victor-ludorum [Sat, 30 Dec 2017 04:51:46 +0000 (10:21 +0530)]
Updating rotcalipers.cpp to resolve issue #10096
Updating the documentation of the rotcalipers.cpp to resolve issue #10096
Alexander Alekhin [Fri, 29 Dec 2017 15:57:57 +0000 (15:57 +0000)]
Merge pull request #10463 from alalek:avx512skx
Alexander Alekhin [Fri, 29 Dec 2017 05:06:52 +0000 (05:06 +0000)]
cmake(opt): AVX512_SKX
Alexander Alekhin [Thu, 28 Dec 2017 16:45:16 +0000 (16:45 +0000)]
Merge pull request #10329 from jasjuang:master
Alexander Alekhin [Thu, 28 Dec 2017 16:43:17 +0000 (16:43 +0000)]
Merge pull request #10410 from GlueCrow:master
Alexander Alekhin [Thu, 28 Dec 2017 16:04:51 +0000 (16:04 +0000)]
Merge pull request #10456 from dkurt:dnn_allocate_mem_for_optimized_concat
Alexander Alekhin [Thu, 28 Dec 2017 15:56:56 +0000 (15:56 +0000)]
Merge pull request #10416 from fenrus75:avx512
Alexander Alekhin [Thu, 28 Dec 2017 15:30:43 +0000 (18:30 +0300)]
Merge pull request #10429 from wxzs5:fix-documentation
Alexander Alekhin [Thu, 28 Dec 2017 15:29:30 +0000 (18:29 +0300)]
Merge pull request #10436 from alalek:test_threads
Alexander Alekhin [Thu, 28 Dec 2017 15:28:29 +0000 (18:28 +0300)]
Merge pull request #10444 from alalek:fix_asan_imgcodecs
Alexander Alekhin [Thu, 28 Dec 2017 02:16:42 +0000 (02:16 +0000)]
cmake: AVX512 -> AVX_512F
Vinay Sharma [Thu, 28 Dec 2017 14:28:34 +0000 (19:58 +0530)]
Merge pull request #10352 from vinay0410:write_pbm
* added write as pbm
* add tests for pbm
* imgcodecs: PBM support
- drop additional PBM parameters
- write: fix P1/P4 mode (no maxval 255 value after width/height)
- write: invert values for P1/P4
- write: P1: compact ASCII mode (no spaces)
- simplify pbm test
- drop .pxm extension (http://netpbm.sourceforge.net/doc/ doesn't know such extension)
Tom Becker [Thu, 28 Dec 2017 14:23:11 +0000 (06:23 -0800)]
Merge pull request #10232 from TomBecker-BD:hough-many-circles
Hough many circles (#10232)
* Add Hui's optimization. Merge with latest changes in OpenCV.
* Use conditional compilation instead of a runtime flag.
* Whitespace.
* Create the sequence for the nonzero edge pixels only if using that approach.
* Improve performance for finding very large numbers of circles
* Return the circles with the larger accumulator values first, as per API documentation.
Use a separate step to check distance between circles. Allows circles to be sorted by strength first. Avoids locking in EstimateRadius which was slowing it down.
Return centers only if maxRadius == 0 as per API documentation.
* Sort the circles so results are deterministic. Otherwise the order of circles with the same strength depends on parallel processing completion order.
* Add test for HoughCircles.
* Add beads test.
* Wrap the non-zero points structure in a common interface so the code can use either a vector or a matrix.
* Remove the special case for skipping the radius search if maxRadius==0.
* Add performance tests.
* Use NULL instead of nullptr.
OpenCV should compile with C++98 compiler.
* Put test suite name first.
Use different test suite names for each test to avoid an error from the test runner.
* Address build bot errors and warnings.
* Skip radius search if maxRadius < 0.
* Dynamically switch to NZPointList when it will be faster than NZPointSet.
* Fix compile error: missing 'typename' prior to dependent type name.
* Fix compile error: missing 'typename' prior to dependent type name.
This time fix it the non C++ 11 way.
* Fix compile error: no type named 'const_reference' in 'class cv::NZPointList'
* Disable ManySmallCircles tests. Failing on Mac.
* Change beads image to JPEG for smaller file size.
Try enabling the ManySmallCircles tests again.
* Remove ManySmallCircles tests. They are failing on the Mac build.
* Fix expectations to check all circles.
* Changing case on a case-insensitive file system
Step 1: remove the old file names
* Changing case on a case-insensitive file system
Step 2: add them back with the new names
* Fix cmpAccum function to be strictly weak ordered.
* Add tests for many small circles.
* imgproc(perf): fix HoughCircles tests
* imgproc(houghCircles): refactor code
- simplify NZPointList
- drop broken (de-synchronization of 'current'/'mi' fields) NZPointSet iterator
- NZPointSet iterator is replaced to direct area scan
- use SIMD intrinsics
- avoid std exceptions (build for embedded systems)
Dmitry Kurtaev [Thu, 28 Dec 2017 13:04:09 +0000 (16:04 +0300)]
Allocate new memory for optimized concat to prevent collisions.
Add a flag to disable memory reusing in dnn module.
Alexander Alekhin [Thu, 28 Dec 2017 12:48:26 +0000 (12:48 +0000)]
Merge pull request #10452 from mshabunin:mediasdk-device
Maksim Shabunin [Thu, 28 Dec 2017 10:43:35 +0000 (13:43 +0300)]
Changed VA device in MediaSDK session initialization
Alexander Alekhin [Thu, 28 Dec 2017 04:43:41 +0000 (04:43 +0000)]
Merge pull request #10445 from pengli:dnn
Li Peng [Wed, 27 Dec 2017 15:15:50 +0000 (23:15 +0800)]
Add ocl version FasterRCNN accuracy test
Signed-off-by: Li Peng <peng.li@intel.com>
Alexander Alekhin [Wed, 27 Dec 2017 18:24:43 +0000 (18:24 +0000)]
imgcodecs(png): resolve ASAN issue with vars scope and setjmp() call
Alexander Alekhin [Thu, 28 Dec 2017 02:30:59 +0000 (05:30 +0300)]
Merge pull request #10435 from DabatoIsCool:master
Alexander Alekhin [Thu, 28 Dec 2017 01:26:14 +0000 (01:26 +0000)]
Merge pull request #10424 from dkurt:fix_concat_optim
Alexander Alekhin [Wed, 27 Dec 2017 14:56:29 +0000 (14:56 +0000)]
dnn(test): avoid calling of cv::setNumThreads() in tests directly
It is not necessary by default.
Also it breaks test system command-line parameters: --perf_threads / --test_threads
Alexander Alekhin [Wed, 27 Dec 2017 14:51:03 +0000 (14:51 +0000)]
ts(feature): add "--test_threads=<N>" command-line option
Arthur Williams [Wed, 27 Dec 2017 14:09:25 +0000 (14:09 +0000)]
Fixed #10433
Alexander Alekhin [Wed, 27 Dec 2017 13:18:10 +0000 (13:18 +0000)]
Merge pull request #10428 from pengli:dnn
wxzs5 [Wed, 27 Dec 2017 03:57:58 +0000 (11:57 +0800)]
Use class' method to set attribute value
wxzs5 [Wed, 27 Dec 2017 03:57:17 +0000 (11:57 +0800)]
Fix functions‘ class attribution error
Arjan van de Ven [Mon, 25 Dec 2017 21:08:22 +0000 (21:08 +0000)]
Provide a few AVX512 optimized functions for the DNN module
This patch adds AVX512 optimized fastConv as well as the hookups
needed to get these called in the convolution_layer.
AVX512 fastConv is code-identical on a C level to the AVX2 one,
but is measurably faster due to AVX512 having more registers available
to cache results in.
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
Alexander Alekhin [Tue, 26 Dec 2017 14:26:58 +0000 (17:26 +0300)]
Merge pull request #10419 from shortcipher3:master
Alexander Alekhin [Tue, 26 Dec 2017 14:21:59 +0000 (14:21 +0000)]
Merge pull request #10418 from mypopydev:master
Dmitry Kurtaev [Tue, 26 Dec 2017 13:49:33 +0000 (16:49 +0300)]
Limit Concat layer optimization
Li Peng [Tue, 26 Dec 2017 09:41:48 +0000 (17:41 +0800)]
dnn(ocl4dnn): update pre-tuned kernel config
Signed-off-by: Li Peng <peng.li@intel.com>
Alexander Alekhin [Tue, 26 Dec 2017 05:51:16 +0000 (05:51 +0000)]
Merge pull request #10404 from alalek:build_fix_msvs2010
Jun Zhao [Tue, 26 Dec 2017 02:25:36 +0000 (10:25 +0800)]
opencl/cvtclr_dx: fix not compile-time constants issue.
fix the "initializing global variables with values that are not
compile-time constants" issue in Intel SDK for OpenCL. The root cause
is when initializing global variables with value, the variable need is
compile-time constants.
Thanks Zheng, Yang <yang.zheng@intel.com>,
Chodor, Jaroslaw <jaroslaw.chodor@intel.com> give a help.
Signed-off-by: Liu,Kaixuan <kaixuan.liu@intel.com>
Signed-off-by: Jun Zhao <jun.zhao@intel.com>
Chris Hall [Tue, 26 Dec 2017 05:01:48 +0000 (22:01 -0700)]
JS examples - FFT didn't work for non-square images because rows/cols were switched, Histogram example misspelled point
Arjan van de Ven [Mon, 25 Dec 2017 21:06:52 +0000 (21:06 +0000)]
Add basic plumbing for AVX512 support
The opencv infrastructure mostly has the basics for supporting avx512 math functions,
but it wasn't hooked up (likely due to lack of users)
In order to compile the DNN functions for AVX512, a few things need to be hooked up
and this patch does that
Signed-off-by: Arjan van de Ven <arjan@linux.intel.com>
GlueCrow [Sat, 23 Dec 2017 13:57:17 +0000 (21:57 +0800)]
Optimize OpenCL BackgroundSubstractionMOG2
Alexander Alekhin [Sat, 23 Dec 2017 00:06:34 +0000 (00:06 +0000)]
build: fix MSVS2010 build error
Alexander Alekhin [Fri, 22 Dec 2017 23:22:50 +0000 (23:22 +0000)]
Merge tag '3.4.0'
Alexander Alekhin [Fri, 22 Dec 2017 19:46:21 +0000 (19:46 +0000)]
OpenCV version++
3.4.0
Alexander Alekhin [Fri, 22 Dec 2017 19:43:59 +0000 (22:43 +0300)]
Merge pull request #10398 from alalek:ml_simplify_simulated_annealing
Alexander Alekhin [Fri, 22 Dec 2017 18:50:04 +0000 (18:50 +0000)]
ml(ANN_MLP): ensure that train() call is always successful
Alexander Alekhin [Fri, 22 Dec 2017 16:18:50 +0000 (16:18 +0000)]
Merge pull request #10401 from terfendail:resize_linear_revert
Alexander Alekhin [Fri, 22 Dec 2017 15:58:56 +0000 (15:58 +0000)]
Merge pull request #10402 from dkurt:dnn_tf_quantized
Alexander Alekhin [Fri, 22 Dec 2017 15:48:40 +0000 (15:48 +0000)]
Merge pull request #10385 from pengli:dnn
Dmitry Kurtaev [Thu, 21 Dec 2017 14:15:48 +0000 (17:15 +0300)]
TensorFlow weights dequantization
Vitaly Tuzov [Fri, 22 Dec 2017 14:10:03 +0000 (17:10 +0300)]
Reverted calls to linear resize back to generic version for floating point matrices
Alexander Alekhin [Fri, 22 Dec 2017 14:07:02 +0000 (14:07 +0000)]
Merge pull request #10397 from mshabunin:fix-incorrect-assert
Li Peng [Fri, 22 Dec 2017 10:49:47 +0000 (18:49 +0800)]
add one more convolution kernel tuning candidate
Signed-off-by: Li Peng <peng.li@intel.com>
Alexander Alekhin [Fri, 22 Dec 2017 12:33:23 +0000 (15:33 +0300)]
ml: simplify interfaces of SimulatedAnnealingSolver
Vadim Pisarevsky [Fri, 22 Dec 2017 13:35:36 +0000 (13:35 +0000)]
Merge pull request #10396 from berak:fix_superres_sample
Vadim Pisarevsky [Fri, 22 Dec 2017 13:29:37 +0000 (13:29 +0000)]
Merge pull request #10265 from dkurt:nms_for_region_layer
Vadim Pisarevsky [Fri, 22 Dec 2017 13:26:31 +0000 (13:26 +0000)]
Merge pull request #10394 from alalek:cmake_fix_pch_pic_pie
Vadim Pisarevsky [Fri, 22 Dec 2017 13:26:05 +0000 (13:26 +0000)]
Merge pull request #10387 from terfendail:resize23_perftest
Vadim Pisarevsky [Fri, 22 Dec 2017 13:23:55 +0000 (13:23 +0000)]
Merge pull request #10392 from terfendail:bitexact_fallback
Alexander Alekhin [Fri, 22 Dec 2017 13:17:11 +0000 (13:17 +0000)]
Merge pull request #10375 from tomoaki0705:buildWarningMSVC
Alexander Alekhin [Fri, 22 Dec 2017 12:52:56 +0000 (12:52 +0000)]
Merge pull request #10390 from alalek:ocl_option_buffer_rect
Maksim Shabunin [Fri, 22 Dec 2017 12:20:13 +0000 (15:20 +0300)]
Replaced incorrect CV_Assert calls with CV_Error
berak [Fri, 22 Dec 2017 11:34:25 +0000 (12:34 +0100)]
samples: check for valid input in gpu/super_resolution.cpp
Vitaly Tuzov [Fri, 22 Dec 2017 10:23:56 +0000 (13:23 +0300)]
Added fallback to generic linear resize in case bit-exact resize of provided matrix isn't supported
Alexander Alekhin [Fri, 22 Dec 2017 10:46:45 +0000 (13:46 +0300)]
cmake: fix -fPIC/-fPIE handling in precompiled headers (PCH)
Vadim Pisarevsky [Fri, 22 Dec 2017 10:54:49 +0000 (10:54 +0000)]
Merge pull request #10386 from terfendail:resizeexact_c3
Alexander Alekhin [Fri, 22 Dec 2017 09:28:41 +0000 (12:28 +0300)]
ocl: workaround option to disable usage of buffer "Rect" operations
Alexander Alekhin [Fri, 22 Dec 2017 10:04:05 +0000 (10:04 +0000)]
Merge pull request #10389 from wxzs5:yangli
Vadim Pisarevsky [Fri, 22 Dec 2017 09:56:45 +0000 (09:56 +0000)]
Merge pull request #10364 from dkurt:dnn_smooth_tf_data_layout
wxzs5 [Fri, 22 Dec 2017 09:13:57 +0000 (17:13 +0800)]
Remove redundant return variable
Vitaly Tuzov [Fri, 22 Dec 2017 07:08:30 +0000 (10:08 +0300)]
Disabled universal intrinsic based implementation for bit-exact resize of 3-channel images
Tomoaki Teshima [Thu, 21 Dec 2017 23:42:21 +0000 (08:42 +0900)]
clean up the code
* disable the warning in CMake, not int the code using pragma
Alexander Alekhin [Thu, 21 Dec 2017 19:02:13 +0000 (19:02 +0000)]
Merge pull request #10374 from tomoaki0705:removeGstreamerTest
Vadim Pisarevsky [Thu, 21 Dec 2017 18:56:53 +0000 (18:56 +0000)]
Merge pull request #10316 from terfendail:bitexact_c234
Vadim Pisarevsky [Thu, 21 Dec 2017 18:48:38 +0000 (18:48 +0000)]
Merge pull request #10369 from alalek:issue_10351
Alexander Alekhin [Thu, 21 Dec 2017 15:36:23 +0000 (15:36 +0000)]
Merge pull request #10370 from pengli:dnn
Tomoaki Teshima [Thu, 21 Dec 2017 15:14:10 +0000 (00:14 +0900)]
avoid the test which is too strict
* confirmed test failure on Jetson TX1 and TX2
* show the performance but not bit exact result
Li Peng [Thu, 21 Dec 2017 14:01:44 +0000 (22:01 +0800)]
cleanup unnecessary macros in convolution ocl kernel
Signed-off-by: Li Peng <peng.li@intel.com>
Li Peng [Tue, 19 Dec 2017 14:28:18 +0000 (22:28 +0800)]
refactor candidate generation of convolution auto-tuning
Signed-off-by: Li Peng <peng.li@intel.com>
jasjuang [Thu, 21 Dec 2017 09:33:25 +0000 (01:33 -0800)]
handle legacy VTK functions
Dmitry Kurtaev [Fri, 8 Dec 2017 11:14:17 +0000 (14:14 +0300)]
Refactor NMS procedure at RegionLayer
Vadim Pisarevsky [Thu, 21 Dec 2017 08:33:58 +0000 (08:33 +0000)]
Merge pull request #10366 from alalek:ocl_fix_cvtcolor_read
Vadim Pisarevsky [Thu, 21 Dec 2017 08:33:18 +0000 (08:33 +0000)]
Merge pull request #10340 from alalek:log_level_option
Alexander Alekhin [Thu, 21 Dec 2017 01:10:24 +0000 (01:10 +0000)]
imgcodecs(pxm): fix memcpy size
Vitaly Tuzov [Wed, 20 Dec 2017 15:11:21 +0000 (18:11 +0300)]
Added performance tests for linear resize of 2 and 3-channel images
Vadim Pisarevsky [Wed, 20 Dec 2017 14:30:26 +0000 (14:30 +0000)]
Merge pull request #10331 from arrybn:python_dnn_net
Vadim Pisarevsky [Wed, 20 Dec 2017 14:23:01 +0000 (14:23 +0000)]
Merge pull request #10354 from catree:add_python_sample_show_extrinsics
LaurentBerger [Wed, 20 Dec 2017 14:24:46 +0000 (15:24 +0100)]
Calcerror uses now weighted samples (#10346)
* Calcerror uses now sample weights
* catree comment in #10319
Alexander Alekhin [Wed, 20 Dec 2017 14:19:46 +0000 (14:19 +0000)]
imgproc(ocl): fix RGB2RGBA kernel out of range access
Vadim Pisarevsky [Wed, 20 Dec 2017 14:19:46 +0000 (14:19 +0000)]
Merge pull request #10356 from dkurt:dnn_rfcn
Vadim Pisarevsky [Wed, 20 Dec 2017 14:19:17 +0000 (14:19 +0000)]
Merge pull request #10360 from sturkmen72:update_links
Vadim Pisarevsky [Wed, 20 Dec 2017 14:18:41 +0000 (14:18 +0000)]
Merge pull request #10358 from catree:add_tutorial_homography_panorama_stitching
Vitaly Tuzov [Thu, 7 Dec 2017 17:27:58 +0000 (20:27 +0300)]
Added universal intrinsics based implementations for CV_8UC2, CV_8UC3, CV_8UC4 bit-exact resizes.
Dmitry Kurtaev [Wed, 20 Dec 2017 11:13:40 +0000 (14:13 +0300)]
Manage TensorFlow's NHWC data layout is smoother
catree [Tue, 19 Dec 2017 16:03:00 +0000 (17:03 +0100)]
Update left_intrinsics.yml file with chessboard square_size=0.025 and with extrinsic parameters. Add camera_calibration_show_extrinsics.py, a Python sample that allows to display the extrinsics saved during the camera calibration process (cpp-example-calibration).
Dmitry Kurtaev [Tue, 19 Dec 2017 16:43:49 +0000 (19:43 +0300)]
R-FCN models support
Suleyman TURKMEN [Tue, 19 Dec 2017 23:25:16 +0000 (02:25 +0300)]
Update links
Alexander Alekhin [Sun, 17 Dec 2017 22:28:33 +0000 (22:28 +0000)]
core(logger): add log level configuration option
catree [Tue, 19 Dec 2017 21:45:35 +0000 (22:45 +0100)]
Add Demo 5: Basic panorama stitching from a rotating camera in the homography tutorial.
Alexander Alekhin [Tue, 19 Dec 2017 14:04:55 +0000 (14:04 +0000)]
Merge pull request #10341 from pengli:dnn
Alexander Alekhin [Tue, 19 Dec 2017 13:07:59 +0000 (13:07 +0000)]
Merge pull request #10348 from ElenaGvozdeva:Canny_HAL
Vadim Pisarevsky [Tue, 19 Dec 2017 11:03:06 +0000 (11:03 +0000)]
Merge pull request #10347 from dkurt:dnn_remove_deprecated_importer
Li Peng [Tue, 19 Dec 2017 08:59:46 +0000 (16:59 +0800)]
add ocl accuracy test for tf mobilenet ssd
Signed-off-by: Li Peng <peng.li@intel.com>
Li Peng [Tue, 19 Dec 2017 09:59:13 +0000 (17:59 +0800)]
add depthwise convolution kernel
Signed-off-by: Li Peng <peng.li@intel.com>