In order for POSIT to work, you need at least four non-coplanar 3D model points and their respective matchings in the 2D image. We will add a termination criteria to that, since POSIT is an iterative algorithm, which generally is a number of iterations or a distance parameter. We will then call the cvPOSIT function, included in calib3d_c.h, which yields the rotation matrix and the translation vector.
As an example, we will follow the tutorial from Javier Barandiaran, which uses POSIT to obtain the pose of a cube. The model is created with four points. It is initialized with the following code:
float cubeSize = 10.0; std::vector<CvPoint3D32f> modelPoints; modelPoints.push_back(cvPoint3D32f(0.0f, 0.0f, 0.0f)); modelPoints.push_back(cvPoint3D32f(0.0f, ...
