Abstract class for registration algorithms. More...

#include <matchBase.h>

Inheritance diagram for gr::MatchBase< PointType, _TransformVisitor, OptExts >:

Collaboration diagram for gr::MatchBase< PointType, _TransformVisitor, OptExts >:

Classes
class	Options

struct	PosMutablePoint
	A convenience class used to wrap (any) PointType to allow mutation of position of point samples for internal computations. More...

Public Types
using	LogLevel = Utils::LogLevel

using	MatrixType = Eigen::Matrix< Scalar, 4, 4 >

using	OptionsType = gr::Utils::CRTP< OptExts..., Options >

using	Scalar = typename PointType::Scalar

using	TransformVisitor = _TransformVisitor

using	VectorType = typename PointType::VectorType

Public Member Functions
template<typename InputRange1 , typename InputRange2 , template< typename > typename Sampler>
Scalar	ComputeTransformation (const InputRange1 &P, const InputRange2 &Q, Eigen::Ref< MatrixType > transformation, const Sampler< PointType > &sampler, TransformVisitor &v)
	Computes an approximation of the best LCP (directional) from Q to P and the rigid transformation that realizes it. The input sets may or may not contain normal information for any point. More...

const std::vector< PosMutablePoint > &	getFirstSampled () const
	Read access to the sampled clouds used for the registration. More...

const std::vector< PosMutablePoint > &	getSecondSampled () const
	Read access to the sampled clouds used for the registration. More...

EIGEN_MAKE_ALIGNED_OPERATOR_NEW	MatchBase (const OptionsType &options, const Utils::Logger &logger)

virtual	~MatchBase ()

Protected Member Functions
template<typename Coordinates >
bool	ComputeRigidTransformation (const Coordinates &ref, const Coordinates &candidate, const Eigen::Matrix< Scalar, 3, 1 > &centroid1, Eigen::Matrix< Scalar, 3, 1 > centroid2, Eigen::Ref< MatrixType > transform, Scalar &rms_, bool computeScale) const
	Computes the best rigid transformation between three corresponding pairs. The transformation is characterized by rotation matrix, translation vector and a center about which we rotate. The set of pairs is potentially being updated by the best permutation of the second set. Returns the RMS of the fit. The method is being called with n points but it applies the fit for only 3 after the best permutation is selected in the second set (see bellow). This is done because the solution for planar points is much simpler. The method is the closed-form solution by Horn: people.csail.mit.edu/bkph/papers/Absolute_Orientation.pdf. More...

template<typename InputRange1 , typename InputRange2 , template< typename > class Sampler>
void	init (const InputRange1 &P, const InputRange2 &Q, const Sampler< PointType > &sampler)
	Initializes the internal state of the Base class. More...

virtual void	Initialize ()
	Initializes the data structures and needed values before the match computation. This method is called once the internal state of the Base class as been set. More...

template<Utils::LogLevel level, typename... Args>
void	Log (Args...args) const

Scalar	MeanDistance () const
	Computes the mean distance between points in Q and their nearest neighbor. We need this for normalization of the user delta (See the paper) to the "scale" of the set. More...

bool	SelectRandomTriangle (Scalar max_base_diameter, int &base1, int &base2, int &base3)
	Selects a random triangle in the set P (then we add another point to keep the base as planar as possible). We apply a simple heuristic that works in most practical cases. The idea is to accept maximum distance, computed by the estimated overlap, multiplied by the diameter of P, and try to have a triangle with all three edges close to this distance. Wide triangles helps to make the transformation robust while too large triangles makes the probability of having all points in the inliers small so we try to trade-off. More...

Protected Attributes
VectorType	centroid_P_ {VectorType::Zero()}
	The centroid of P. More...

VectorType	centroid_Q_ {VectorType::Zero()}
	The centroid of Q. More...

KdTree< Scalar >	kd_tree_
	KdTree used to compute the LCP. More...

const Utils::Logger &	logger_

OptionsType	options_

Scalar	P_diameter_ {Scalar( 0 )}
	The diameter of P. More...

Scalar	P_mean_distance_ {Scalar( 0 )}
	Mean distance between points and their nearest neighbor in the set P. Used to normalize the "delta" which is given in terms of this distance. More...

VectorType	qcentroid1_ {VectorType::Zero()}

VectorType	qcentroid2_ {VectorType::Zero()}

std::mt19937	randomGenerator_

std::vector< PosMutablePoint >	sampled_P_3D_
	Sampled P (3D coordinates). More...

std::vector< PosMutablePoint >	sampled_Q_3D_
	Sampled Q (3D coordinates). More...

Eigen::Matrix< Scalar, 4, 4 >	transform_ {Eigen::Matrix<Scalar, 4, 4>::Identity()}
	The transformation matrix by wich we transform Q to P. More...

Static Protected Attributes
static constexpr int	kNumberOfDiameterTrials = 1000

Detailed Description

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>
class gr::MatchBase< PointType, _TransformVisitor, OptExts >

Abstract class for registration algorithms.

Member Typedef Documentation

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

using gr::MatchBase< PointType, _TransformVisitor, OptExts >::LogLevel = Utils::LogLevel

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

using gr::MatchBase< PointType, _TransformVisitor, OptExts >::MatrixType = Eigen::Matrix<Scalar, 4, 4>

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

using gr::MatchBase< PointType, _TransformVisitor, OptExts >::OptionsType = gr::Utils::CRTP < OptExts ... , Options >

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

using gr::MatchBase< PointType, _TransformVisitor, OptExts >::Scalar = typename PointType::Scalar

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

using gr::MatchBase< PointType, _TransformVisitor, OptExts >::TransformVisitor = _TransformVisitor

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

using gr::MatchBase< PointType, _TransformVisitor, OptExts >::VectorType = typename PointType::VectorType

Constructor & Destructor Documentation

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

EIGEN_MAKE_ALIGNED_OPERATOR_NEW gr::MatchBase< PointType, _TransformVisitor, OptExts >::MatchBase	(	const OptionsType &	options,
		const Utils::Logger &	logger
	)

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

virtual gr::MatchBase< PointType, _TransformVisitor, OptExts >::~MatchBase ( )

virtual

Member Function Documentation

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

template<typename Coordinates >

bool gr::MatchBase< PointType, _TransformVisitor, OptExts >::ComputeRigidTransformation	(	const Coordinates &	ref,
		const Coordinates &	candidate,
		const Eigen::Matrix< Scalar, 3, 1 > &	centroid1,
		Eigen::Matrix< Scalar, 3, 1 >	centroid2,
		Eigen::Ref< MatrixType >	transform,
		Scalar &	rms_,
		bool	computeScale
	)		const

protected

Computes the best rigid transformation between three corresponding pairs. The transformation is characterized by rotation matrix, translation vector and a center about which we rotate. The set of pairs is potentially being updated by the best permutation of the second set. Returns the RMS of the fit. The method is being called with n points but it applies the fit for only 3 after the best permutation is selected in the second set (see bellow). This is done because the solution for planar points is much simpler. The method is the closed-form solution by Horn: people.csail.mit.edu/bkph/papers/Absolute_Orientation.pdf.

Template Parameters

Coordinates Struct with operator[](int i) ->i[0:2]

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

template<typename InputRange1 , typename InputRange2 , template< typename > typename Sampler>

Scalar gr::MatchBase< PointType, _TransformVisitor, OptExts >::ComputeTransformation	(	const InputRange1 &	P,
		const InputRange2 &	Q,
		Eigen::Ref< MatrixType >	transformation,
		const Sampler< PointType > &	sampler,
		TransformVisitor &	v
	)

inline

Computes an approximation of the best LCP (directional) from Q to P and the rigid transformation that realizes it. The input sets may or may not contain normal information for any point.

Parameters

[in]	P	The first input set.
[in]	Q	The second input set.
[out]	transformation	Rigid transformation matrix (4x4) that brings Q to the (approximate) optimal LCP. Initial value is considered as a guess

Returns: the computed LCP measure as a fraction of the size of P ([0..1]). Computes an approximation of the best LCP (directional) from Q to P and the rigid transformation that realizes it. The input sets may or may not contain normal information for any point.

Parameters

[in]	P	The first input set.
[in]	Q	The second input set.
[out]	transformation	Rigid transformation matrix (4x4) that brings Q to the (approximate) optimal LCP. Initial value is considered as a guess

Returns: the computed LCP measure as a fraction of the size of P ([0..1]).

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

const std::vector<PosMutablePoint>& gr::MatchBase< PointType, _TransformVisitor, OptExts >::getFirstSampled ( ) const

inline

Read access to the sampled clouds used for the registration.

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

const std::vector<PosMutablePoint>& gr::MatchBase< PointType, _TransformVisitor, OptExts >::getSecondSampled ( ) const

inline

Read access to the sampled clouds used for the registration.

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

template<typename InputRange1 , typename InputRange2 , template< typename > class Sampler>

void gr::MatchBase< PointType, _TransformVisitor, OptExts >::init	(	const InputRange1 &	P,
		const InputRange2 &	Q,
		const Sampler< PointType > &	sampler
	)

protected

Initializes the internal state of the Base class.

Parameters

P	The first input set.
Q	The second input set.
sampler	The sampler used to sample the input sets.

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

virtual void gr::MatchBase< PointType, _TransformVisitor, OptExts >::Initialize ( )

inlineprotectedvirtual

Initializes the data structures and needed values before the match computation. This method is called once the internal state of the Base class as been set.

Reimplemented in gr::Match4pcsBase< _Functor, _PointType, _TransformVisitor, _PairFilteringFunctor, PairFilteringOptions >.

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

template<Utils::LogLevel level, typename... Args>

void gr::MatchBase< PointType, _TransformVisitor, OptExts >::Log ( Args... args ) const

inlineprotected

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

Scalar gr::MatchBase< PointType, _TransformVisitor, OptExts >::MeanDistance ( ) const

protected

Computes the mean distance between points in Q and their nearest neighbor. We need this for normalization of the user delta (See the paper) to the "scale" of the set.

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

bool gr::MatchBase< PointType, _TransformVisitor, OptExts >::SelectRandomTriangle	(	Scalar	max_base_diameter,
		int &	base1,
		int &	base2,
		int &	base3
	)

protected

Selects a random triangle in the set P (then we add another point to keep the base as planar as possible). We apply a simple heuristic that works in most practical cases. The idea is to accept maximum distance, computed by the estimated overlap, multiplied by the diameter of P, and try to have a triangle with all three edges close to this distance. Wide triangles helps to make the transformation robust while too large triangles makes the probability of having all points in the inliers small so we try to trade-off.

Parameters

max_base_diameter Maximum size allowed between two points of the base

Member Data Documentation

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

VectorType gr::MatchBase< PointType, _TransformVisitor, OptExts >::centroid_P_ {VectorType::Zero()}

protected

The centroid of P.

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

VectorType gr::MatchBase< PointType, _TransformVisitor, OptExts >::centroid_Q_ {VectorType::Zero()}

protected

The centroid of Q.

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

KdTree<Scalar> gr::MatchBase< PointType, _TransformVisitor, OptExts >::kd_tree_

protected

KdTree used to compute the LCP.

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

constexpr int gr::MatchBase< PointType, _TransformVisitor, OptExts >::kNumberOfDiameterTrials = 1000

staticprotected

Todo:: Rationnalize use and name of this variable

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

const Utils::Logger& gr::MatchBase< PointType, _TransformVisitor, OptExts >::logger_

protected

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

OptionsType gr::MatchBase< PointType, _TransformVisitor, OptExts >::options_

protected

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

Scalar gr::MatchBase< PointType, _TransformVisitor, OptExts >::P_diameter_ {Scalar( 0 )}

protected

The diameter of P.

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

Scalar gr::MatchBase< PointType, _TransformVisitor, OptExts >::P_mean_distance_ {Scalar( 0 )}

protected

Mean distance between points and their nearest neighbor in the set P. Used to normalize the "delta" which is given in terms of this distance.

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

VectorType gr::MatchBase< PointType, _TransformVisitor, OptExts >::qcentroid1_ {VectorType::Zero()}

protected

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

VectorType gr::MatchBase< PointType, _TransformVisitor, OptExts >::qcentroid2_ {VectorType::Zero()}

protected

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

std::mt19937 gr::MatchBase< PointType, _TransformVisitor, OptExts >::randomGenerator_

protected

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

std::vector<PosMutablePoint> gr::MatchBase< PointType, _TransformVisitor, OptExts >::sampled_P_3D_

protected

Sampled P (3D coordinates).

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

std::vector<PosMutablePoint> gr::MatchBase< PointType, _TransformVisitor, OptExts >::sampled_Q_3D_

protected

Sampled Q (3D coordinates).

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>

Eigen::Matrix<Scalar, 4, 4> gr::MatchBase< PointType, _TransformVisitor, OptExts >::transform_ {Eigen::Matrix<Scalar, 4, 4>::Identity()}

protected

The transformation matrix by wich we transform Q to P.

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

/home/travis/build/STORM-IRIT/OpenGR/src/gr/algorithms/matchBase.h

Classes

Public Types

Public Member Functions

Protected Member Functions

Protected Attributes

Static Protected Attributes

Detailed Description

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts> class gr::MatchBase< PointType, _TransformVisitor, OptExts >

Member Typedef Documentation

Constructor & Destructor Documentation

Member Function Documentation

Member Data Documentation

template<typename PointType, typename _TransformVisitor = DummyTransformVisitor, template< class, class > class... OptExts>
class gr::MatchBase< PointType, _TransformVisitor, OptExts >