Windows 10上のソースコードはOpenCVをコンパイルしてCUDA|compile opencv with CUDAをサポートする
16603 ワード
この記事は個人ブログからhttps://kezunlin.me/post/6580691f/あ、お読みください.
compile opencv with CUDA support on windows 10
Series Part 1: compile opencv on ubuntu 16.04 Part 2: compile opencv with CUDA support on windows 10 Part 3: opencv mat for loop Part 4: speed up opencv image processing with openmp
Guide
requirements: windows: 10 opencv: 3.1.0 nvidia driver: gtx 1060 382.05 (gtx 970m) GPU arch(s): sm_61 (sm_52) cuda: 8.0 cudnn: 5.0.5 cmake: 3.10.0 vs: vs2015 64
nvidia cuda CC
see cuda compute capacity
ノートパソコンのグラフィックスとデスクトップのコンピューティング能力には差があります.
cpu vs gpu
for opencv functions
get source
Get opencv 3.1.0 for git and fix some bugs
compile
config
configure with
for opencv
for caffe
the
複数の異なるコンピューティング能力のGPUで実行可能なプログラムをサポートするためにopencv/caffeコンパイル中に複数の異なるアーキテクチャをサポートする必要があります.
configure and output:
Notice for
build
Open
this may take hours to finish.
errors
possible solutions
With
I actually got it to work both on my laptop and my desktop (GTX960M and GTX970 respectively) running with OpenCV 3.2 and the latest version of CUDA 8.0 for Win10 in Visual Studio 15 Community! What I did was to enable
my solution:
configure and build again. this time cost only about 1 minutes.
after error fixed,build results
demo
cuda-module
GPU modules includes class
The GPU module is designed as a host-level API. This means that if you have pre-compiled OpenCV GPU binaries, you are not required to have the CUDA Toolkit installed or write any extra code to make use of the GPU.
CMakeLists.txt
demo.cpp
In the sample below an image is loaded from local file, next it is uploaded to GPU, thresholded, downloaded and displayed.
cpu vs gpu time cost(1)解像度が特に大きくないピクチャ間のORB特徴マッチングについては、GPU版よりもCPU演算が速い(GPUへのアップロードに時間がかかるため) .(2)しかし解像度の大きい画像に対して、あるいはGPUがCPUより良い機械(例えばNvidia Jetsonシリーズ)に対して、GPU版のORBアルゴリズムはCPU版のプログラムよりもっと効率的です.
problems
(1)cudaバージョンのopencv caffeネットワークを用いた最初の作成は非常に時間がかかり,後のネットワークの作成は非常に速い.
(2)opencvのgpuコードはcpuコードより遅く,初回起動に20 s程度かかる.(事実はコンパイルされたcaffeとGPUの計算力の不一致によるものである)
reasons
Your problem is that CUDA needs to initialize! And it will generally takes between serveral seconds
Why first function call is slow?That is because of initialization overheads. On first GPU function call
The first gpu function call is always takes more time, because CUDA initialize context for device. The following calls will be faster.
Not Reasons:(1) CPU clockspeed is 10x faster than GPU clockspeed.(2) memory transfer times between host (CPU) and device (GPU) (upload,downloa data)
deploy
runtime errors
gtx 1060コンパイルopencv caffe gtx 970 mで実行中にエラーが発生しました
im 2 colはcaffeのソースファイルであり、
reasons
see what-is-the-purpose-of-using-multiple-arch-flags-in-nvidias-nvcc-compiler
Roughly speaking, the code compilation flow goes like this:
The virtual architecture (e.g.
win7/win10 deploy compile opencv caffe on windows 10 for GTX 1060 deoply on windows 7 for GTX 1080 Ti successfully
for win7, if we install
Solutions: use older drivers
linux/window performance
(1)apiはlinuxで平均3 msかかる.同じコードはwindowsで平均14 msかかります
(2)vsコンパイルオープンコード最適化前後の性能差は5倍近く,125 ms vs 25 ms
(3)cmakeコンパイルRELEASEオプションデフォルトでコード最適化-O 3がオン
Reference Building OpenCV for Tegra with CUDA opencv with cuda introduction ref doc for opencv gpu opencv 3.1.0 doc std::tuple errors when Building OpenCV (Main Branch) for Microsoft VS 2015 (x64) opencv github issues building-opencv-3-cuda-errors why-opencv-gpu-code-is-slower-than-cpu huge-time-to-upload-data-to-gpu https://github.com/floydhub/dl-docker/issues/12 Cuda kernel failed. Error: invalid device function offical cuda-c-programming-guide what-is-the-purpose-of-using-multiple-arch-flags-in-nvidias-nvcc-compiler
History 20180713: created.
Copyright Post author: kezunlin Post link: https://kezunlin.me/post/6580691f/ Copyright Notice: All articles in this blog are licensed under CC BY-NC-SA 3.0 unless stating additionally.
compile opencv with CUDA support on windows 10
Series
Guide
requirements:
nvidia cuda CC
see cuda compute capacity
ノートパソコンのグラフィックスとデスクトップのコンピューティング能力には差があります.
cpu vs gpu
for opencv functions
get source
Get opencv 3.1.0 for git and fix some bugs
git clone https://github.com/opencv/opencv.git
cd opencv
git checkout -b v3.1.0 3.1.0
# fix bugs for 3.1.0
git cherry-pick 10896
git cherry-pick cdb9c
git cherry-pick 24dbb
git branch
master
* v3.1.0
compile
mkdir build && cd build && cmake-gui ..
config
configure with
VS 2015 win64 with options BUILD_SHARED_LIBS ON
CMAKE_CONFIGURATION_TYPES Release # Release
CMAKE_CXX_FLAGS_RELEASE /MD /O2 /Ob2 /DNDEBUG /MP # for multiple processor
WITH_VTK OFF
BUILD_PERF_TESTS OFF # if ON, build errors occur
WITH_CUDA ON
CUDA_TOOLKIT_ROOT_DIR C:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v8.0
#CUDA_ARCH_BIN 3.0 3.5 5.0 5.2 6.0 6.1 # very time-consuming
CUDA_ARCH_PTX 3.0
for opencv
CUDA_ARCH_BIN 3.0 3.5 5.0 5.2 6.0 6.1 relate with -gencode;arch=compute_30,code=sm_30;-gencode;arch=compute_35,code=sm_35;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_52,code=sm_52;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;
CUDA_ARCH_PTX 3.0 relate with -gencode;arch=compute_30,code=compute_30;
for caffe
the
CUDA_ARCH_BIN parameter specifies multiple architectures so as to support a variety of GPU boards. otherwise, cuda programs will not run with other type of GPU boards. 複数の異なるコンピューティング能力のGPUで実行可能なプログラムをサポートするためにopencv/caffeコンパイル中に複数の異なるアーキテクチャをサポートする必要があります.
eg. CUDA_ARCH_BIN 3.0 3.5 5.0 5.2 6.0 6.1であるため、コンパイルプロセスは非常に時間がかかります.コンパイル中にreleaseバージョンをリリースする必要があるGPUアーキテクチャをできるだけ選択して構成コンパイルを行います.configure and output:
Selecting Windows SDK version 10.0.14393.0 to target Windows 10.0.17134.
found IPP (ICV version): 9.0.1 [9.0.1]
at: C:/compile/opencv/3rdparty/ippicv/unpack/ippicv_win
CUDA detected: 8.0
CUDA NVCC target flags: -gencode;arch=compute_30,code=sm_30;-gencode;arch=compute_30,code=compute_30
Could NOT find Doxygen (missing: DOXYGEN_EXECUTABLE)
To enable PlantUML support, set PLANTUML_JAR environment variable or pass -DPLANTUML_JAR= option to cmake
Could NOT find PythonInterp: Found unsuitable version "1.4", but required is at least "3.4" (found C:/Users/zunli/.babun/cygwin/bin/python)
Could NOT find PythonInterp: Found unsuitable version "1.4", but required is at least "3.2" (found C:/Users/zunli/.babun/cygwin/bin/python)
Could NOT find Matlab (missing: MATLAB_MEX_SCRIPT MATLAB_INCLUDE_DIRS MATLAB_ROOT_DIR MATLAB_LIBRARIES MATLAB_LIBRARY_DIRS MATLAB_MEXEXT MATLAB_ARCH MATLAB_BIN)
General configuration for OpenCV 3.1.0 =====================================
Version control: 3.1.0-3-g5e9beb8
Platform:
Host: Windows 10.0.17134 AMD64
CMake: 3.10.0
CMake generator: Visual Studio 14 2015 Win64
CMake build tool: C:/Program Files (x86)/MSBuild/14.0/bin/MSBuild.exe
MSVC: 1900
C/C++:
Built as dynamic libs?: YES
C++ Compiler: C:/Program Files (x86)/Microsoft Visual Studio 14.0/VC/bin/x86_amd64/cl.exe (ver 19.0.24215.1)
C++ flags (Release): /DWIN32 /D_WINDOWS /W4 /GR /EHa /D _CRT_SECURE_NO_DEPRECATE /D _CRT_NONSTDC_NO_DEPRECATE /D _SCL_SECURE_NO_WARNINGS /Gy /bigobj /Oi /wd4251 /wd4324 /wd4275 /wd4589 /MP8 /MD /O2 /Ob2 /DNDEBUG /MP /Zi
C++ flags (Debug): /DWIN32 /D_WINDOWS /W4 /GR /EHa /D _CRT_SECURE_NO_DEPRECATE /D _CRT_NONSTDC_NO_DEPRECATE /D _SCL_SECURE_NO_WARNINGS /Gy /bigobj /Oi /wd4251 /wd4324 /wd4275 /wd4589 /MP8 /MDd /Zi /Ob0 /Od /RTC1
C Compiler: C:/Program Files (x86)/Microsoft Visual Studio 14.0/VC/bin/x86_amd64/cl.exe
C flags (Release): /DWIN32 /D_WINDOWS /W3 /D _CRT_SECURE_NO_DEPRECATE /D _CRT_NONSTDC_NO_DEPRECATE /D _SCL_SECURE_NO_WARNINGS /Gy /bigobj /Oi /MP8 /MD /O2 /Ob2 /DNDEBUG /Zi
C flags (Debug): /DWIN32 /D_WINDOWS /W3 /D _CRT_SECURE_NO_DEPRECATE /D _CRT_NONSTDC_NO_DEPRECATE /D _SCL_SECURE_NO_WARNINGS /Gy /bigobj /Oi /MP8 /MDd /Zi /Ob0 /Od /RTC1
Linker flags (Release): /machine:x64 /INCREMENTAL:NO /debug
Linker flags (Debug): /machine:x64 /debug /INCREMENTAL
Precompiled headers: YES
Extra dependencies: comctl32 gdi32 ole32 setupapi ws2_32 vfw32 cudart nppc nppi npps cufft -LC:/Program Files/NVIDIA GPU Computing Toolkit/CUDA/v8.0/lib/x64
3rdparty dependencies: zlib libjpeg libwebp libpng libtiff libjasper IlmImf
OpenCV modules:
To be built: cudev core cudaarithm flann imgproc ml video cudabgsegm cudafilters cudaimgproc cudawarping imgcodecs photo shape videoio cudacodec highgui objdetect ts features2d calib3d cudafeatures2d cudalegacy cudaobjdetect cudaoptflow cudastereo stitching superres videostab python2
Disabled: world
Disabled by dependency: -
Unavailable: java python3 viz
Windows RT support: NO
GUI:
QT: NO
Win32 UI: YES
OpenGL support: NO
VTK support: NO
Media I/O:
ZLib: build (ver 1.2.8)
JPEG: build (ver 90)
WEBP: build (ver 0.3.1)
PNG: build (ver 1.6.19)
TIFF: build (ver 42 - 4.0.2)
JPEG 2000: build (ver 1.900.1)
OpenEXR: build (ver 1.7.1)
GDAL: NO
Video I/O:
Video for Windows: YES
DC1394 1.x: NO
DC1394 2.x: NO
FFMPEG: YES (prebuilt binaries)
codec: YES (ver 56.41.100)
format: YES (ver 56.36.101)
util: YES (ver 54.27.100)
swscale: YES (ver 3.1.101)
resample: NO
gentoo-style: YES
GStreamer: NO
OpenNI: NO
OpenNI PrimeSensor Modules: NO
OpenNI2: NO
PvAPI: NO
GigEVisionSDK: NO
DirectShow: YES
Media Foundation: NO
XIMEA: NO
Intel PerC: NO
Parallel framework: Concurrency
Other third-party libraries:
Use IPP: 9.0.1 [9.0.1]
at: C:/compile/opencv/3rdparty/ippicv/unpack/ippicv_win
Use IPP Async: NO
Use Eigen: NO
Use Cuda: YES (ver 8.0)
Use OpenCL: YES
Use custom HAL: NO
NVIDIA CUDA
Use CUFFT: YES
Use CUBLAS: NO
USE NVCUVID: NO
NVIDIA GPU arch: 30 35 50 52 60 61
NVIDIA PTX archs: 30
Use fast math: NO
OpenCL:
Version: dynamic
Include path: C:/compile/opencv/3rdparty/include/opencl/1.2
Use AMDFFT: NO
Use AMDBLAS: NO
Python 2:
Interpreter: C:/Python27/python.exe (ver 2.7.13)
Libraries: C:/Python27/libs/python27.lib (ver 2.7.13)
numpy: C:/Python27/lib/site-packages/numpy/core/include (ver 1.11.3)
packages path: C:/Python27/Lib/site-packages
Python 3:
Interpreter: NO
Python (for build): C:/Python27/python.exe
Java:
ant: NO
JNI: C:/Program Files/Java/jdk1.8.0_161/include C:/Program Files/Java/jdk1.8.0_161/include/win32 C:/Program Files/Java/jdk1.8.0_161/include
Java wrappers: NO
Java tests: NO
Matlab: Matlab not found or implicitly disabled
Documentation:
Doxygen: NO
PlantUML: NO
Tests and samples:
Tests: YES
Performance tests: NO
C/C++ Examples: NO
Install path: C:/compile/opencv/build/install
cvconfig.h is in: C:/compile/opencv/build
-----------------------------------------------------------------
Configuring done
Generating done
Notice for
gencode CUDA NVCC target flags: -gencode;arch=compute_30,code=sm_30;-gencode;arch=compute_35,code=sm_35;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_52,code=sm_52;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_30,code=compute_30
build
Open
OpenCV.sln with VS 2015 and build release version. this may take hours to finish.
errors
possible solutions
With
BUILD_PERF_TESTS and BUILD_TESTS disabled, I managed to build OpenCV 3.1 with CUDA 8.0 on Windows 10 with VS2015 x64 arch target. Without building test/performance modules, the build process costs less time as well : ) I actually got it to work both on my laptop and my desktop (GTX960M and GTX970 respectively) running with OpenCV 3.2 and the latest version of CUDA 8.0 for Win10 in Visual Studio 15 Community! What I did was to enable
WITH_CUBLAS aswell as WITH_CUDA . I also turned off BUILD_PERF_TESTS and BUILD_TESTS . The configuration was built using the Visual Studio 14 2015 C++ compiler. my solution:
disable `BUILD_PERF_TESTS`
configure and build again. this time cost only about 1 minutes.
after error fixed,build results
demo
cuda-module
OpenCV GPU module is written using CUDA , therefore it benefits from the CUDA ecosystem. GPU modules includes class
cv::cuda::GpuMat which is a primary container for data kept in GPU memory. It’s interface is very similar with cv::Mat , its CPU counterpart. All GPU functions receive GpuMat as input and output arguments. This allows to invoke several GPU algorithms without downloading data. GPU module API interface is also kept similar with CPU interface where possible. So developers who are familiar with Opencv on CPU could start using GPU straightaway. The GPU module is designed as a host-level API. This means that if you have pre-compiled OpenCV GPU binaries, you are not required to have the CUDA Toolkit installed or write any extra code to make use of the GPU.
CMakeLists.txt
find_package(OpenCV REQUIRED COMPONENTS core highgui imgproc features2d calib3d
cudaarithm cudabgsegm cudafilters cudaimgproc cudawarping cudafeatures2d # for cuda-enabled
) #
MESSAGE( [Main] " OpenCV_INCLUDE_DIRS = ${OpenCV_INCLUDE_DIRS}")
MESSAGE( [Main] " OpenCV_LIBS = ${OpenCV_LIBS}")demo.cpp
In the sample below an image is loaded from local file, next it is uploaded to GPU, thresholded, downloaded and displayed.
#include
#include
#include
#include
#include
#include
int test_opencv_gpu()
{
try
{
cv::Mat src_host = cv::imread("file.png", CV_LOAD_IMAGE_GRAYSCALE);
cv::cuda::GpuMat dst, src;
src.upload(src_host);
cv::cuda::threshold(src, dst, 128.0, 255.0, CV_THRESH_BINARY);
cv::Mat result_host;
dst.download(result_host);
cv::imshow("Result", result_host);
cv::waitKey();
}
catch (const cv::Exception& ex)
{
std::cout << "Error: " << ex.what() << std::endl;
}
return 0;
} cpu vs gpu time cost
problems
(1)cudaバージョンのopencv caffeネットワークを用いた最初の作成は非常に時間がかかり,後のネットワークの作成は非常に速い.
(2)opencvのgpuコードはcpuコードより遅く,初回起動に20 s程度かかる.(事実はコンパイルされたcaffeとGPUの計算力の不一致によるものである)
reasons
Your problem is that CUDA needs to initialize! And it will generally takes between serveral seconds
Why first function call is slow?That is because of initialization overheads. On first GPU function call
Cuda Runtime API is initialized implicitly. The first gpu function call is always takes more time, because CUDA initialize context for device. The following calls will be faster.
Not Reasons:(1) CPU clockspeed is 10x faster than GPU clockspeed.(2) memory transfer times between host (CPU) and device (GPU) (upload,downloa data)
deploy
runtime errors
gtx 1060コンパイルopencv caffe gtx 970 mで実行中にエラーが発生しました
im2col.cu Check failed: error == cudaSuccess (8 vs. 0) invalid device function gtx 1060 sm_61
gtx 970m sm_52
im 2 colはcaffeのソースファイルであり、
gtx 970mの計算能力が実行可能なファイルの実行をサポートしていないことを示している.reasons
see what-is-the-purpose-of-using-multiple-arch-flags-in-nvidias-nvcc-compiler
Roughly speaking, the code compilation flow goes like this:
CUDA C/C++ device code source --> PTX --> SASS The virtual architecture (e.g.
compute_20 , whatever is specified by -arch compute ...) determines what type of PTX code will be generated. The additional switches (e.g. -code sm_21 ) determine what type of SASS code will be generated. SASS is actually executable object code for a GPU (machine language). An executable can contain multiple versions of SASS and/or PTX, and there is a runtime loader mechanism that will pick appropriate versions based on the GPU actually being used. win7/win10 deploy
for win7, if we install
398.82-desktop-win8-win7-64bit-international-whql.exe ,errors may occur: > nvidia-smi.exe
Failed to initialize NVML: Unknown error
Solutions: use older drivers
385.69 linux/window performance
(1)apiはlinuxで平均3 msかかる.同じコードはwindowsで平均14 msかかります
(2)vsコンパイルオープンコード最適化前後の性能差は5倍近く,125 ms vs 25 ms
(3)cmakeコンパイルRELEASEオプションデフォルトでコード最適化-O 3がオン
Reference
History
Copyright