Question

背景

我正在制作立体相机＆amp; amp;我试图找到两点云的Homography Matrix。当我试图在3D中找到单应性时，它主要失败了。这花费了太多时间。所以我想通过在2D＆amp;中找到Homography来加速这个过程。然后使用ICP算法拟合目标点云。

问题

如何将Homography Matrix从2D图像映射到3D Point云？我见过this answer＆amp;试图改变点云，但它失败了。有人可以指导我，如何在3D中使用2D单应矩阵？如果你分享一些文件会非常有用！

注意：

我的相机已校准＆amp;我有内在函数和＆amp;外在矩阵。所以我可以在3D中找到2D点。

样本数据

1）LeftImage，LeftPointCloud 2）RightImage，RightPointCloud

这是我的简短code。

    UMat mSource_Img, mTarget_Img;
    string leftImageName = "PointCloudFrame_0.bmp";
    string rightImageName = "PointCloudFrame_1.bmp";
    string leftPCloudName = "PointCloudFrame_0.pcd";
    string rightPCloudName = "PointCloudFrame_1.pcd";

    imread(leftImageName, IMREAD_GRAYSCALE).copyTo(mSource_Img);
    if(mSource_Img.empty())
    {
        cout << "Couldn't load " << leftImageName << endl;
        return EXIT_FAILURE;
    }

    imread(rightImageName, IMREAD_GRAYSCALE).copyTo(mTarget_Img);
    if(mTarget_Img.empty())
    {
        cout << "Couldn't load " << rightImageName << endl;
        return EXIT_FAILURE;
    }

    Timer t_H;
    t_H.start();

    SURFFeatureDetector FD_surf;
    Mat H_2D;
    FD_surf.FindHomographyMatrix(mSource_Img,mTarget_Img,H_2D,0,1);

    Mat mTransformed_Img;
    warpPerspective(mSource_Img, mTransformed_Img, H_2D, mTarget_Img.size());

    t_H.stop();
    cout<<"Time taken for Homography Estimation: "<<t_H.getElapsedTimeInMilliSec();

    imshow("mSource_Img", mSource_Img);
    imshow("mTarget_Img", mTarget_Img);
    addWeighted(mTarget_Img,0.5,mTransformed_Img,0.5,1.0,mTransformed_Img);
    imshow("source_Transformed", mTransformed_Img);

    pcl::PointCloud<pcl::PointXYZRGB>::Ptr source_pcd (new pcl::PointCloud<pcl::PointXYZRGB>);  
    pcl::PointCloud<pcl::PointXYZRGB>::Ptr source_transformed_pcd (new pcl::PointCloud<pcl::PointXYZRGB>);  
    pcl::PointCloud<pcl::PointXYZRGB>::Ptr target_pcd (new pcl::PointCloud<pcl::PointXYZRGB>);  
    pcl::io::loadPCDFile (leftPCloudName, *source_pcd);
    pcl::io::loadPCDFile (rightPCloudName, *target_pcd);

    cout<<"\nHomography 2D:\n"<<H_2D<<"\n";

    pcl::visualization::CloudViewer PclViewer("Stereo Point Cloud Viewer");
    Eigen::Matrix4f RotationMat3D=Eigen::Matrix4f::Identity();

    for (int i = 0; i < 3; i++)
    {
        for (int j = 0; j < 3; j++)
        {
            RotationMat3D(i,j)=(float)H_2D.at<double>(i,j);

        }
    }
    cout<<"\Rotation 3D:\n"<<RotationMat3D<<"\n";
    //RotationMat3D*=RotationMat2D;

    pcl::transformPointCloud (*source_pcd, *source_transformed_pcd, RotationMat3D);

    if(!PclViewer.wasStopped())
        PclViewer.showCloud(source_transformed_pcd);

    waitKey(0);

如何将Homography矩阵从2D转换为3D

0 个答案: