Alexander Alekhin [Fri, 15 Dec 2017 12:01:12 +0000 (15:01 +0300)]
Merge pull request #10315 from alalek:issue_10147_debug
dnn: more debug information
Sayed Adel [Fri, 15 Dec 2017 11:03:46 +0000 (13:03 +0200)]
core:ppc Fix several issues for VSX (#10303)
- fix conversion intrinsics compatibility with xlc
- implement odd-elements 2 to 4 conversion intrinsics
- improve implementation of universal intrinsic v_popcount
- rename FORCE_INLINE to VSX_FINLINE in vsx_utils.hpp
LaurentBerger [Fri, 15 Dec 2017 10:57:39 +0000 (11:57 +0100)]
Simulated Annealing for ANN_MLP training method (#10213)
* Simulated Annealing for ANN_MLP training method
* EXPECT_LT
* just to test new data
* manage RNG
* Try again
* Just run buildbot with new data
* try to understand
* Test layer
* New data- new test
* Force RNG in backprop
* Use Impl to avoid virtual method
* reset all weights
* try to solve ABI
* retry
* ABI solved?
* till problem with dynamic_cast
* Something is wrong
* Solved?
* disable backprop test
* remove ANN_MLP_ANNEALImpl
* Disable weight in varmap
* Add example for SimulatedAnnealing
zhijackchen [Fri, 15 Dec 2017 09:23:44 +0000 (01:23 -0800)]
Merge pull request #10283 from zhijackchen:exr_export
* Fix issue #10114
Convert table change
From:
CV_8U -> HALF
CV_8S -> HALF
CV_16U -> UINT
CV_16S -> UINT
CV_32S -> UINT
CV_32F -> FLOAT
To:
CV_8U -> HALF
CV_8S -> HALF
CV_16U -> UINT
CV_16S -> FLOAT
CV_32S -> FLOAT loss precision
CV_32F -> FLOAT
Signed integer can't be presented well with UINT. Even adjust bias, CV16S and CV32S will be confused when load from exr file.
Also fix CV_8S negative value incorrect bug
* EXR import and export
imread() from EXR returns CV_32F only
imwrite() accepts CV_32 cv::Mat only and stores FLOAT images by default. Add imwrite() flag to store in HALF format.
* fix compiling error
* clean up
* fix EXR import issues
Alexander Alekhin [Fri, 15 Dec 2017 08:53:34 +0000 (08:53 +0000)]
Merge pull request #10304 from catree:add_tutorial_homography
Alexander Alekhin [Fri, 15 Dec 2017 08:43:10 +0000 (08:43 +0000)]
Merge pull request #10309 from seiko2plus:issue10308
Sayed Adel [Fri, 15 Dec 2017 02:47:52 +0000 (02:47 +0000)]
3rdparty:libpng[1.6.34] Add missing support for PowerPC/VSX
Vadim Pisarevsky [Thu, 14 Dec 2017 21:00:46 +0000 (21:00 +0000)]
Merge pull request #10307 from savuor:fix/expect_softfloat
Tomoaki Teshima [Thu, 14 Dec 2017 21:00:35 +0000 (06:00 +0900)]
core: remove raw SSE2/NEON implementation from convert.cpp (#9831)
* remove raw SSE2/NEON implementation from convert.cpp
* remove raw implementation from Cvt_SIMD
* remove raw implementation from cvtScale_SIMD
* remove raw implementation from cvtScaleAbs_SIMD
* remove duplicated implementation cvt_<float, short>
* remove duplicated implementation cvtScale_<short, short, float>
* add "from double" version of Cvt_SIMD
* modify the condition of test ConvertScaleAbs
* Update convert.cpp
fixed crash in cvtScaleAbs(8s=>8u)
* fixed compile error on Win32
* fixed several test failures because of accuracy loss in cvtScale(int=>int)
* fixed NEON implementation of v_cvt_f64(int=>double) intrinsic
* another attempt to fix test failures
* keep trying to fix the test failures and just introduced compile warnings
* fixed one remaining test (subtractScalar)
Rostislav Vasilikhin [Thu, 14 Dec 2017 18:03:25 +0000 (21:03 +0300)]
SoftFloat tests: assert => expect
Alexander Alekhin [Thu, 14 Dec 2017 16:20:30 +0000 (19:20 +0300)]
dnn: more debug information
Alexander Alekhin [Thu, 14 Dec 2017 15:10:31 +0000 (15:10 +0000)]
Merge pull request #10274 from alalek:update_3rdparty
catree [Wed, 13 Dec 2017 20:41:40 +0000 (21:41 +0100)]
Add tutorial and codes for the homography tutorial.
Alexander Alekhin [Fri, 8 Dec 2017 15:19:30 +0000 (18:19 +0300)]
3rdparty: update libwebp 0.6.1
Alexander Alekhin [Fri, 8 Dec 2017 12:47:54 +0000 (15:47 +0300)]
3rdparty: update libtiff 4.0.9
Alexander Alekhin [Fri, 8 Dec 2017 15:47:47 +0000 (18:47 +0300)]
3rdparty: eliminate jasper build warning
Alexander Alekhin [Thu, 7 Dec 2017 15:30:33 +0000 (18:30 +0300)]
cmake(3rdparty): BUILD_ options before WITH_ options
Suleyman TURKMEN [Sun, 22 Oct 2017 10:06:42 +0000 (13:06 +0300)]
3rdparty: update libjpeg 9b
Alexander Alekhin [Thu, 7 Dec 2017 16:16:11 +0000 (19:16 +0300)]
3rdparty: libpng patch
Suleyman TURKMEN [Thu, 5 Oct 2017 17:36:38 +0000 (20:36 +0300)]
3rdparty: update libpng 1.6.34
Suleyman TURKMEN [Thu, 5 Oct 2017 16:54:13 +0000 (19:54 +0300)]
3rdparty: update zlib 1.2.11
Vadim Pisarevsky [Thu, 14 Dec 2017 11:05:09 +0000 (11:05 +0000)]
Merge pull request #10294 from alalek:fix_copyToMask_dst_initialization
Vadim Pisarevsky [Thu, 14 Dec 2017 11:01:02 +0000 (11:01 +0000)]
Merge pull request #10299 from alalek:cmake_imported_configuration_remap
Vadim Pisarevsky [Wed, 13 Dec 2017 17:31:40 +0000 (17:31 +0000)]
Merge pull request #10302 from alalek:issue_10254
Vadim Pisarevsky [Wed, 13 Dec 2017 17:30:23 +0000 (17:30 +0000)]
Merge pull request #10293 from alalek:fix_persistence_with_deprecated_traits
Alexander Alekhin [Wed, 13 Dec 2017 15:30:21 +0000 (15:30 +0000)]
Merge pull request #10290 from tomoaki0705:fixVS2012Round
Alexander Alekhin [Wed, 13 Dec 2017 15:21:55 +0000 (18:21 +0300)]
core(test): avoid filename duplicates in tests
Tomoaki Teshima [Tue, 12 Dec 2017 12:27:46 +0000 (21:27 +0900)]
fix build error on Visual Studio 2012
* round doesn't exists in standard library of Visual Studio 2012
* apply the correct computation of ROI
Alexander Alekhin [Wed, 13 Dec 2017 13:52:31 +0000 (13:52 +0000)]
Merge pull request #10295 from alalek:cmake_workaround_download_log_issue
Alexander Alekhin [Wed, 13 Dec 2017 13:51:09 +0000 (13:51 +0000)]
Merge pull request #10300 from alalek:cmake_timestamp_update
Alexander Alekhin [Wed, 13 Dec 2017 13:50:24 +0000 (13:50 +0000)]
Merge pull request #10282 from alalek:cmake_tbb_version
Alexander Alekhin [Wed, 13 Dec 2017 12:16:03 +0000 (15:16 +0300)]
cmake: update timestamp status
- avoid unnecessary rebuilding of OpenCV libraries
- use timestamp of the first launch of CMake
- to return to previous behavior use `-UOPENCV_TIMESTAMP` CMake option
Vitaly Tuzov [Wed, 13 Dec 2017 12:00:38 +0000 (15:00 +0300)]
Implementation of bit-exact resize. Internal calls to linear resize updated to use bit-exact version. (#9468)
Alexander Alekhin [Wed, 13 Dec 2017 11:14:28 +0000 (14:14 +0300)]
cmake: rework MAP_IMPORTED_CONFIG to prevent unintended build failures
Introduce CMake variable OPENCV_MAP_IMPORTED_CONFIG which controls this behaviour
Added defaults for MSVS targets
Vadim Pisarevsky [Wed, 13 Dec 2017 07:56:34 +0000 (07:56 +0000)]
Merge pull request #10297 from tomoaki0705:suppressOclWarning
Tomoaki Teshima [Wed, 13 Dec 2017 06:07:51 +0000 (15:07 +0900)]
suppress warnings on OpenCL build
* stop re-enabling the warning C4127
* disabling is done in CMakeLists.txt
Alexander Alekhin [Tue, 12 Dec 2017 15:56:09 +0000 (18:56 +0300)]
cmake: add option to workaround issue observed with old CMake
CMake message contains this: "Invalid escape sequence \"
Alexander Alekhin [Tue, 12 Dec 2017 15:32:57 +0000 (18:32 +0300)]
core: fix copyTo(with mask) dst initialization
Alexander Alekhin [Tue, 12 Dec 2017 14:58:04 +0000 (14:58 +0000)]
Merge pull request #9804 from woodychow:optimize_cveigen
Alexander Alekhin [Tue, 12 Dec 2017 14:07:03 +0000 (17:07 +0300)]
core: fix persistence with deprecated traits
Alexander Alekhin [Mon, 11 Dec 2017 12:22:48 +0000 (15:22 +0300)]
cmake: generate tbb version file
Alexander Alekhin [Mon, 11 Dec 2017 16:04:44 +0000 (19:04 +0300)]
core(test): Core_Eigen.vector_32 use relative norm
Alexander Alekhin [Mon, 11 Dec 2017 16:08:48 +0000 (19:08 +0300)]
core(test): update eps in testEigen<float>: 1e-6 => 1e-4
Alexander Alekhin [Mon, 11 Dec 2017 14:03:44 +0000 (17:03 +0300)]
core(test): refactor Java Eigenvalues/PCA tests
Alexander Alekhin [Mon, 4 Dec 2017 23:28:05 +0000 (02:28 +0300)]
core(test): refactor PCA test
- CV_L2 -> relative NORM_L2
- eigenEps: 1e-6 ==> 1e-4
- evalEps: 1e-6 ==> 1e-5
- evecEps: 1e-3 ==> 5e-3
- RNG seed: 12345
- drop non-informative legacy test code (ts->printf, etc)
Alexander Alekhin [Tue, 12 Dec 2017 09:33:49 +0000 (09:33 +0000)]
Merge pull request #10284 from alalek:dnn_debug_error
Alexander Alekhin [Mon, 11 Dec 2017 20:00:12 +0000 (20:00 +0000)]
dnn: more debug info
Alexander Alekhin [Tue, 12 Dec 2017 08:58:14 +0000 (08:58 +0000)]
Merge pull request #9260 from Cartucho:add_python_signatures
Alexander Alekhin [Mon, 11 Dec 2017 12:48:03 +0000 (12:48 +0000)]
Merge pull request #10269 from terfendail:softdouble_round
Vadim Pisarevsky [Mon, 11 Dec 2017 12:47:52 +0000 (12:47 +0000)]
Merge pull request #10215 from dkurt:dnn_js
Vadim Pisarevsky [Mon, 11 Dec 2017 12:44:21 +0000 (12:44 +0000)]
Merge pull request #10277 from pengli:dnn
Vadim Pisarevsky [Mon, 11 Dec 2017 12:42:45 +0000 (12:42 +0000)]
Merge pull request #10263 from mshabunin:embedded-build
Vadim Pisarevsky [Mon, 11 Dec 2017 12:39:00 +0000 (12:39 +0000)]
Merge pull request #10280 from alalek:python_cv2_to_cv
Vadim Pisarevsky [Mon, 11 Dec 2017 12:34:22 +0000 (12:34 +0000)]
Merge pull request #10231 from alalek:ocl_refactor_program_api
Vitaly Tuzov [Fri, 8 Dec 2017 12:38:22 +0000 (15:38 +0300)]
Added implementation of softdouble rounding to int64_t
Alexander Alekhin [Mon, 11 Dec 2017 09:55:03 +0000 (12:55 +0300)]
python: 'cv2.' -> 'cv.' via 'import cv2 as cv'
Maksim Shabunin [Tue, 28 Nov 2017 09:53:40 +0000 (12:53 +0300)]
Build for embedded systems
Li Peng [Mon, 11 Dec 2017 09:25:41 +0000 (17:25 +0800)]
add command parser to caffe googlenet sample
Signed-off-by: Li Peng <peng.li@intel.com>
Alexander Alekhin [Sun, 10 Dec 2017 20:54:29 +0000 (20:54 +0000)]
Merge pull request #10112 from haritha1313:bugfix_doc1
Alexander Alekhin [Sat, 9 Dec 2017 18:54:52 +0000 (18:54 +0000)]
python: 'sub-module' for binding sources and documentation meta information
Alexander Alekhin [Sat, 9 Dec 2017 04:38:50 +0000 (04:38 +0000)]
doc: finalize Python signatures injection
Haritha [Thu, 7 Dec 2017 06:42:29 +0000 (12:12 +0530)]
crop parameter usage in blobFromImage() calls
tribta [Sun, 19 Nov 2017 21:22:00 +0000 (16:22 -0500)]
add methods
tribta [Fri, 17 Nov 2017 17:56:47 +0000 (12:56 -0500)]
Sample for functions
Alexander Alekhin [Tue, 5 Sep 2017 15:55:01 +0000 (18:55 +0300)]
doc: update Python signatures generation
- drop dependency on 'import cv2', use pyopencv_signatures.json instead
- try to make generator idempotent
tribta [Fri, 28 Jul 2017 19:52:56 +0000 (20:52 +0100)]
Using __doc__ to add Python signatures to the docs.
Alexander Alekhin [Fri, 8 Dec 2017 20:58:16 +0000 (20:58 +0000)]
Merge pull request #10260 from native-api:ffmpeg_load_from_package
Vadim Pisarevsky [Fri, 8 Dec 2017 18:48:48 +0000 (18:48 +0000)]
Merge pull request #10266 from ChristofKaufmann:warp_documentation
Vadim Pisarevsky [Fri, 8 Dec 2017 18:46:49 +0000 (18:46 +0000)]
Merge pull request #10268 from dkurt:fix_scale_layer
Ivan Pozdeev [Thu, 7 Dec 2017 21:12:08 +0000 (00:12 +0300)]
GetModuleFileName() doesn't reset last error on success
Dmitry Kurtaev [Sat, 2 Dec 2017 20:52:35 +0000 (23:52 +0300)]
JavaScript bindings for dnn module
Dmitry Kurtaev [Fri, 8 Dec 2017 14:34:59 +0000 (17:34 +0300)]
Scale layer in case of 2D inputs
Elena Gvozdeva [Fri, 8 Dec 2017 13:36:24 +0000 (16:36 +0300)]
Merge pull request #10172 from ElenaGvozdeva:eg/HAL_sobel
* add HAL for SobelFilter
* add HAL for pyrDown
* add HAL for Scharr
Alexander Alekhin [Fri, 8 Dec 2017 11:35:11 +0000 (11:35 +0000)]
Merge pull request #10253 from alalek:cmake_disable_optimization
Christof Kaufmann [Fri, 8 Dec 2017 11:33:51 +0000 (12:33 +0100)]
doc: Add type restriction note for geometric transformations
This is added according to #7862.
Alexander Alekhin [Fri, 8 Dec 2017 11:20:07 +0000 (11:20 +0000)]
Merge pull request #10255 from dkurt:dnn_roi_pooling
Alexander Alekhin [Fri, 8 Dec 2017 11:02:49 +0000 (11:02 +0000)]
Merge pull request #10264 from terfendail:createsamples
Vitaly Tuzov [Fri, 8 Dec 2017 09:23:44 +0000 (12:23 +0300)]
updated createsamples app utility functions to new c++ API from old c
Alexander Alekhin [Fri, 8 Dec 2017 08:49:10 +0000 (08:49 +0000)]
Merge pull request #10262 from Riyuzakii:10209
Riyuzakii [Fri, 8 Dec 2017 06:40:00 +0000 (12:10 +0530)]
Remove typo/repeated code block
Dmitry Kurtaev [Mon, 4 Dec 2017 15:45:30 +0000 (18:45 +0300)]
ROIPooling layer
Alexander Alekhin [Thu, 7 Dec 2017 15:02:41 +0000 (15:02 +0000)]
Merge pull request #8198 from mshabunin:cmakeex
Alexander Alekhin [Thu, 7 Dec 2017 14:54:37 +0000 (17:54 +0300)]
cmake: CV_DISABLE_OPTIMIZATION disables IPP, OpenCL, Eigen and LAPACK too
Alexander Alekhin [Thu, 7 Dec 2017 12:56:04 +0000 (15:56 +0300)]
ocl: opencl_custom_kernel.cpp example
Alexander Alekhin [Thu, 7 Dec 2017 09:54:18 +0000 (09:54 +0000)]
Merge pull request #10243 from berak:dnn_fix_mobilenet_sample
Alexander Alekhin [Thu, 7 Dec 2017 09:43:16 +0000 (09:43 +0000)]
Merge pull request #10244 from dkurt:issue_10239
Dmitry Kurtaev [Wed, 6 Dec 2017 18:58:36 +0000 (21:58 +0300)]
Fix conv/deconv/fc layers FLOPS computation
Maksim Shabunin [Thu, 7 Dec 2017 07:28:58 +0000 (10:28 +0300)]
Added option to disable gc-sections flag
Pavel Rojtberg [Wed, 25 Jan 2017 11:20:57 +0000 (12:20 +0100)]
Hid symbols in static builds, added LTO flags, removed exports from ts
Alexander Alekhin [Wed, 6 Dec 2017 16:26:51 +0000 (16:26 +0000)]
Merge pull request #10241 from alalek:ios_fix_build_warnings
Alexander Alekhin [Wed, 6 Dec 2017 15:50:12 +0000 (15:50 +0000)]
Merge pull request #10228 from pengli:dnn_new
Alexander Alekhin [Wed, 6 Dec 2017 15:41:18 +0000 (15:41 +0000)]
Merge pull request #10240 from alalek:dnn_perf_ssd
Li Peng [Thu, 30 Nov 2017 12:43:49 +0000 (20:43 +0800)]
detection_output layer ocl implementation
Signed-off-by: Li Peng <peng.li@intel.com>
Li Peng [Wed, 29 Nov 2017 14:35:02 +0000 (22:35 +0800)]
region layer ocl implementation
Signed-off-by: Li Peng <peng.li@intel.com>
Li Peng [Wed, 29 Nov 2017 13:09:31 +0000 (21:09 +0800)]
reorg layer ocl implementation
Signed-off-by: Li Peng <peng.li@intel.com>
Li Peng [Tue, 28 Nov 2017 15:40:46 +0000 (23:40 +0800)]
support axis in concat layer ocl path
Signed-off-by: Li Peng <peng.li@intel.com>
Li Peng [Tue, 5 Dec 2017 13:58:30 +0000 (21:58 +0800)]
reshape layer ocl implementation
Signed-off-by: Li Peng <peng.li@intel.com>
berak [Wed, 6 Dec 2017 13:16:37 +0000 (14:16 +0100)]
dnn: add a check for webcam properties in ssd_mobilenet sample
Alexander Alekhin [Wed, 6 Dec 2017 13:48:51 +0000 (13:48 +0000)]
Merge pull request #10236 from alalek:ocl_skip_unstable_tests
Alexander Alekhin [Wed, 6 Dec 2017 11:51:05 +0000 (14:51 +0300)]
dnn: SSD performance test