0

I work on 2d image point and 3d model point correspondences. Firstly, I use below code : 

cv::Mat cameraMatrix(3, 3, cv::DataType<double>::type);

cv::setIdentity(cameraMatrix);

std::cout << "Initial cameraMatrix: " << cameraMatrix << std::endl;

cv::Mat distCoeffs(4, 1, cv::DataType<double>::type);
distCoeffs.at<double>(0) = 0;
distCoeffs.at<double>(1) = 0;
distCoeffs.at<double>(2) = 0;
distCoeffs.at<double>(3) = 0;

cv::Mat rvec(3, 1, cv::DataType<double>::type);
cv::Mat tvec(3, 1, cv::DataType<double>::type);

cv::solvePnP(objectPoints, imagePoints, cameraMatrix, distCoeffs, rvec, tvec);

std::cout << "rvec: " << rvec << std::endl;
std::cout << "tvec: " << tvec << std::endl;

std::vector<cv::Point2f> projectedPoints;
cv::projectPoints(objectPoints, rvec, tvec, cameraMatrix, distCoeffs, projectedPoints);

I expect that positions of projectedPoints[i] is equal to position of imagepoints[i]. But when I write their values, they are different.

for (unsigned int i = 0; i < projectedPoints.size(); ++i)
{
    std::cout << "Image point: " << imagePoints[i] << " Projected to " << projectedPoints[i] << std::endl;
}

enter image description here

Why these values are different?

zakjma
  • 2,030
  • 12
  • 40
  • 81
  • identity is no good camera intrinsics matrix. does your image have distortion? in the end, probably your assumptions dont hold: known distortion coeffs and known camera intrinsics. if your scene doesnt work for those values you cant expect the projection to make sense. – Micka Dec 04 '15 at 17:36
  • I take this code from https://github.com/daviddoria/Examples/blob/master/c%2B%2B/OpenCV/SolvePNP/SolvePNP.cxx. I have no image. – zakjma Dec 05 '15 at 06:37

0 Answers0