kcp::KCP Class Reference

The KCP-TEASER registration approach. More...

#include <kcp/solver.hpp>

Inheritance diagram for kcp::KCP:

Collaboration diagram for kcp::KCP:

Classes
struct	Params
	Type of parameters for the KCP-TEASER solver. More...

Public Types
using	TEASER = teaser::RobustRegistrationSolver
	Type alias of the TEASER++ solver.

Public Member Functions
	KCP (KCP::Params params)
	Construct a new KCP object. More...
KCP::Params &	get_params ()
	Get the parameters. More...
const Correspondences &	get_initial_correspondences () const
	Get the initial set of correspondences. More...
const std::vector< int > &	get_inlier_correspondence_indices () const
	Get the inlier correspondence indices. More...
virtual void	solve (const Eigen::MatrixX3d &src, const Eigen::MatrixX3d &dst, const Eigen::MatrixXd &src_feature, const Eigen::MatrixXd &dst_feature) override
	The main function to trigger the KCP-TEASER registration approach. More...
Public Member Functions inherited from kcp::AbstractSolver
	AbstractSolver ()
	Constructor.
virtual const Eigen::Matrix4d &	get_solution () const
	Get the solution of the registration. More...

Protected Attributes
TEASER	solver
	The TEASER++ solver.
KCP::Params	params
	Parameters for the KCP-TEASER solver.
Correspondences	initial_correspondences
	The initial set of k-closest-points correspondences.
std::vector< int >	inlier_correspondence_indices
	The inlier correspondence indices with respect to initial_correspondences. The set is estimated by the maximum clique pruning method.
Protected Attributes inherited from kcp::AbstractSolver
Eigen::Matrix4d	solution
	The estimation result.

Detailed Description

The KCP-TEASER registration approach.

KCP is an efficient and effective local point cloud registration approach targeting for real-world 3D LiDAR scan matching problem. A simple (and naive) understanding is: ICP iteratively considers the closest point of each source point, but KCP considers the k closest points of each source point in the beginning, and outlier correspondences are mainly rejected by the maximum clique pruning method (provided by TEASER++).

See also

Yu-Kai Lin, Wen-Chieh Lin, Chieh-Chih Wang, KCP: k-Closest Points and Maximum Clique Pruning for Efficient and Effective 3D Laser Scan Matching. To appear in IEEE Robotics and Automation Letters (RA-L), 2022.

Constructor & Destructor Documentation

KCP()

kcp::KCP::KCP ( KCP::Params  params )

inline

Construct a new KCP object.

Parameters

params

KCP-TEASER parameters.

Member Function Documentation

get_initial_correspondences()

const Correspondences& kcp::KCP::get_initial_correspondences ( ) const

inline

Get the initial set of correspondences.

Returns

const Correspondences& Initial set of correspondences.

get_inlier_correspondence_indices()

const std::vector<int>& kcp::KCP::get_inlier_correspondence_indices ( ) const

inline

Get the inlier correspondence indices.

Returns

const std::vector<int>& Inlier correspondence indices.

get_params()

KCP::Params& kcp::KCP::get_params ( )

inline

Get the parameters.

Returns

Parameters.

solve()

void kcp::KCP::solve ( const Eigen::MatrixX3d &  src, const Eigen::MatrixX3d &  dst, const Eigen::MatrixXd &  src_feature, const Eigen::MatrixXd &  dst_feature  )

overridevirtual

The main function to trigger the KCP-TEASER registration approach.

Parameters

src	The source point cloud.
dst	The target point cloud.
src_feature	The source feature cloud.
dst_feature	The target feature cloud.

Implements kcp::AbstractSolver.

---

The documentation for this class was generated from the following files:

• kcp/include/kcp/solver.hpp
• kcp/src/solver.cpp