Point Cloud Library (PCL)
1.11.1
|
44 #include <pcl/registration/icp.h>
45 #include <pcl/registration/transformation_estimation_lm.h>
66 template <
typename Po
intSource,
typename Po
intTarget,
typename Scalar =
float>
76 using Ptr = shared_ptr< IterativeClosestPointNonLinear<PointSource, PointTarget, Scalar> >;
77 using ConstPtr = shared_ptr< const IterativeClosestPointNonLinear<PointSource, PointTarget, Scalar> >;
85 reg_name_ =
"IterativeClosestPointNonLinear";
IterativeClosestPointNonLinear is an ICP variant that uses Levenberg-Marquardt optimization backend.
typename Registration< PointSource, PointTarget, Scalar >::Matrix4 Matrix4
IterativeClosestPoint provides a base implementation of the Iterative Closest Point algorithm.
int min_number_correspondences_
The minimum number of correspondences that the algorithm needs before attempting to estimate the tran...
Eigen::Matrix< Scalar, 4, 4 > Matrix4
shared_ptr< IterativeClosestPointNonLinear< PointSource, PointTarget, Scalar > > Ptr
IterativeClosestPointNonLinear()
Empty constructor.
TransformationEstimationPtr transformation_estimation_
A TransformationEstimation object, used to calculate the 4x4 rigid transformation.
shared_ptr< const IterativeClosestPointNonLinear< PointSource, PointTarget, Scalar > > ConstPtr
std::string reg_name_
The registration method name.