This post describes how to install OpenCV with CUDA support on a PC with Ubuntu 14.04. To install OpenCV with CUDA support, CUDA needs to be installed in a first step. If you already have a CUDA installation you can jump to the OpenCV installation.
The installation was tested on Ubuntu 14.04 LTS with CUDA 7.5 and OpenCV 3.0
First Step: Installation of CUDA
The current CUDA version is 7.5. It can be downloaded from the Nvidia developers page (CUDA-Toolkit).
Installation of the CUDA Toolkit
sudo dpkg -i cuda-repo-ubuntu1404_7.5-18_amd64.deb
sudo apt-get update
sudo apt-get install cuda
Environmental Variables
The CUDA toolkit needs to be added to the path and library path so it can be used by other applications. This can be done in the .bashrc file or in the /etc/environment file. In our case it is defined in the /etc/environment file.
PATH="...:/usr/local/cuda/bin"
LD_LIBRARY_PATH="/usr/local/cuda/lib64"
Test Installation
To load the new environmental variables restart the PC. Afterwards it should be possible to generate the provided samples with the command
cuda-install-samples-7.5.sh .
With the example
deviceQuery it is possible to verify the CUDA installation, which is run as followed.
cd NVIDIA-CUDA-7.5_Samples/1_Utilities/deviceQuery
make
./deviceQuery
This should return an output similar to the following.
./deviceQuery Starting...
CUDA Device Query (Runtime API) version (CUDART static linking)
Detected 1 CUDA Capable device(s)
Device 0: "Quadro K2200"
CUDA Driver Version / Runtime Version 7.5 / 7.5
CUDA Capability Major/Minor version number: 5.0
Total amount of global memory: 4095 MBytes (4294246400 bytes)
( 5) Multiprocessors, (128) CUDA Cores/MP: 640 CUDA Cores
GPU Max Clock rate: 1124 MHz (1.12 GHz)
Memory Clock rate: 2505 Mhz
Memory Bus Width: 128-bit
L2 Cache Size: 2097152 bytes
Maximum Texture Dimension Size (x,y,z) 1D=(65536), 2D=(65536, 65536), 3D=(4096, 4096, 4096)
Maximum Layered 1D Texture Size, (num) layers 1D=(16384), 2048 layers
Maximum Layered 2D Texture Size, (num) layers 2D=(16384, 16384), 2048 layers
Total amount of constant memory: 65536 bytes
Total amount of shared memory per block: 49152 bytes
Total number of registers available per block: 65536
Warp size: 32
Maximum number of threads per multiprocessor: 2048
Maximum number of threads per block: 1024
Max dimension size of a thread block (x,y,z): (1024, 1024, 64)
Max dimension size of a grid size (x,y,z): (2147483647, 65535, 65535)
Maximum memory pitch: 2147483647 bytes
Texture alignment: 512 bytes
Concurrent copy and kernel execution: Yes with 1 copy engine(s)
Run time limit on kernels: Yes
Integrated GPU sharing Host Memory: No
Support host page-locked memory mapping: Yes
Alignment requirement for Surfaces: Yes
Device has ECC support: Disabled
Device supports Unified Addressing (UVA): Yes
Device PCI Domain ID / Bus ID / location ID: 0 / 3 / 0
Compute Mode:
< Default (multiple host threads can use ::cudaSetDevice() with device simultaneously) >
deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 7.5, CUDA Runtime Version = 7.5, NumDevs = 1, Device0 = Quadro K2200
Result = PASS
Second Step: Installation of OpenCV
Requirements OpenCV
sudo apt-get install libopencv-dev build-essential checkinstall cmake pkg-config yasm libtiff4-dev libjpeg-dev libjasper-dev libavcodec-dev libavformat-dev libswscale-dev libdc1394-22-dev libxine-dev libgstreamer0.10-dev libgstreamer-plugins-base0.10-dev libv4l-dev python-dev python-numpy libtbb-dev libqt4-dev libgtk2.0-dev libfaac-dev libmp3lame-dev libopencore-amrnb-dev libopencore-amrwb-dev libtheora-dev libvorbis-dev libxvidcore-dev x264 v4l-utils
Get OpenCV Repository
Clone the official OpenCV repository from github. It will create the folder opencv and copy the repository in it.
git clone https://github.com/Itseez/opencv.git
It will take a moment to download the repository.
Build OpenCV
cd opencv
mkdir release
cd release
cmake -D CMAKE_BUILD_TYPE=RELEASE -D CMAKE_INSTALL_PREFIX=/usr/local -D WITH_TBB=ON -D BUILD_NEW_PYTHON_SUPPORT=ON -D WITH_V4L=ON -D INSTALL_C_EXAMPLES=ON -D INSTALL_PYTHON_EXAMPLES=ON -D BUILD_EXAMPLES=ON -D WITH_QT=ON -D WITH_OPENGL=ON -D ENABLE_FAST_MATH=1 -D CUDA_FAST_MATH=1 -D WITH_CUBLAS=1 ..
Check the configuration summary in the end. Make sure that CUDA support is activated with the correct version and CUBLAS is activated as well.
-- Use Cuda: YES (ver 7.5)
-- Use OpenCL: YES
--
-- NVIDIA CUDA
-- Use CUFFT: YES
-- Use CUBLAS: YES
-- USE NVCUVID: NO
-- NVIDIA GPU arch: 20 21 30 35
-- NVIDIA PTX archs: 30
-- Use fast math: YES
Install OpenCV
If the configuration looks fine can the installation be started. The installation can take up to a few hours.
make
sudo make install
Configure the library search path
Create the file /etc/ld.so.conf.d/opencv.conf and add the following line.
/usr/local/lib
Add the following line to the ~/.bashrc file
PKG_CONFIG_PATH=$PKG_CONFIG_PATH:/usr/local/lib/pkgconfig
Reload the bashrc file for the current session
source ~/.bashrc
Sources