Configure C++ Samples With CMake and Build Them in Visual Studio in Windows

Introduction
Requirements
- Required
- Optional
Instructions

Introduction 

This tutorial shows how to configure the Zivid C++ Samples with CMake and then build them with Visual Studio in Windows.

Requirements 

Required 

Zivid Software
Git
Zivid C++ Samples
CMake - version 3.5 or higher

Note

The CMake version that is tested is 3.15.3.

Optional 

Some of the samples depend on external libraries, in particular:

Note

Cloning/installation is covered below.

Instructions 

Step by step instructions is provided with screenshots below.

Install external dependencies 

Install required dependencies based on the sample programs you want to run; see the table below.

Sample program	PCL	Eigen	OpenCV	OpenCV_Contrib
ReadPCLVis3D	YES
CaptureWritePCLVis3D	YES
MaskPointCloud	YES		YES
PoseConversions		YES	YES
UtilizeHandEyeCalibration		YES	YES
CaptureUndistortRGB			YES
CreateDepthMap			YES
TransformPointCloudViaArucoMarker				YES

The Eigen version that is tested is 3.3.7. This version can be downloaded or cloned from eigenteam/eigen-git-mirror. To clone it from GitHub, launch the Command Prompt by pressing Win + R keys on the keyboard, then type cmd and press Enter.

Navigate to a location where you want to clone the repository:

cd C:\Users\Public\Downloads

Then run the following command (install Git if necessary):

git clone https://github.com/eigenteam/eigen-git-mirror

Screenshot of cloning Eigen with Git in cmd.

Install PCL 

The PCL version that is tested is 1.9.1. Download and install prebuilt PCL binaries: pcl_1.9.1. Check the option to install 3rd Party Libraries including OpenNI.

Screenshot of PCL installation wizard on Windows.

Install OpenCV 

The OpenCV version that is tested is 4.1.0. Download and install prebuilt OpenCV binaries: opencv_4.1.0.

Screenshot of OpenCV zip-folder extraction on Windows.

Caution

The prebuilt OpenCV binaries are not sufficient for the samples that depend on OpenCV_Contrib. These samples require building OpenCV with extra modules.

Set PATH system variable for PCL and OpenCV 

Some settings must be included in the PATH system variable before you can build Zivid C++ Samples.

On your PC, navigate to Control Panel → System and Security → System → Advanced System Settings → Environment Variables.

Screenshot of system information on Windows.

Under System variables (bottom of the window), select Path variable and click Edit.

Warning

It is important to click Edit and not New in order not to overwrite the existing value(s) of the PATH system variable!

In the Environment variables window, click New (top-right corner of the window, in the section for user variables) and add the following values:

C:\Program Files\PCL 1.9.1\bin
C:\Program Files\OpenNI2\Tools
C:\Users\Public\Downloads\opencv\build\x64\vc15\bin

Note

Make sure to have the correct path for your PCL and OpenCV installations.

Screenshot which shows how to add paths to the PATH variable.

Clone C++ Samples 

This is our GitHub zivid-cpp-samples repository. To clone it, launch the Command Prompt by pressing Win + R keys on the keyboard, then type cmd and press Enter.

Navigate to a location where you want to clone the repository (install Git if not already installed).

cd C:\Users\Public\Downloads

Then run the following command:

git clone https://github.com/zivid/zivid-cpp-samples

Configure the solution with CMake 

Run CMake (cmake-gui.exe), then set the top two entries:

Where is the source code	`C:/Users/Public/Downloads/zivid-cpp-samples/source`
Where to build the binaries	`C:/Users/Public/Downloads/zivid-cpp-samples/build`

Screenshot of CMake GUI with configuration for Zivid C++ samples.

Click Configure. When prompted, click Yes to create a new build directory.

Screenshot of CMake GUI with a note about creation of build directory.

Set the top two entries:

Specify the generator for this project	Visual Studio 15 2017
Optional platform for generator	x64

Screenshot of CMake GUI where you specify generator.

Click Configure.

Screenshot of CMake GUI where you can configure dependencies.

Note

Some of the samples depend on external libraries, in particular, Eigen 3, PCL, or OpenCV.
If you don’t want to configure those, uncheck USE_EIGEN, USE_OPENCV and USE_PCL, respectively.

Check USE_ARUCO to configure the samples that depend on ArUco library (OpenCV_Contrib).