000 (19:09 +0300)]
Merge pull request #13708 from savuor:yuv42x_wide
YUV42x color conversions rewritten to wide intrinsics (#13708)
* a*b+c -> fma
* YUV420sp2RGB initially vectorized
* shorter var names
* loops by 4
* yuv420p2rgb vectorized
* yuv422toRGB vectorized
* reg arrays
* rgb2yuv420 vectorized
* warnings fixed
* try to fix align error
Alexander Alekhin [Fri, 1 Feb 2019 16:07:43 +0000 (16:07 +0000)]
Merge pull request #13722 from alalek:ocl_pass_ptr_for_empty_umat
Alexander Alekhin [Fri, 1 Feb 2019 10:01:39 +0000 (10:01 +0000)]
Merge pull request #13736 from dkurt:dnn_ie_future
Alexander Alekhin [Fri, 1 Feb 2019 09:58:14 +0000 (09:58 +0000)]
Merge pull request #13735 from dkurt:fix_13725
Alexander Alekhin [Fri, 1 Feb 2019 09:57:08 +0000 (09:57 +0000)]
Merge pull request #13716 from dkurt:dnn_conv_weights_clone
Hannah McLaughlin [Thu, 31 Jan 2019 12:34:36 +0000 (12:34 +0000)]
Merge pull request #13718 from lochsh:svm-sigmoid-fix
SVM sigmoid kernel fix (issue #13621) (#13718)
* Added test for sigmoid case for retrieving support vectors
* undo unhelpful test
* add test for sigmoid SVM with data that is easily separable into two concentric circles
* Update sigmoid kernel to use tanh(gamma * <x, y> + coef0) instead of -tanh(gamma * <x, y> + coef0)
* remove unnecessary constraint on coef0
* cleanup
* fixing inappropriate use of doubles
* Add f to float literal
* replace CV_Assert with ASSERT_EQ where appropriate
Dmitry Kurtaev [Thu, 31 Jan 2019 10:38:24 +0000 (13:38 +0300)]
Fix IE tests
Dmitry Kurtaev [Thu, 31 Jan 2019 10:40:54 +0000 (13:40 +0300)]
Align TensorFlow and OpenCV paths to create a text graph
Vitaly Tuzov [Wed, 30 Jan 2019 19:37:27 +0000 (22:37 +0300)]
Merge pull request #13693 from terfendail:spatialgrad_wintr
* spatialGradient() reworked to use wide universal intrinsics
* Moved row pointers inside loops
Alexander Nesterov [Tue, 22 Jan 2019 16:01:48 +0000 (15:01 -0100)]
Update obj. detect sample
Alexander Alekhin [Wed, 30 Jan 2019 11:48:53 +0000 (14:48 +0300)]
ocl: support empty "ptr only" UMat in Kernel::set()
add messages to avoid silent kernel destruction
Ramez Rafla [Tue, 29 Jan 2019 21:14:33 +0000 (16:14 -0500)]
Merge pull request #13710 from ramezrafla:3.4
* Added DMatch bindings
* Added CR which was accidently removed to maintain consistency
* Removed trailing white-space
Alexander Alekhin [Tue, 29 Jan 2019 16:21:08 +0000 (16:21 +0000)]
Merge pull request #13684 from terfendail:lblend_wintr
Alexander Alekhin [Tue, 29 Jan 2019 16:20:24 +0000 (19:20 +0300)]
Merge pull request #13679 from alalek:imgproc_median_blur_cleanup
* imgproc: cleanup medianBlur_8u_O1 code
Unnecessary per-channel buffers: H[c] / lut[c]
* imgproc(medianBlur_8u_O1): use CV_SIMD_WIDTH for alignment
Dmitry Kurtaev [Tue, 29 Jan 2019 11:29:47 +0000 (14:29 +0300)]
Clone convolution layer weights only for fusion
Arnaud Brejeon [Mon, 28 Jan 2019 18:36:00 +0000 (10:36 -0800)]
Merge pull request #13672 from arnaudbrejeon:bug_fix_12961
PyrDown: Fix bug #12961 (#13672)
* Force unaligned pointer and create test
* More cross-platform solution
* MSVC expects a proper order
* Remove useless clang macro
Alexander Alekhin [Mon, 28 Jan 2019 18:34:20 +0000 (18:34 +0000)]
Merge pull request #13692 from dkurt:dnn_do_not_crash_myriad_in_tests
Alexander Alekhin [Mon, 28 Jan 2019 18:32:21 +0000 (18:32 +0000)]
Merge pull request #13707 from dkurt:fix_13575
Dmitry Kurtaev [Mon, 28 Jan 2019 15:44:31 +0000 (18:44 +0300)]
Fix a dnn bug with retrieving all the output blobs
Dmitry Kurtaev [Fri, 25 Jan 2019 12:02:29 +0000 (15:02 +0300)]
Remove ASSERT_ANY_THROW checks fpr Myriad plugin and FP32 networks
Alexander Alekhin [Fri, 25 Jan 2019 11:59:29 +0000 (11:59 +0000)]
Merge pull request #13608 from allnes:dnn_rework
Vitaly Tuzov [Thu, 24 Jan 2019 13:26:09 +0000 (16:26 +0300)]
Added performance test for blendLinear
Vitaly Tuzov [Wed, 23 Jan 2019 14:24:44 +0000 (17:24 +0300)]
blendLinear() reworked to use wide universal intrinsics
Alexander Alekhin [Thu, 24 Jan 2019 16:21:49 +0000 (16:21 +0000)]
Merge pull request #13685 from mshabunin:fix-avfoundation-34