CODEWITHSUNDEEP
eigenpoint-cloud-library
shunyo
shunyo

Reputation: 1307

Converting Eigen::MatrixXd to pcl::PointCloud<pcl::PointXYZ>

My question is related to Creating a PCL point cloud using a container of Eigen Vector3d but I am using a Eigen::MatrixXd instead of Eigen::Vector3d. getMatrixXfMap() is not a part of the member function and so cannot be used instead of getVector3fMap(). How can the types be converted in this case?

pcl::PointCloud<pcl::PointXYZ>::Ptr cloud(new pcl::PointCloud<pcl::PointXYZ>);

// resize to number of points
cloud->points.resize(Pts.rows());

/*DOES NOT WORK */
for(int i=0;i<Pts.rows();i++)
    cloud->points[i].getMatrixXfMap() = Pts[i].cast<float>();

Upvotes: 0

Views: 2892

Answers (2)

Michael Jacob Mathew
Michael Jacob Mathew

Reputation: 185

This is what I did. My source_cloud is a Nx3 Matrix constaining XYZ points in float32 format.

typedef pcl::PointCloud<pcl::PointXYZRGBA> PointCloudRGBA;

Eigen::MatrixXf convertToPointCloud(const Eigen::MatrixXf source_cloud){
    PointCloudRGBA::Ptr cloud(new PointCloudRGBA);
    cloud->points.resize(source_cloud.rows());
    for(int i=0; i<source_cloud.rows(); i++){
        cloud->points[i].getVector3fMap() = Eigen::Vector3d(source_cloud(i, 0), source_cloud(i, 1), source_cloud(i, 2)).cast<float>();
    }
}

Upvotes: 0

Ng Oon-Ee
Ng Oon-Ee

Reputation: 1243

It may not be very sexy, but why not just create each point in your for loop? No need for the resize in that case.

pcl::PointXYZ temp;
temp.x = Pts[i][0];
temp.y = Pts[i][1];
temp.z = Pts[i][2];
cloud.push_back(temp);

Upvotes: 1

Related Questions