Maksim Shabunin [Fri, 2 Feb 2018 15:14:25 +0000 (18:14 +0300)]
Fixed components INTERFACE_INCLUDE_DIRECTORY property in "world" builds
Harshal Nishar [Fri, 2 Feb 2018 14:00:11 +0000 (19:30 +0530)]
Fix in Canny when Sobel apertureSize is 7 (#10743)
* Fixing a bug in Canny implemetation when Sobel aperture size is 7.
* Fixing the bug in Canny accross variants and in test_canny.cpp
* Replacing a tab with white space
Vadim Pisarevsky [Fri, 2 Feb 2018 11:27:26 +0000 (11:27 +0000)]
Merge pull request #10761 from seiko2plus:issue10753
Alexander Alekhin [Fri, 2 Feb 2018 10:48:49 +0000 (10:48 +0000)]
Merge pull request #10751 from nglee:dev
Ya-Chiu Wu [Fri, 2 Feb 2018 10:20:46 +0000 (18:20 +0800)]
Merge pull request #10553 from GlueCrow:bgfg_knn_opencl
Add ocl version BackgroundSubtractorKNN (#10553)
* Add ocl version bgfg_knn
* Add ocl KNN perf test
* ocl KNN: Avoid unnecessary initializing when non-UMat parameters are used
* video: turn off OpenCL for color KNN on Intel devices
due performance degradation
* video: turn off KNN OpenCL on Apple devices with Intel iGPU
due process freeze during clBuildProgram() call
Alexander Alekhin [Fri, 2 Feb 2018 09:34:59 +0000 (12:34 +0300)]
Merge pull request #10763 from alalek:issue_10760
Alexander Alekhin [Fri, 2 Feb 2018 09:26:22 +0000 (12:26 +0300)]
js: filterout OpenCL headers
Sayed Adel [Fri, 2 Feb 2018 03:31:25 +0000 (03:31 +0000)]
core:ppc Fix compile issue with Eigen
Alexander Alekhin [Thu, 1 Feb 2018 18:34:56 +0000 (18:34 +0000)]
Merge pull request #10756 from terfendail:orb_firstlevel
Alexander Alekhin [Thu, 1 Feb 2018 18:32:13 +0000 (18:32 +0000)]
Merge pull request #10706 from alalek:exception_opencv_version
Alexander Alekhin [Thu, 1 Feb 2018 18:31:52 +0000 (18:31 +0000)]
Merge pull request #10705 from alalek:opencv_version_update
Vitaly Tuzov [Thu, 1 Feb 2018 13:58:35 +0000 (16:58 +0300)]
Fixed assertion in ORB::setFirstLevel()
Vadim Pisarevsky [Thu, 1 Feb 2018 13:41:29 +0000 (13:41 +0000)]
Merge pull request #10690 from terfendail:orb_firstlevel
Maksim Shabunin [Thu, 1 Feb 2018 13:30:23 +0000 (16:30 +0300)]
Do not build protobuf without dnn (#10689)
* Do not build protobuf if dnn is disabled
* Added BUILD_LIST cmake option to the cache
* Moved protobuf to the top level
* Fixed static build
* Fixed world build
* fixup! Fixed world build
Vadim Pisarevsky [Thu, 1 Feb 2018 13:28:10 +0000 (13:28 +0000)]
Merge pull request #10730 from dkurt:dnn_faster_rcnn_rfcn_tests
Vadim Pisarevsky [Thu, 1 Feb 2018 13:23:49 +0000 (13:23 +0000)]
Merge pull request #10742 from alalek:fix_winrt_fs_build
Vadim Pisarevsky [Thu, 1 Feb 2018 13:22:09 +0000 (13:22 +0000)]
Merge pull request #10746 from dkurt:dnn_batch_norm_from_nvidia_caffe
Alexander Alekhin [Thu, 1 Feb 2018 13:21:17 +0000 (13:21 +0000)]
Merge pull request #10748 from dkurt:fix_dnn_slice_layer
Alexander Alekhin [Thu, 1 Feb 2018 13:08:36 +0000 (13:08 +0000)]
Merge pull request #10709 from alalek:cleanup_2018
Alexander Alekhin [Thu, 1 Feb 2018 12:39:27 +0000 (12:39 +0000)]
Merge pull request #10754 from dkurt:dnn_ocl_gemv_min_globalsize
Vadim Pisarevsky [Thu, 1 Feb 2018 12:26:50 +0000 (12:26 +0000)]
Merge pull request #10749 from alalek:issue_10736
Aditya Rohan [Thu, 1 Feb 2018 12:21:00 +0000 (17:51 +0530)]
Merge pull request #10732 from Riyuzakii:10598
* Introduces error throw for line drawn with THICKNESS=0
Dmitry Kurtaev [Thu, 1 Feb 2018 08:35:35 +0000 (11:35 +0300)]
ocl4dnnGEMV in case of row_size < 4
Alexander Alekhin [Thu, 1 Feb 2018 10:49:54 +0000 (10:49 +0000)]
Merge pull request #10717 from pengli:dnn
Alexander Alekhin [Thu, 1 Feb 2018 10:31:48 +0000 (10:31 +0000)]
Merge pull request #10733 from alalek:issue_10654
Li Peng [Wed, 31 Jan 2018 11:28:58 +0000 (19:28 +0800)]
mvn layer ocl update
it fuse ocl kernels to reduce kernel enqueue
Signed-off-by: Li Peng <peng.li@intel.com>
Li Peng [Wed, 31 Jan 2018 18:09:13 +0000 (02:09 +0800)]
fix extra spaces in build option
Signed-off-by: Li Peng <peng.li@intel.com>
Li Peng [Thu, 25 Jan 2018 14:15:41 +0000 (22:15 +0800)]
eltwise layer SUM op update
Signed-off-by: Li Peng <peng.li@intel.com>
Namgoo Lee [Wed, 31 Jan 2018 18:45:51 +0000 (03:45 +0900)]
cuda_stream: do not allocate GPU memory by default (fixes #8725)
Alexander Alekhin [Wed, 31 Jan 2018 18:54:08 +0000 (18:54 +0000)]
samples: remove orphan documentation snippet: SBM_Sample.cpp
Dmitry Kurtaev [Wed, 31 Jan 2018 16:12:37 +0000 (19:12 +0300)]
Fix slice layer from TensorFlow
Alexander Alekhin [Wed, 31 Jan 2018 13:56:24 +0000 (16:56 +0300)]
cmake: avoid using of pkg-config in cross-compilation mode unconditionally
- specify PKG_CONFIG_LIBDIR environment variable to new target SYSROOT
Arjan van de Ven [Wed, 31 Jan 2018 13:34:12 +0000 (05:34 -0800)]
Merge pull request #10468 from fenrus75:avx512-2
* Add a 512 bit codepath to the AVX512 fastConv function
this patch adds a 512 wide codepath to the fastConv() function for
AVX512 use.
The basic idea is to process the first N * 16 elements of the vector
with avx512, and then run the rest of the vector using the traditional
AVX2 codepath.
* dnn: use unaligned AVX512 load (OpenCV aligns data on 32-byte boundary)
* dnn: change "vecsize" condition for AVX512
* dnn: fix indentation
Alexander Alekhin [Wed, 31 Jan 2018 13:28:09 +0000 (13:28 +0000)]
Merge pull request #10701 from dkurt:tf_ave_pooling
Dmitry Kurtaev [Wed, 31 Jan 2018 13:25:45 +0000 (16:25 +0300)]
Fix Batch Normalization layer imported from NVIDIA Caffe.
Dmitry Kurtaev [Fri, 26 Jan 2018 13:45:25 +0000 (16:45 +0300)]
Fix padding for average pooling from TensorFlow
Alexander Alekhin [Wed, 31 Jan 2018 12:00:45 +0000 (15:00 +0300)]
winrt: fix build
Alexander Alekhin [Wed, 31 Jan 2018 09:57:29 +0000 (09:57 +0000)]
Merge pull request #10738 from tomoaki0705:fixBuildVisualStudio
Alexander Alekhin [Wed, 31 Jan 2018 09:41:37 +0000 (09:41 +0000)]
Merge pull request #10735 from seiko2plus:coreYieldPPC
Tomoaki Teshima [Wed, 31 Jan 2018 09:22:09 +0000 (18:22 +0900)]
fix build error on Visual Studio 2013 and earlier
Sayed Adel [Wed, 31 Jan 2018 04:03:35 +0000 (04:03 +0000)]
core:ppc Add yield support
Alexander Alekhin [Tue, 30 Jan 2018 18:29:18 +0000 (18:29 +0000)]
cmake: fix gnu.toolchain
- CMAKE_FIND_ROOT_PATH_MODE_PROGRAM: ONLY => NEVER
- update handling of CMAKE_FIND_ROOT_PATH_MODE_* variables
Alexander Alekhin [Tue, 30 Jan 2018 14:13:29 +0000 (14:13 +0000)]
Merge pull request #10691 from alalek:parallel_for_2018
Dmitry Kurtaev [Tue, 30 Jan 2018 10:17:35 +0000 (13:17 +0300)]
Faster-RCNN and RFCN tests
https://github.com/rbgirshick/py-faster-rcnn
https://github.com/YuwenXiong/py-R-FCN
Vadim Pisarevsky [Tue, 30 Jan 2018 09:22:46 +0000 (09:22 +0000)]
Merge pull request #10707 from alalek:update_reports
Vadim Pisarevsky [Tue, 30 Jan 2018 09:20:57 +0000 (09:20 +0000)]
Merge pull request #10721 from alalek:ocl_force_clBuildProgram
Li Peng [Wed, 24 Jan 2018 12:26:16 +0000 (20:26 +0800)]
slice layer ocl support
Signed-off-by: Li Peng <peng.li@intel.com>
Alexander Alekhin [Mon, 29 Jan 2018 13:24:59 +0000 (16:24 +0300)]
Merge pull request #10719 from alisentas:master
Fix cv::CommandLineParser::check() documentation
Alexander Alekhin [Mon, 29 Jan 2018 12:49:50 +0000 (15:49 +0300)]
ocl: force clBuildProgram() call after clCreateProgramWithBinary()
Ali Sentas [Mon, 29 Jan 2018 12:14:21 +0000 (15:14 +0300)]
Fix cv::CommandLineParser::check() documentation
Alexander Alekhin [Mon, 29 Jan 2018 09:03:36 +0000 (09:03 +0000)]
Merge pull request #10713 from alalek:issue_10710
Alexander Alekhin [Sun, 28 Jan 2018 01:54:25 +0000 (01:54 +0000)]
dnn(ocl): fix build options for Apple OpenCL
Alexander Alekhin [Sat, 27 Jan 2018 23:22:44 +0000 (23:22 +0000)]
cleanup unused/outdated files/scripts/links
Alexander Alekhin [Sat, 27 Jan 2018 13:37:38 +0000 (13:37 +0000)]
opencv_version: dump detected HW features
Alexander Alekhin [Sat, 27 Jan 2018 13:09:28 +0000 (13:09 +0000)]
opencv_version: dump OpenCL information via opencv_version
fix missing "opencv2/core/opencl" headers from core module (updated install list)
Alexander Alekhin [Sat, 27 Jan 2018 14:30:37 +0000 (14:30 +0000)]
perf reports: remove units from table cells
- moved to table title
- can be restored via '--show_units' option
- fix microseconds: mks -> us
Alexander Alekhin [Sat, 27 Jan 2018 13:51:16 +0000 (13:51 +0000)]
core: add OpenCV version information into Exception messages
Woody Chow [Sat, 27 Jan 2018 05:46:54 +0000 (14:46 +0900)]
Merge pull request #10697 from woodychow:tbb_task_arena
* Use Intel TBB's task arena if possible
Pavel Rojtberg [Fri, 26 Jan 2018 19:05:13 +0000 (20:05 +0100)]
Merge pull request #10667 from paroj:stereo_calib_ex
calib3d: add stereoCalibrateExtended (#10667)
* cvCalibrateCamera2Internal: simplify per view error computation
* calib3d: add stereoCalibrateExtended
- allow CALIB_USE_EXTRINSIC_GUESS
- returns per view errors
* calib3d: add stereoCalibrateExtended test
Alexander Alekhin [Fri, 26 Jan 2018 19:01:48 +0000 (19:01 +0000)]
Merge pull request #10688 from LaurentBerger:calcback
Alexander Alekhin [Fri, 26 Jan 2018 18:59:47 +0000 (21:59 +0300)]
Merge pull request #10700 from alalek:cpu_dispatch_axv512
* cmake: enable CPU dispatching for AVX512 (SKX)
* cmake: update handling of unsupported flags/modes
Alexander Alekhin [Fri, 26 Jan 2018 16:22:38 +0000 (16:22 +0000)]
Merge pull request #10693 from travbid:IntrinsicParams_fix
LaurentBerger [Thu, 25 Jan 2018 11:23:31 +0000 (12:23 +0100)]
solve issue #10687
Alexander Alekhin [Fri, 26 Jan 2018 10:04:41 +0000 (10:04 +0000)]
Merge pull request #10681 from hmaarrfk:python_stitching_scans
Alexander Alekhin [Fri, 19 Jan 2018 14:26:29 +0000 (17:26 +0300)]
core: fix pthreads performance
OpenCV pthreads-based implementation changes:
- rework worker threads pool, allow to execute job by the main thread too
- rework synchronization scheme (wait for job completion, threads 'pong' answer is not required)
- allow "active wait" (spin) by worker threads and by the main thread
- use _mm_pause() during active wait (support for Hyper-Threading technology)
- use sched_yield() to avoid preemption of still working other workers
- don't use getTickCount()
- optional builtin thread pool profiler (disabled by compilation flag)
Alexander Alekhin [Fri, 26 Jan 2018 04:02:21 +0000 (04:02 +0000)]
Merge pull request #10676 from dkurt:dnn_for_newer_mobilenet_ssd
Travers [Fri, 26 Jan 2018 01:04:45 +0000 (14:04 +1300)]
IntrinsicParams operator+ fix
Maksim Shabunin [Thu, 25 Jan 2018 13:47:47 +0000 (13:47 +0000)]
Merge pull request #10619 from mshabunin:fix-run-py-without-java
Vitaly Tuzov [Thu, 25 Jan 2018 12:23:59 +0000 (15:23 +0300)]
Fixed issue in ORB detection if firstLevel property is set above 0
Alexander Alekhin [Wed, 24 Jan 2018 10:40:56 +0000 (13:40 +0300)]
perf: update message
Alexander Alekhin [Thu, 25 Jan 2018 10:18:36 +0000 (10:18 +0000)]
Merge pull request #10661 from alalek:parallel_kmeans
Maksim Shabunin [Thu, 25 Jan 2018 09:57:43 +0000 (09:57 +0000)]
Merge pull request #10679 from mshabunin:vs-defense-flags
Alexander Alekhin [Thu, 25 Jan 2018 07:18:25 +0000 (07:18 +0000)]
Merge pull request #10672 from pengli:dnn
Li Peng [Tue, 23 Jan 2018 15:52:41 +0000 (23:52 +0800)]
mvn, batch_norm and relu layer fusion
Signed-off-by: Li Peng <peng.li@intel.com>
Li Peng [Tue, 23 Jan 2018 12:44:28 +0000 (20:44 +0800)]
convolution and tanh layer fusion
Signed-off-by: Li Peng <peng.li@intel.com>
Mark Harfouche [Wed, 24 Jan 2018 20:42:45 +0000 (12:42 -0800)]
Exported a high level stitcher the DLL
allows Stitcher to be used for scans from within python.
I had to use very strange notation because I couldn't export the `enum`
`Mode` making the Cpython generated code unable to compile.
```c++
class Stitcher {
public:
enum Mode
{
PANORAMA = 0,
SCANS = 1,
};
...
```
Also removed duplicate code from the `createStitcher` function making
use of the `Stitcher::create` function
Alexander Alekhin [Wed, 24 Jan 2018 17:49:20 +0000 (17:49 +0000)]
Merge pull request #10675 from LaurentBerger:tutopy_arithm
Alexander Alekhin [Wed, 24 Jan 2018 17:48:11 +0000 (17:48 +0000)]
Merge pull request #10677 from alalek:issue_10668
Maksim Shabunin [Wed, 24 Jan 2018 14:17:36 +0000 (17:17 +0300)]
VS with hardening: added guard flag, moved dynamicbase and safeseh to linker flags
Jozef Mlich [Wed, 24 Jan 2018 14:03:23 +0000 (15:03 +0100)]
Merge pull request #10663 from jmlich:master
* hogsvm compatibility with python3
Alexander Alekhin [Wed, 24 Jan 2018 11:53:04 +0000 (14:53 +0300)]
core(lapack): fix build issues related to 'extern "C"'
Dmitry Kurtaev [Wed, 24 Jan 2018 11:01:42 +0000 (14:01 +0300)]
PriorBox layer with explicit normalized sizes
LaurentBerger [Wed, 24 Jan 2018 08:09:49 +0000 (09:09 +0100)]
bad image file
Elena Gvozdeva [Tue, 23 Jan 2018 22:48:06 +0000 (01:48 +0300)]
add HAL for FAST (#10362)
* add HAL for FAST
* add new interface
Vadim Pisarevsky [Tue, 23 Jan 2018 10:42:05 +0000 (10:42 +0000)]
Merge pull request #10635 from csukuangfj:doc-checkVector
Vadim Pisarevsky [Tue, 23 Jan 2018 10:38:14 +0000 (10:38 +0000)]
Merge pull request #10644 from csukuangfj:doc-fundamental-matrix
Pavel Rojtberg [Tue, 23 Jan 2018 10:36:30 +0000 (11:36 +0100)]
persistence: replace arbitrary limit of cn to 4 by CV_CN_MAX (#10636)
* persistence: replace arbitrary limit of cn to 4 by CV_CN_MAX
* python: added persistence test, remove temp files
* fixup! python: added persistence test, remove temp files
* fixup! python: added persistence test, remove temp files
Alexander Alekhin [Tue, 23 Jan 2018 10:16:55 +0000 (10:16 +0000)]
Merge pull request #10655 from dkurt:dnn_tests
Dmitry Kurtaev [Sat, 20 Jan 2018 18:55:25 +0000 (21:55 +0300)]
OpenCV face detection network test
Alexander Alekhin [Mon, 22 Jan 2018 11:14:02 +0000 (14:14 +0300)]
core: kmeans refactoring
- reduce scope of i,k,j variables
- use cv::AutoBuffer
- template<bool onlyDistance> class KMeansDistanceComputer
- eliminate manual unrolling: CV_ENABLE_UNROLLED
Alexander Alekhin [Wed, 17 Jan 2018 14:39:23 +0000 (17:39 +0300)]
core: fix kmeans multi-threaded performance
Alexander Alekhin [Thu, 18 Jan 2018 15:11:30 +0000 (18:11 +0300)]
core(perf): refactor kmeans test
- don't use RNG for "task size" parameters (N, K, dims)
- add "good" kmeans test data (without singularities: K > unique points)
Vitaly Tuzov [Mon, 22 Jan 2018 10:26:32 +0000 (13:26 +0300)]
bitexact gaussianblur implementation (#10345)
* Bit-exact implementation of GaussianBlur smoothing
* Added universal intrinsics based implementation for bit-exact CV_8U GaussianBlur smoothing.
* Added parallel_for to evaluation of bit-exact GaussianBlur
* Added custom implementations for 3x3 and 5x5 bit-exact GaussianBlur
Alexander Alekhin [Sun, 21 Jan 2018 17:25:37 +0000 (17:25 +0000)]
Merge pull request #10657 from GlueCrow:mog2_init_fix
Glue Crow [Sun, 21 Jan 2018 15:02:05 +0000 (23:02 +0800)]
ocl: Avoid unnecessary initializing when non-UMat parameters are used
Gregory Morse [Sun, 21 Jan 2018 06:26:48 +0000 (07:26 +0100)]
Merge pull request #10649 from GregoryMorse:patch-3
* Fix for QT image window rectangle
* Update window_QT.h
* Update window_QT.cpp
* trailing whitespace
* highgui: fix QT getWindowImageRect()
Maksim Shabunin [Wed, 17 Jan 2018 13:35:44 +0000 (16:35 +0300)]
run.py: simplified scripts, fixed most of PEP8 warnings
Fangjun Kuang [Fri, 19 Jan 2018 12:23:09 +0000 (13:23 +0100)]
Improve the doc for fundamental matrix.
Vadim Pisarevsky [Fri, 19 Jan 2018 11:12:19 +0000 (11:12 +0000)]
Merge pull request #10629 from GlueCrow:ocl_mog2_test_fix
Namgoo Lee [Fri, 19 Jan 2018 11:06:05 +0000 (20:06 +0900)]
cv::cuda::cvtColor bug fix (#10640)
* cuda::cvtColor bug fix
Fixed bug in conversion formula between RGB space and LUV space.
Testing with opencv_test_cudaimgproc.exe, this commit reduces the number
of failed tests from 191 to 95. (96 more tests pass)
* Rename variables