Revision history [back]

Hi. Classical situation:

cv::Mat Rvec;
  cv::Mat_<float> Tvec;
  cv::Mat raux,taux;

  cv::solvePnP(points3d, points2d, calibration.getIntrinsic(), calibration.getDistorsion(),raux,taux);
  raux.convertTo(Rvec,CV_32F);
  taux.convertTo(Tvec ,CV_32F);

  cv::Mat_<float> rotMat(3,3); 
  cv::Rodrigues(Rvec, rotMat);

  // Copy to transformation matrix
  for (int col=0; col<3; col++)
  {
    for (int row=0; row<3; row++)
    {        
     pose3d.r().mat[row][col] = rotMat(row,col); // Copy rotation component
    }
    pose3d.t().data[col] = Tvec(col); // Copy translation component
  }

  pose3d = pose3d.getInverted();

And after it it's easy to draw something in OpenGL with setting MV matrix as pose3d (Matrix4x4). But, i getting crazy to find solution for getting camera position and rotation in world coordinate system to setup camera without setting MV matrix. This is required for drawing camera position in Unity editor application (e.g. here is the image, here is the photo camera and here is the player = scene render camera ). Thank you, folks. It's the biggest stuck for me as a programmer ever :(