Alexander Alekhin [Mon, 11 Nov 2019 18:24:05 +0000 (18:24 +0000)]
Merge remote-tracking branch 'upstream/3.4' into merge-3.4
Alexander Alekhin [Sun, 10 Nov 2019 09:35:59 +0000 (09:35 +0000)]
Merge pull request #15875 from gyadam:fix-windows-install-doc
Alexander Alekhin [Sun, 10 Nov 2019 09:25:29 +0000 (09:25 +0000)]
Merge pull request #15877 from mshabunin:fix-gst-relative-34
Alexander Alekhin [Sun, 10 Nov 2019 09:22:45 +0000 (09:22 +0000)]
Merge pull request #15835 from alalek:cmake_cpu_optimizations_fix_15802
Alexander Alekhin [Sun, 10 Nov 2019 09:22:22 +0000 (09:22 +0000)]
Merge pull request #15826 from alalek:cmake_fix_itt_define_condition
Maksim Shabunin [Sat, 9 Nov 2019 11:19:23 +0000 (11:19 +0000)]
Fixed relative paths handling in cap_gstreamer:
Lubov Batanina [Sat, 9 Nov 2019 11:11:09 +0000 (14:11 +0300)]
Merge pull request #15811 from l-bat:eltwise_div
Supported ONNX Squeeze, ReduceL2 and Eltwise::DIV
* Support eltwise div
* Fix test
* OpenCL support added
* refactoring
* fix code style
* Only squeeze with axes supported
Adam Gyarmati [Sat, 9 Nov 2019 03:05:46 +0000 (19:05 -0800)]
Fix Windows installation script error
Fix an error during Windows installation caused by trying to create the already existing Build directory. Also excluding intermediate steps for Install directory creation.
Alexander Alekhin [Fri, 8 Nov 2019 22:01:07 +0000 (22:01 +0000)]
Merge pull request #15861 from dkurt:dnn_fix_get_input_layers
Alexander Alekhin [Fri, 8 Nov 2019 13:20:09 +0000 (13:20 +0000)]
Merge pull request #15319 from cancerberoSgx:fs
Sebastián Gurín [Fri, 16 Aug 2019 02:09:37 +0000 (23:09 -0300)]
expose FS
collinbrake [Fri, 8 Nov 2019 09:27:37 +0000 (04:27 -0500)]
Merge pull request #15832 from collinbrake:feature_grammar_fixes_4
* Grammar fixes for python core operations docs
* fixed whitespace error
* reverted changes
Alexander Alekhin [Thu, 7 Nov 2019 16:48:41 +0000 (16:48 +0000)]
Merge pull request #15848 from alalek:backport_test_15842
Alexander Alekhin [Thu, 7 Nov 2019 14:21:35 +0000 (14:21 +0000)]
Merge pull request #15868 from alalek:issue_15857
Alexander Alekhin [Thu, 7 Nov 2019 14:04:33 +0000 (14:04 +0000)]
Merge pull request #15866 from mshabunin:add-security-md
Alexander Alekhin [Thu, 7 Nov 2019 13:49:34 +0000 (13:49 +0000)]
Merge pull request #15867 from mshabunin:fix-gapi-install
Alexander Alekhin [Thu, 7 Nov 2019 11:01:51 +0000 (14:01 +0300)]
core(persistence): add more checks for implementation limitations
Maksim Shabunin [Thu, 7 Nov 2019 11:03:46 +0000 (14:03 +0300)]
Install: added missing gapi headers
Maksim Shabunin [Thu, 7 Nov 2019 10:10:42 +0000 (13:10 +0300)]
Added SECURITY.md
Dimitri Gerin [Wed, 6 Nov 2019 18:05:35 +0000 (21:05 +0300)]
Fix dnn::getLayerInputs
Alexander Alekhin [Wed, 6 Nov 2019 13:47:37 +0000 (13:47 +0000)]
Merge pull request #15854 from czgdp1807:rem-typo
Alexander Smorkalov [Wed, 6 Nov 2019 13:37:37 +0000 (13:37 +0000)]
Merge pull request #15159 from SSE4:fix_find_openexr
czgdp1807 [Wed, 6 Nov 2019 11:45:31 +0000 (17:15 +0530)]
removed typo
SSE4 [Tue, 5 Nov 2019 09:13:21 +0000 (16:13 +0700)]
- fix FindOpenEXR to respect OPENEXR_ROOT
Signed-off-by: SSE4 <tomskside@gmail.com>
Alexander Alekhin [Wed, 6 Nov 2019 09:54:29 +0000 (09:54 +0000)]
Merge pull request #15814 from i-murzov:3.4-ocl-cleanup
Alexander Alekhin [Wed, 6 Nov 2019 08:12:40 +0000 (08:12 +0000)]
Merge pull request #15813 from i-murzov:3.4-ocl-empty-platform
Igor Murzov [Wed, 30 Oct 2019 15:24:32 +0000 (18:24 +0300)]
Fix OpenCL device detection when some OpenCL platform has no devices
It's not an error if some OpenCL platform has no devices. This makes
OpenCL device detection work correctly in the following scenario:
$ OPENCV_OPENCL_DEVICE=:GPU: ./opencv_test_dnn
OpenCV version: 4.1.2-dev
OpenCV VCS version: 4.1.2-80-g467748ee98-dirty
Build type: Debug
Compiler: /usr/bin/g++ (ver 7.4.0)
Parallel framework: pthreads
CPU features: SSE SSE2 SSE3 *SSE4.1 *SSE4.2 *FP16 *AVX *AVX2 *AVX512-SKX?
Intel(R) IPP version: ippIP AVX2 (l9) 2019.0.0 Gold (-) Jul 24 2018
OpenCL Platforms:
AMD Accelerated Parallel Processing
Portable Computing Language
CPU: pthread-AMD Ryzen 7 2700X Eight-Core Processor (OpenCL 1.2 pocl HSTR: pthread-x86_64-pc-linux-gnu-znver1)
NVIDIA CUDA
dGPU: GeForce GTX 1080 (OpenCL 1.2 CUDA)
Current OpenCL device:
Type = dGPU
Name = GeForce GTX 1080
Version = OpenCL 1.2 CUDA
Driver version = 430.26
Alexander Alekhin [Tue, 5 Nov 2019 16:21:04 +0000 (16:21 +0000)]
Merge pull request #15851 from alalek:fixup_15842
Chip Kerchner [Tue, 5 Nov 2019 15:52:35 +0000 (10:52 -0500)]
Merge pull request #15828 from ChipKerchner:momentsToHal
* Convert moments in tile algorithms to HAL (1.3x faster for VSX).
* Adding NEON code back in for non 64-bit platforms.
* Remove floats from post processing.
TH3CHARLie [Mon, 4 Nov 2019 13:27:48 +0000 (21:27 +0800)]
core(test): add test for YAML parse multiple documents
- added removal of temporary file
Alexander Alekhin [Tue, 5 Nov 2019 15:28:15 +0000 (18:28 +0300)]
core(persistence): fix processing of multiple documents
Alexander Alekhin [Tue, 5 Nov 2019 15:22:42 +0000 (18:22 +0300)]
revert changes in modules/core/src/persistence_yml.cpp (PR15842)
Igor Murzov [Wed, 30 Oct 2019 15:35:38 +0000 (18:35 +0300)]
Simplify OpenCL info dumping code:
* Reduce code nesting
* Drop redundant .c_str() calls
TH3CHARLie [Mon, 4 Nov 2019 13:27:48 +0000 (21:27 +0800)]
Merge pull request #15842 from TH3CHARLie:yaml-fix
* fix yaml parse
* add test for YAML parse multiple documents
* remove trailing whitespace in test
Alexander Alekhin [Mon, 4 Nov 2019 09:58:29 +0000 (09:58 +0000)]
Merge remote-tracking branch 'upstream/3.4' into merge-3.4
Oleg Alexandrov [Sun, 3 Nov 2019 13:37:25 +0000 (05:37 -0800)]
Merge pull request #15838 from oleg-alexandrov:patch-2
Correct stereoRectify documentation
Alexander Alekhin [Sat, 2 Nov 2019 18:50:46 +0000 (18:50 +0000)]
Merge pull request #15834 from berak:python_fix_type_error
Alexander Alekhin [Sat, 2 Nov 2019 10:32:23 +0000 (10:32 +0000)]
cmake(cpu optmizations): fix cleanup of OPENCV_DEPENDANT_TARGETS_* vars
berak [Sat, 2 Nov 2019 07:17:07 +0000 (08:17 +0100)]
python: fix type error msg
Gael Colas [Fri, 1 Nov 2019 19:37:34 +0000 (12:37 -0700)]
Merge pull request #15821 from ColasGael:colasg-viz-color
Fix wrong definition of viz::Color::navy()
Oleg Alexandrov [Fri, 1 Nov 2019 19:34:11 +0000 (12:34 -0700)]
Merge pull request #15820 from oleg-alexandrov:patch-1
Clarify stereoRectify() doc
The function stereoRectify() takes as input a coordinate transform between two cameras. It is ambiguous how it goes. I clarified that it goes from the second camera to the first.
yuriyluxriot [Fri, 1 Nov 2019 19:33:12 +0000 (21:33 +0200)]
Merge pull request #15812 from yuriyluxriot:fls_replaces_tls
* Use FlsAlloc/FlsFree/FlsGetValue/FlsSetValue instead of TlsAlloc/TlsFree/TlsGetValue/TlsSetValue to implment TLS value cleanup when thread has been terminated on Windows Vista and above
* Fix 32-bit build
* Fixed calling convention of cleanup callback
* WINAPI changed to NTAPI
* Use proper guard macro
Chip Kerchner [Fri, 1 Nov 2019 19:30:48 +0000 (15:30 -0400)]
Merge pull request #15555 from ChipKerchner:flipVectorize
* Vectorize flipHoriz and flipVert functions.
* Change v_load_mirror_1 to use vec_revb for VSX
* Only use vec_revb in ISA3.0
* Removing vec_revb code since some of the older compilers don't fully support it.
* Use new v_reverse intrinsic and cleanup code.
* Ensure there are no alignment issues with copies
Alexander Alekhin [Fri, 1 Nov 2019 16:07:15 +0000 (19:07 +0300)]
Merge pull request #15831 from alalek:fix_pylint_master
Alexander Alekhin [Fri, 1 Nov 2019 15:59:35 +0000 (18:59 +0300)]
pylint: eliminate warnings
Alexander Alekhin [Fri, 1 Nov 2019 12:07:49 +0000 (15:07 +0300)]
cmake: fix ITT define condition
Alexander Alekhin [Thu, 31 Oct 2019 20:39:41 +0000 (20:39 +0000)]
Merge pull request #15819 from alalek:ts_unblock_reporting_of_disabled_tests
Alexander Alekhin [Thu, 31 Oct 2019 20:39:16 +0000 (20:39 +0000)]
Merge pull request #15805 from i-murzov:3.4
Ciprian Alexandru Pitis [Thu, 31 Oct 2019 20:38:49 +0000 (21:38 +0100)]
Merge pull request #15799 from Cpitis:feature/parallelization
Parallelize pyrDown & calcSharrDeriv
* ::pyrDown has been parallelized
* CalcSharrDeriv parallelized
* Fixed whitespace
* Set granularity based on amount of threads enabled
* Granularity changed to cv::getNumThreads, now each thread should receive 1/n sized stripes
* imgproc: move PyrDownInvoker<CastOp>::operator() implementation
* imgproc(pyramid): remove syloopboundary()
* video: SharrDerivInvoker replace 'Mat*' => 'Mat&' fields
CJ Smith [Thu, 31 Oct 2019 19:29:04 +0000 (15:29 -0400)]
Merge pull request #15789 from CJSmith-0141:15779-scale-bug-in-stereo-match-sample
* Changes disparity image to float representation
Signed-off-by: Connor James Smith <cjs.connor.smith@gmail.com>
* samples: update disparity multiplier handling in stereo_match.cpp
Oleg Alexandrov [Thu, 31 Oct 2019 19:28:01 +0000 (12:28 -0700)]
Merge pull request #15780 from oleg-alexandrov:master
* Doc bugfix
The documentation page StereoBinaryBM and StereoBinarySGBM says that it returns a disparity that is scaled multiplied by 16. This scaling must be undone before calling reprojectImageTo3D, otherwise the results are wrong. The function reprojectImageTo3D() could do this scaling internally, maybe, but at least the documentation must explain that this has to be done.
* calib3d: update reprojectImageTo3D documentation
* calib3d: add StereoBM/StereoSGBM into notes list
Alexander Alekhin [Thu, 31 Oct 2019 19:24:06 +0000 (19:24 +0000)]
Merge pull request #15478 from terfendail:wintr_stereosgbm
Dizhenin Vlad [Thu, 31 Oct 2019 19:09:33 +0000 (22:09 +0300)]
Merge pull request #15608 from SimpleVlad:3.4
* Add flags for build js
* Add poi.json
* Rebase whitelist into JSON file
* Rework generator of white_list
* Fix small typos
* Transfer opencv_js.josn in opencv_js.config.py
* Edit OPENCV_JS_WHITELIST
* Write comment
* Add description
* Fix typos in desc
* flag's append deleeted
* Fix whitespace
* variable deleted
* fix comment on lines 229 and 235
Alexander Alekhin [Thu, 31 Oct 2019 12:10:42 +0000 (15:10 +0300)]
ts: do not block reporting of launched "DISABLED_" tests
If tests are run through GTest option `--gtest_also_run_disabled_tests`
Alexander Alekhin [Thu, 31 Oct 2019 11:10:45 +0000 (11:10 +0000)]
Merge pull request #15763 from akhakim:dynamic_set_kernel_window_size
Anna Khakimova [Wed, 30 Oct 2019 12:25:32 +0000 (15:25 +0300)]
GAPI Fluid: Dynamic window size
atalaman [Wed, 30 Oct 2019 10:33:39 +0000 (13:33 +0300)]
Merge pull request #15751 from TolyaTalamanov:at/refactor-render-tests
* Refactor render tests
* Fix comment to review
* Move ocv render tests stuff to specific file
* Add OCV prefix for render tests
* Fix comments to review
Igor Murzov [Fri, 25 Oct 2019 16:45:11 +0000 (19:45 +0300)]
Fix wording in some tutorials
Alexander Alekhin [Tue, 29 Oct 2019 20:46:51 +0000 (20:46 +0000)]
Merge remote-tracking branch 'upstream/3.4' into merge-3.4
Alexander Alekhin [Tue, 29 Oct 2019 20:40:35 +0000 (20:40 +0000)]
Merge pull request #15692 from alalek:core_tls_handle_thread_termination
Alexander Alekhin [Tue, 29 Oct 2019 19:10:00 +0000 (19:10 +0000)]
Merge pull request #15791 from alalek:android_camera2_issue_14915
Chip Kerchner [Tue, 29 Oct 2019 10:42:20 +0000 (06:42 -0400)]
Merge pull request #15623 from ChipKerchner:optimizeHOGpipeline
* Use circular lut hustory buffer in computeGradient of HOG
* Initialize prefetch data outside main loop. Avoid code duplication.
Alexander Alekhin [Mon, 28 Oct 2019 20:11:36 +0000 (20:11 +0000)]
Merge pull request #15800 from alalek:issue_15796
Alexander Alekhin [Mon, 28 Oct 2019 18:53:14 +0000 (18:53 +0000)]
gapi: fix opencv_world build
Alexander Alekhin [Sun, 27 Oct 2019 11:14:17 +0000 (11:14 +0000)]
android: use .getRowStride() in JavaCamera2View
Alexander Alekhin [Mon, 28 Oct 2019 13:31:16 +0000 (13:31 +0000)]
Merge pull request #15777 from dbudniko:dbudniko/add_own_empty
Alexander Alekhin [Sun, 27 Oct 2019 22:45:19 +0000 (22:45 +0000)]
Merge pull request #15793 from Cherubin7th:3.4
André Lippok [Sun, 27 Oct 2019 16:43:31 +0000 (17:43 +0100)]
Fixed typo in assertion
Alexander Alekhin [Sat, 26 Oct 2019 21:39:08 +0000 (21:39 +0000)]
Merge pull request #15761 from alalek:core_trace_itt_parameter
Alexander Alekhin [Tue, 22 Oct 2019 13:59:59 +0000 (16:59 +0300)]
core(trace): add ITT control parameter
- OPENCV_TRACE_ITT_ENABLE
Dmitry Budnikov [Fri, 25 Oct 2019 14:50:09 +0000 (17:50 +0300)]
add empty implementation
Alexander Alekhin [Fri, 25 Oct 2019 13:49:46 +0000 (13:49 +0000)]
Merge pull request #15764 from ChipKerchner:demosaicingToHal
Vitaly Tuzov [Thu, 11 Jul 2019 22:34:19 +0000 (01:34 +0300)]
StereoSGBM algorithm updated to use wide universal intrinsics
Alexander Alekhin [Thu, 24 Oct 2019 18:17:40 +0000 (18:17 +0000)]
Merge remote-tracking branch 'upstream/3.4' into merge-3.4
Alexander Alekhin [Thu, 24 Oct 2019 17:58:22 +0000 (17:58 +0000)]
Merge pull request #15770 from terfendail:stereobm_fix
Vitaly Tuzov [Thu, 24 Oct 2019 14:54:49 +0000 (17:54 +0300)]
Fixed StereoBM uniqueness check
Alexander Alekhin [Thu, 24 Oct 2019 07:06:32 +0000 (07:06 +0000)]
Merge pull request #15752 from dkurt:fix_15750
Alexander Alekhin [Sun, 13 Oct 2019 11:14:41 +0000 (11:14 +0000)]
core(tls): implement releasing of TLS on thread termination
- move TLS & instrumentation code out of core/utility.hpp
- (*) TLSData lost .gather() method (to dispose thread data on thread termination)
- use TLSDataAccumulator for reliable collecting of thread data
- prefer using of .detachData() + .cleanupDetachedData() instead of .gather() method
(*) API is broken: replace TLSData => TLSDataAccumulator if gather required
(objects disposal on threads termination is not available in accumulator mode)
Alexander Alekhin [Wed, 23 Oct 2019 17:07:32 +0000 (17:07 +0000)]
Merge pull request #15736 from akhakim:fix_for_ADE-859_using_namespace_cv_G_TYPED_KERNEL_fail_c++
Alexander Alekhin [Wed, 23 Oct 2019 17:02:22 +0000 (17:02 +0000)]
Merge pull request #15754 from float13:patch-1
ChipKerchner [Wed, 23 Oct 2019 15:47:07 +0000 (10:47 -0500)]
Convert demosaic functions to HAL
Alexander Alekhin [Wed, 23 Oct 2019 12:05:28 +0000 (12:05 +0000)]
Merge pull request #15762 from rgarnov:return_exported_mkdatanode
float13 [Tue, 22 Oct 2019 01:36:40 +0000 (21:36 -0400)]
Tutorial - Make required input args positional.
I think it would help to change all 3 of the the input file arguments to be "positional" for consistency with the other tutorials. This also simplifies the command line input to run this tutorial by reducing typing, and helpfully prints the "usage" info if any of the 3 required inputs are missing.
I'm new to OpenCV and working through the tutorials. I kept getting runtime errors with this one until I realized that the arguments weren't positional, and I was missing the "--input1", "--input2, "--input3" flags preceding the filenames. All of the previous tutorials had required filenames as positional arguments and didn't require this.
The original code would require each input to be specified like this:
./compareHist_Demo --input1 filename1 --input2 filename2 --input3 filename3
But with this change, the above command is simplified to:
./compareHist_Demo filename1 filename2 filename3
This avoids a confusing runtime error to make things simpler for newcomers like me :)
Anna Khakimova [Wed, 23 Oct 2019 09:17:49 +0000 (12:17 +0300)]
GAPI:Fix for using cv makes G_TYPED_KERNEL fail
Alexander Alekhin [Tue, 22 Oct 2019 21:40:10 +0000 (21:40 +0000)]
Merge pull request #15759 from alalek:issue_15758
Ruslan Garnov [Tue, 22 Oct 2019 15:57:31 +0000 (18:57 +0300)]
Returned GModel::mkDataNode() overload for external backends
Chip Kerchner [Tue, 22 Oct 2019 13:37:20 +0000 (09:37 -0400)]
Merge pull request #15738 from ChipKerchner:bugInt64x2Comparison
Fixing bug with comparison of v_int64x2 or v_uint64x2
* Casting v_uint64x2 to v_float64x2 and comparing does NOT work in all cases. Rewrite using epi64 instructions - faster too.
* Fix bad merge.
* Fix equal comparsion for non-SSE4.1. Add test cases for v_int64x2 comparisons.
* Try to fix merge conflict.
* Only test v_int64x2 comparisons if CV_SIMD_64F
* Fix compiler warning.
Alexander Alekhin [Tue, 22 Oct 2019 13:31:40 +0000 (16:31 +0300)]
gapi: fix build
- gcc 4.8.4 (ARMv7)
Alexander Alekhin [Tue, 22 Oct 2019 12:52:13 +0000 (12:52 +0000)]
Merge pull request #15739 from dkurt:darknet_shortcut_asymm
Alexander Alekhin [Tue, 22 Oct 2019 12:51:29 +0000 (12:51 +0000)]
Merge pull request #15756 from dankamongmen:dankamongmen/stitching_surf_default
nick black [Tue, 22 Oct 2019 06:16:29 +0000 (02:16 -0400)]
stitching_detailed: use correct match_conf default
The usage function states that the default for match_conf is
0.65 if the default SURF feature finder is used, and 0.3 for
orbs. Indeed, if --feature orbs is used, match_conf is set
to 0.3f. This is a NOP, because the real default is also set
to 0.3f. Change it to 0.65f when SURF is in play.
anton-potapov [Mon, 21 Oct 2019 19:33:18 +0000 (22:33 +0300)]
Merge pull request #15735 from anton-potapov:gapi_async_documentaion
* G-API: Doxygen documentatation for Async API
* G-API: Doxygen documentatation for Async API
- renamed local variable (reading parameter async) async ->
asyncNumReq in object_detection DNN sample
to avoid Doxygen erroneous linking the sample to cv::gapi::wip::async
documentation
Dmitry Kurtaev [Mon, 21 Oct 2019 16:09:44 +0000 (19:09 +0300)]
Fix custom IE layers in case of no MKLDNN plugin
Yashas Samaga B L [Mon, 21 Oct 2019 11:28:00 +0000 (16:58 +0530)]
Merge pull request #14827 from YashasSamaga:cuda4dnn-csl-low
CUDA backend for the DNN module
* stub cuda4dnn design
* minor fixes for tests and doxygen
* add csl public api directory to module headers
* add low-level CSL components
* add high-level CSL components
* integrate csl::Tensor into backbone code
* switch to CPU iff unsupported; otherwise, fail on error
* add fully connected layer
* add softmax layer
* add activation layers
* support arbitary rank TensorDescriptor
* pass input wrappers to `initCUDA()`
* add 1d/2d/3d-convolution
* add pooling layer
* reorganize and refactor code
* fixes for gcc, clang and doxygen; remove cxx14/17 code
* add blank_layer
* add LRN layer
* add rounding modes for pooling layer
* split tensor.hpp into tensor.hpp and tensor_ops.hpp
* add concat layer
* add scale layer
* add batch normalization layer
* split math.cu into activations.cu and math.hpp
* add eltwise layer
* add flatten layer
* add tensor transform api
* add asymmetric padding support for convolution layer
* add reshape layer
* fix rebase issues
* add permute layer
* add padding support for concat layer
* refactor and reorganize code
* add normalize layer
* optimize bias addition in scale layer
* add prior box layer
* fix and optimize normalize layer
* add asymmetric padding support for pooling layer
* add event API
* improve pooling performance for some padding scenarios
* avoid over-allocation of compute resources to kernels
* improve prior box performance
* enable layer fusion
* add const layer
* add resize layer
* add slice layer
* add padding layer
* add deconvolution layer
* fix channelwise ReLU initialization
* add vector traits
* add vectorized versions of relu, clipped_relu, power
* add vectorized concat kernels
* improve concat_with_offsets performance
* vectorize scale and bias kernels
* add support for multi-billion element tensors
* vectorize prior box kernels
* fix address alignment check
* improve bias addition performance of conv/deconv/fc layers
* restructure code for supporting multiple targets
* add DNN_TARGET_CUDA_FP64
* add DNN_TARGET_FP16
* improve vectorization
* add region layer
* improve tensor API, add dynamic ranks
1. use ManagedPtr instead of a Tensor in backend wrapper
2. add new methods to tensor classes
- size_range: computes the combined size of for a given axis range
- tensor span/view can be constructed from a raw pointer and shape
3. the tensor classes can change their rank at runtime (previously rank was fixed at compile-time)
4. remove device code from tensor classes (as they are unused)
5. enforce strict conditions on tensor class APIs to improve debugging ability
* fix parametric relu activation
* add squeeze/unsqueeze tensor API
* add reorg layer
* optimize permute and enable 2d permute
* enable 1d and 2d slice
* add split layer
* add shuffle channel layer
* allow tensors of different ranks in reshape primitive
* patch SliceOp to allow Crop Layer
* allow extra shape inputs in reshape layer
* use `std::move_backward` instead of `std::move` for insert in resizable_static_array
* improve workspace management
* add spatial LRN
* add nms (cpu) to region layer
* add max pooling with argmax ( and a fix to limits.hpp)
* add max unpooling layer
* rename DNN_TARGET_CUDA_FP32 to DNN_TARGET_CUDA
* update supportBackend to be more rigorous
* remove stray include from preventing non-cuda build
* include op_cuda.hpp outside condition #if
* refactoring, fixes and many optimizations
* drop DNN_TARGET_CUDA_FP64
* fix gcc errors
* increase max. tensor rank limit to six
* add Interp layer
* drop custom layers; use BackendNode
* vectorize activation kernels
* fixes for gcc
* remove wrong assertion
* fix broken assertion in unpooling primitive
* fix build errors in non-CUDA build
* completely remove workspace from public API
* fix permute layer
* enable accuracy and perf. tests for DNN_TARGET_CUDA
* add asynchronous forward
* vectorize eltwise ops
* vectorize fill kernel
* fixes for gcc
* remove CSL headers from public API
* remove csl header source group from cmake
* update min. cudnn version in cmake
* add numerically stable FP32 log1pexp
* refactor code
* add FP16 specialization to cudnn based tensor addition
* vectorize scale1 and bias1 + minor refactoring
* fix doxygen build
* fix invalid alignment assertion
* clear backend wrappers before allocateLayers
* ignore memory lock failures
* do not allocate internal blobs
* integrate NVTX
* add numerically stable half precision log1pexp
* fix indentation, following coding style, improve docs
* remove accidental modification of IE code
* Revert "add asynchronous forward"
This reverts commit
1154b9da9da07e9b52f8a81bdcea48cf31c56f70.
* [cmake] throw error for unsupported CC versions
* fix rebase issues
* add more docs, refactor code, fix bugs
* minor refactoring and fixes
* resolve warnings/errors from clang
* remove haveCUDA() checks from supportBackend()
* remove NVTX integration
* changes based on review comments
* avoid exception when no CUDA device is present
* add color code for CUDA in Net::dump
Alexander Alekhin [Sun, 20 Oct 2019 13:08:40 +0000 (13:08 +0000)]
Merge pull request #15747 from TH3CHARLie:imshow-fix
Alexander Alekhin [Sun, 20 Oct 2019 13:08:28 +0000 (13:08 +0000)]
Merge pull request #15743 from collinbrake:feature_grammar_fixes_2
Alexander Alekhin [Sun, 20 Oct 2019 13:06:27 +0000 (13:06 +0000)]
Merge pull request #14518 from SSteve:intersectConvexConvex-example
Steve Nicholson [Thu, 9 May 2019 00:54:17 +0000 (17:54 -0700)]
Add documentation and example program for intersectConvexConvex
TH3CHARLie [Sat, 19 Oct 2019 12:34:18 +0000 (20:34 +0800)]
fix incorrect imshow behavior
Dmitry Kurtaev [Sat, 19 Oct 2019 09:54:15 +0000 (12:54 +0300)]
Fix Darknet eltwise