Abstract base class for Correspondence Grouping algorithms. More...

#include <pcl/recognition/cg/correspondence_grouping.h>

Inheritance diagram for pcl::CorrespondenceGrouping< PointModelT, PointSceneT >:

Collaboration diagram for pcl::CorrespondenceGrouping< PointModelT, PointSceneT >:

Public Types
using	SceneCloud = pcl::PointCloud<PointSceneT>
using	SceneCloudPtr = typename SceneCloud::Ptr
using	SceneCloudConstPtr = typename SceneCloud::ConstPtr
Public Types inherited from pcl::PCLBase< PointModelT >
using	PointCloud
using	PointCloudPtr
using	PointCloudConstPtr
using	PointIndicesPtr
using	PointIndicesConstPtr

Public Member Functions
	CorrespondenceGrouping ()
	Empty constructor.
	~CorrespondenceGrouping () override
	destructor.
virtual void	setSceneCloud (const SceneCloudConstPtr &scene)
	Provide a pointer to the scene dataset.
SceneCloudConstPtr	getSceneCloud () const
	Getter for the scene dataset.
virtual void	setModelSceneCorrespondences (const CorrespondencesConstPtr &corrs)
	Provide a pointer to the precomputed correspondences between points in the input dataset and points in the scene dataset.
CorrespondencesConstPtr	getModelSceneCorrespondences () const
	Getter for the precomputed correspondences between points in the input dataset and points in the scene dataset.
std::vector< double >	getCharacteristicScales () const
	Getter for the vector of characteristic scales associated to each cluster.
void	cluster (std::vector< Correspondences > &clustered_corrs)
	Clusters the input correspondences belonging to different model instances.
Public Member Functions inherited from pcl::PCLBase< PointModelT >
	PCLBase ()
	Empty constructor.
virtual	~PCLBase ()=default
	Destructor.
virtual void	setInputCloud (const PointCloudConstPtr &cloud)
	Provide a pointer to the input dataset.
PointCloudConstPtr const	getInputCloud () const
	Get a pointer to the input point cloud dataset.
virtual void	setIndices (const IndicesPtr &indices)
	Provide a pointer to the vector of indices that represents the input data.
IndicesPtr	getIndices ()
	Get a pointer to the vector of indices used.
const PointModelT &	operator[] (std::size_t pos) const
	Override PointCloud operator[] to shorten code.

Protected Member Functions
virtual void	clusterCorrespondences (std::vector< Correspondences > &clustered_corrs)=0
	The actual clustering method, should be implemented by each subclass.
bool	initCompute ()
	This method should get called before starting the actual computation.
bool	deinitCompute ()
	This method should get called after finishing the actual computation.
Protected Member Functions inherited from pcl::PCLBase< PointModelT >
bool	initCompute ()
	This method should get called before starting the actual computation.
bool	deinitCompute ()
	This method should get called after finishing the actual computation.

Protected Attributes
SceneCloudConstPtr	scene_
	The scene cloud.
CorrespondencesConstPtr	model_scene_corrs_
	The correspondences between points in the input and the scene datasets.
std::vector< double >	corr_group_scale_
	characteristic scale associated to each correspondence subset; if the cg algorithm can not handle scale invariance, the size of the vector will be 0.
Protected Attributes inherited from pcl::PCLBase< PointModelT >
PointCloudConstPtr	input_
	The input point cloud dataset.
IndicesPtr	indices_
	A pointer to the vector of point indices to use.
bool	use_indices_
	Set to true if point indices are used.
bool	fake_indices_
	If no set of indices are given, we construct a set of fake indices that mimic the input PointCloud.

Detailed Description

template<typename PointModelT, typename PointSceneT>
class pcl::CorrespondenceGrouping< PointModelT, PointSceneT >

Abstract base class for Correspondence Grouping algorithms.

Author: Tommaso Cavallari, Federico Tombari, Aitor Aldoma

Definition at line 54 of file correspondence_grouping.h.

Member Typedef Documentation

◆ SceneCloud

template<typename PointModelT, typename PointSceneT>

using pcl::CorrespondenceGrouping< PointModelT, PointSceneT >::SceneCloud = pcl::PointCloud<PointSceneT>

Definition at line 57 of file correspondence_grouping.h.

◆ SceneCloudConstPtr

template<typename PointModelT, typename PointSceneT>

using pcl::CorrespondenceGrouping< PointModelT, PointSceneT >::SceneCloudConstPtr = typename SceneCloud::ConstPtr

Definition at line 59 of file correspondence_grouping.h.

◆ SceneCloudPtr

template<typename PointModelT, typename PointSceneT>

using pcl::CorrespondenceGrouping< PointModelT, PointSceneT >::SceneCloudPtr = typename SceneCloud::Ptr

Definition at line 58 of file correspondence_grouping.h.

Constructor & Destructor Documentation

◆ CorrespondenceGrouping()

template<typename PointModelT, typename PointSceneT>

pcl::CorrespondenceGrouping< PointModelT, PointSceneT >::CorrespondenceGrouping ( )

inline

Empty constructor.

Definition at line 62 of file correspondence_grouping.h.

References scene_.

◆ ~CorrespondenceGrouping()

template<typename PointModelT, typename PointSceneT>

pcl::CorrespondenceGrouping< PointModelT, PointSceneT >::~CorrespondenceGrouping ( )

inlineoverride

destructor.

Definition at line 65 of file correspondence_grouping.h.

References model_scene_corrs_, and scene_.

Member Function Documentation

◆ cluster()

template<typename PointModelT, typename PointSceneT>

void pcl::CorrespondenceGrouping< PointModelT, PointSceneT >::cluster ( std::vector< Correspondences > & clustered_corrs )

Clusters the input correspondences belonging to different model instances.

Parameters

[out] clustered_corrs a vector containing the correspondences for each instance of the model found within the input data.

Definition at line 49 of file correspondence_grouping.hpp.

References clusterCorrespondences(), corr_group_scale_, deinitCompute(), and initCompute().

◆ clusterCorrespondences()

template<typename PointModelT, typename PointSceneT>

virtual void pcl::CorrespondenceGrouping< PointModelT, PointSceneT >::clusterCorrespondences ( std::vector< Correspondences > & clustered_corrs )

protectedpure virtual

The actual clustering method, should be implemented by each subclass.

Parameters

[out] clustered_corrs a vector containing the correspondences for each instance of the model found within the input data.

Implemented in pcl::GeometricConsistencyGrouping< PointModelT, PointSceneT >, and pcl::Hough3DGrouping< PointModelT, PointSceneT, PointModelRfT, PointSceneRfT >.

Referenced by cluster().

◆ deinitCompute()

template<typename PointModelT, typename PointSceneT>

bool pcl::CorrespondenceGrouping< PointModelT, PointSceneT >::deinitCompute ( )

inlineprotected

This method should get called after finishing the actual computation.

Definition at line 191 of file correspondence_grouping.h.

Referenced by cluster(), pcl::GeometricConsistencyGrouping< PointModelT, PointSceneT >::recognize(), and pcl::Hough3DGrouping< PointModelT, PointSceneT, PointModelRfT, PointSceneRfT >::recognize().

◆ getCharacteristicScales()

template<typename PointModelT, typename PointSceneT>

std::vector< double > pcl::CorrespondenceGrouping< PointModelT, PointSceneT >::getCharacteristicScales ( ) const

inline

Getter for the vector of characteristic scales associated to each cluster.

Returns: the vector of characteristic scales (assuming scale = model / scene)

Definition at line 119 of file correspondence_grouping.h.

References corr_group_scale_.

◆ getModelSceneCorrespondences()

template<typename PointModelT, typename PointSceneT>

CorrespondencesConstPtr pcl::CorrespondenceGrouping< PointModelT, PointSceneT >::getModelSceneCorrespondences ( ) const

inline

Getter for the precomputed correspondences between points in the input dataset and points in the scene dataset.

Returns: the correspondences between the model and the scene.

Definition at line 109 of file correspondence_grouping.h.

References model_scene_corrs_.

◆ getSceneCloud()

template<typename PointModelT, typename PointSceneT>

SceneCloudConstPtr pcl::CorrespondenceGrouping< PointModelT, PointSceneT >::getSceneCloud ( ) const

inline

Getter for the scene dataset.

Returns: the const boost shared pointer to a PointCloud message.

Definition at line 86 of file correspondence_grouping.h.

References scene_.

◆ initCompute()

template<typename PointModelT, typename PointSceneT>

bool pcl::CorrespondenceGrouping< PointModelT, PointSceneT >::initCompute ( )

inlineprotected

This method should get called before starting the actual computation.

Internally, initCompute() does the following:

checks if an input dataset is given, and returns false otherwise
checks if a scene dataset is given, and returns false otherwise
checks if the model-scene correspondences have been given, and returns false otherwise

Definition at line 159 of file correspondence_grouping.h.

References pcl::PCLBase< PointT >::initCompute(), pcl::PCLBase< PointModelT >::input_, model_scene_corrs_, and scene_.

Referenced by cluster(), pcl::GeometricConsistencyGrouping< PointModelT, PointSceneT >::recognize(), and pcl::Hough3DGrouping< PointModelT, PointSceneT, PointModelRfT, PointSceneRfT >::recognize().

◆ setModelSceneCorrespondences()

template<typename PointModelT, typename PointSceneT>

virtual void pcl::CorrespondenceGrouping< PointModelT, PointSceneT >::setModelSceneCorrespondences ( const CorrespondencesConstPtr & corrs )

inlinevirtual

Provide a pointer to the precomputed correspondences between points in the input dataset and points in the scene dataset.

The correspondences are going to be clustered into different model hypotheses by the algorithm.

Parameters

[in] corrs the correspondences between the model and the scene.

Reimplemented in pcl::Hough3DGrouping< PointModelT, PointSceneT, PointModelRfT, PointSceneRfT >.

Definition at line 98 of file correspondence_grouping.h.

References model_scene_corrs_.

◆ setSceneCloud()

template<typename PointModelT, typename PointSceneT>

virtual void pcl::CorrespondenceGrouping< PointModelT, PointSceneT >::setSceneCloud ( const SceneCloudConstPtr & scene )

inlinevirtual

Provide a pointer to the scene dataset.

Parameters

[in] scene the const boost shared pointer to a PointCloud message.

Reimplemented in pcl::Hough3DGrouping< PointModelT, PointSceneT, PointModelRfT, PointSceneRfT >.

Definition at line 76 of file correspondence_grouping.h.

References scene_.

Member Data Documentation

◆ corr_group_scale_

template<typename PointModelT, typename PointSceneT>

std::vector<double> pcl::CorrespondenceGrouping< PointModelT, PointSceneT >::corr_group_scale_

protected

characteristic scale associated to each correspondence subset; if the cg algorithm can not handle scale invariance, the size of the vector will be 0.

Definition at line 142 of file correspondence_grouping.h.

Referenced by cluster(), and getCharacteristicScales().

◆ model_scene_corrs_

template<typename PointModelT, typename PointSceneT>

CorrespondencesConstPtr pcl::CorrespondenceGrouping< PointModelT, PointSceneT >::model_scene_corrs_

protected

The correspondences between points in the input and the scene datasets.

Definition at line 138 of file correspondence_grouping.h.

Referenced by pcl::GeometricConsistencyGrouping< PointModelT, PointSceneT >::clusterCorrespondences(), pcl::Hough3DGrouping< PointModelT, PointSceneT, PointModelRfT, PointSceneRfT >::clusterCorrespondences(), getModelSceneCorrespondences(), pcl::Hough3DGrouping< PointModelT, PointSceneT, PointModelRfT, PointSceneRfT >::houghVoting(), initCompute(), setModelSceneCorrespondences(), pcl::Hough3DGrouping< PointModelT, PointSceneT, PointModelRfT, PointSceneRfT >::setModelSceneCorrespondences(), and ~CorrespondenceGrouping().

◆ scene_

template<typename PointModelT, typename PointSceneT>

SceneCloudConstPtr pcl::CorrespondenceGrouping< PointModelT, PointSceneT >::scene_

protected

The scene cloud.

Definition at line 133 of file correspondence_grouping.h.

Referenced by pcl::GeometricConsistencyGrouping< PointModelT, PointSceneT >::clusterCorrespondences(), pcl::Hough3DGrouping< PointModelT, PointSceneT, PointModelRfT, PointSceneRfT >::clusterCorrespondences(), CorrespondenceGrouping(), getSceneCloud(), pcl::Hough3DGrouping< PointModelT, PointSceneT, PointModelRfT, PointSceneRfT >::houghVoting(), initCompute(), setSceneCloud(), pcl::Hough3DGrouping< PointModelT, PointSceneT, PointModelRfT, PointSceneRfT >::setSceneCloud(), and ~CorrespondenceGrouping().

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

pcl/recognition/cg/correspondence_grouping.h
pcl/recognition/impl/cg/correspondence_grouping.hpp

Public Types

Public Member Functions

Protected Member Functions

Protected Attributes

Detailed Description

Member Typedef Documentation

◆ SceneCloud

◆ SceneCloudConstPtr

◆ SceneCloudPtr

Constructor & Destructor Documentation

◆ CorrespondenceGrouping()

◆ ~CorrespondenceGrouping()

Member Function Documentation

◆ cluster()

◆ clusterCorrespondences()

◆ deinitCompute()

◆ getCharacteristicScales()

◆ getModelSceneCorrespondences()

◆ getSceneCloud()

◆ initCompute()

◆ setModelSceneCorrespondences()

◆ setSceneCloud()

Member Data Documentation

◆ corr_group_scale_

◆ model_scene_corrs_

◆ scene_