Generic class for extracting the persistent features from an input point cloud It can be given any Feature estimator instance and will compute the features of the input over a multiscale representation of the cloud and output the unique ones over those scales. More...

#include <multiscale_feature_persistence.h>

Inheritance diagram for pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >:

[legend]

List of all members.

Public Types
typedef boost::shared_ptr < const MultiscaleFeaturePersistence < PointSource, PointFeature > >	ConstPtr
typedef pcl::PointCloud < PointFeature >	FeatureCloud
typedef pcl::PointCloud < PointFeature >::Ptr	FeatureCloudPtr
typedef pcl::Feature < PointSource, PointFeature > ::Ptr	FeatureEstimatorPtr
typedef boost::shared_ptr < const pcl::PointRepresentation < PointFeature > >	FeatureRepresentationConstPtr
typedef boost::shared_ptr < MultiscaleFeaturePersistence < PointSource, PointFeature > >	Ptr
Public Member Functions
void	computeFeaturesAtAllScales ()
	Method that calls computeFeatureAtScale () for each scale parameter.
void	determinePersistentFeatures (FeatureCloud &output_features, boost::shared_ptr< std::vector< int > > &output_indices)
	Central function that computes the persistent features.
float	getAlpha ()
	Get the value of the alpha parameter.
NormType	getDistanceMetric ()
	Returns the distance metric that is currently used to calculate the difference between feature vectors.
FeatureEstimatorPtr	getFeatureEstimator ()
	Getter method for the feature estimator.
FeatureRepresentationConstPtr const	getPointRepresentation ()
	Get a pointer to the feature representation used when converting features into k-D vectors.
std::vector< float >	getScalesVector ()
	Method for getting the scale parameters vector.
	MultiscaleFeaturePersistence ()
	Empty constructor.
void	setAlpha (float alpha)
	Sets the alpha parameter.
void	setDistanceMetric (NormType distance_metric)
	Method for setting the distance metric that will be used for computing the difference between feature vectors.
void	setFeatureEstimator (FeatureEstimatorPtr feature_estimator)
	Setter method for the feature estimator.
void	setPointRepresentation (const FeatureRepresentationConstPtr &feature_representation)
	Provide a pointer to the feature representation to use to convert features to k-D vectors.
void	setScalesVector (std::vector< float > &scale_values)
	Method for setting the scale parameters for the algorithm.
virtual	~MultiscaleFeaturePersistence ()
	Empty destructor.
Private Member Functions
void	calculateMeanFeature ()
	Method that averages all the features at all scales in order to obtain the global mean feature; this value is stored in the mean_feature field.
virtual void	computeFeatureAtScale (float &scale, FeatureCloudPtr &features)
	Method to compute the features for the point cloud at the given scale.
float	distanceBetweenFeatures (const std::vector< float > &a, const std::vector< float > &b)
	Function that calculates the scalar difference between two features.
void	extractUniqueFeatures ()
	Selects the so-called 'unique' features from the cloud of features at each level. These features are the ones that fall outside the standard deviation * alpha_.
bool	initCompute ()
	Checks if all the necessary input was given and the computations can successfully start.
Private Attributes
float	alpha_
	Parameter that determines if a feature is to be considered unique or not.
NormType	distance_metric_
	Parameter that determines which distance metric is to be usedto calculate the difference between feature vectors.
FeatureEstimatorPtr	feature_estimator_
	the feature estimator that will be used to determine the feature set at each scale level
FeatureRepresentationConstPtr	feature_representation_
std::vector< FeatureCloudPtr >	features_at_scale_
std::vector< std::vector < std::vector< float > > >	features_at_scale_vectorized_
std::vector< float >	mean_feature_
std::vector< float >	scale_values_
	The general parameter for determining each scale level.
std::vector< std::list< size_t > >	unique_features_indices_
	Two structures in which to hold the results of the unique feature extraction process. They are superfluous with respect to each other, but improve the time performance of the algorithm.
std::vector< std::vector< bool > >	unique_features_table_

Detailed Description

template<typename PointSource, typename PointFeature>
class pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >

Generic class for extracting the persistent features from an input point cloud It can be given any Feature estimator instance and will compute the features of the input over a multiscale representation of the cloud and output the unique ones over those scales.

Please refer to the following publication for more details: Radu Bogdan Rusu, Zoltan Csaba Marton, Nico Blodow, and Michael Beetz Persistent Point Feature Histograms for 3D Point Clouds Proceedings of the 10th International Conference on Intelligent Autonomous Systems (IAS-10) 2008, Baden-Baden, Germany

Author:: Alexandru-Eugen Ichim

Definition at line 64 of file multiscale_feature_persistence.h.

Member Typedef Documentation

template<typename PointSource, typename PointFeature>

typedef boost::shared_ptr<const MultiscaleFeaturePersistence<PointSource, PointFeature> > pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::ConstPtr

Definition at line 68 of file multiscale_feature_persistence.h.

template<typename PointSource, typename PointFeature>

typedef pcl::PointCloud<PointFeature> pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::FeatureCloud

Definition at line 69 of file multiscale_feature_persistence.h.

template<typename PointSource, typename PointFeature>

typedef pcl::PointCloud<PointFeature>::Ptr pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::FeatureCloudPtr

Definition at line 70 of file multiscale_feature_persistence.h.

template<typename PointSource, typename PointFeature>

typedef pcl::Feature<PointSource, PointFeature>::Ptr pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::FeatureEstimatorPtr

Definition at line 71 of file multiscale_feature_persistence.h.

template<typename PointSource, typename PointFeature>

typedef boost::shared_ptr<const pcl::PointRepresentation <PointFeature> > pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::FeatureRepresentationConstPtr

Definition at line 72 of file multiscale_feature_persistence.h.

template<typename PointSource, typename PointFeature>

typedef boost::shared_ptr<MultiscaleFeaturePersistence<PointSource, PointFeature> > pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::Ptr

Definition at line 67 of file multiscale_feature_persistence.h.

Constructor & Destructor Documentation

template<typename PointSource , typename PointFeature >

pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::MultiscaleFeaturePersistence ( )

Empty constructor.

Definition at line 47 of file multiscale_feature_persistence.hpp.

template<typename PointSource, typename PointFeature>

virtual pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::~MultiscaleFeaturePersistence ( ) [inline, virtual]

Empty destructor.

Definition at line 80 of file multiscale_feature_persistence.h.

Member Function Documentation

template<typename PointSource , typename PointFeature >

void pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::calculateMeanFeature ( ) [private]

Method that averages all the features at all scales in order to obtain the global mean feature; this value is stored in the mean_feature field.

Definition at line 137 of file multiscale_feature_persistence.hpp.

template<typename PointSource , typename PointFeature >

void pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::computeFeatureAtScale	(	float &	scale,
		FeatureCloudPtr &	features
	)		`[private, virtual]`

Method to compute the features for the point cloud at the given scale.

Definition at line 118 of file multiscale_feature_persistence.hpp.

template<typename PointSource , typename PointFeature >

void pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::computeFeaturesAtAllScales ( )

Method that calls computeFeatureAtScale () for each scale parameter.

Definition at line 93 of file multiscale_feature_persistence.hpp.

template<typename PointSource , typename PointFeature >

void pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::determinePersistentFeatures	(	FeatureCloud &	output_features,
		boost::shared_ptr< std::vector< int > > &	output_indices
	)

Central function that computes the persistent features.

Parameters:

output_features	a cloud containing the persistent features
output_indices	vector containing the indices of the points in the input cloud that have persistent features, under a one-to-one correspondence with the output_features cloud

Definition at line 195 of file multiscale_feature_persistence.hpp.

template<typename PointSource , typename PointFeature >

float pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::distanceBetweenFeatures	(	const std::vector< float > &	a,
		const std::vector< float > &	b
	)		`[private]`

Function that calculates the scalar difference between two features.

Returns:: the difference as a floating point type

Definition at line 128 of file multiscale_feature_persistence.hpp.

template<typename PointSource , typename PointFeature >

void pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::extractUniqueFeatures ( ) [private]

Selects the so-called 'unique' features from the cloud of features at each level. These features are the ones that fall outside the standard deviation * alpha_.

Definition at line 158 of file multiscale_feature_persistence.hpp.

template<typename PointSource, typename PointFeature>

float pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::getAlpha ( ) [inline]

Get the value of the alpha parameter.

Definition at line 135 of file multiscale_feature_persistence.h.

template<typename PointSource, typename PointFeature>

NormType pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::getDistanceMetric ( ) [inline]

Returns the distance metric that is currently used to calculate the difference between feature vectors.

Definition at line 145 of file multiscale_feature_persistence.h.

template<typename PointSource, typename PointFeature>

FeatureEstimatorPtr pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::getFeatureEstimator ( ) [inline]

Getter method for the feature estimator.

Definition at line 115 of file multiscale_feature_persistence.h.

template<typename PointSource, typename PointFeature>

FeatureRepresentationConstPtr const pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::getPointRepresentation ( ) [inline]

Get a pointer to the feature representation used when converting features into k-D vectors.

Definition at line 125 of file multiscale_feature_persistence.h.

template<typename PointSource, typename PointFeature>

std::vector<float> pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::getScalesVector ( ) [inline]

Method for getting the scale parameters vector.

Definition at line 103 of file multiscale_feature_persistence.h.

template<typename PointSource , typename PointFeature >

bool pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::initCompute ( ) [private]

Checks if all the necessary input was given and the computations can successfully start.

Reimplemented from pcl::PCLBase< PointSource >.

Definition at line 67 of file multiscale_feature_persistence.hpp.

template<typename PointSource, typename PointFeature>

void pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::setAlpha ( float alpha ) [inline]

Sets the alpha parameter.

Parameters:

alpha value to replace the current alpha with

Definition at line 131 of file multiscale_feature_persistence.h.

template<typename PointSource, typename PointFeature>

void pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::setDistanceMetric ( NormType distance_metric ) [inline]

Method for setting the distance metric that will be used for computing the difference between feature vectors.

Parameters:

distance_metric the new distance metric chosen from the NormType enum

Definition at line 141 of file multiscale_feature_persistence.h.

template<typename PointSource, typename PointFeature>

void pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::setFeatureEstimator ( FeatureEstimatorPtr feature_estimator ) [inline]

Setter method for the feature estimator.

Parameters:

feature_estimator pointer to the feature estimator instance that will be used

Note:: the feature estimator instance should already have the input data given beforehand and everything set, ready to be given the compute () command

Definition at line 111 of file multiscale_feature_persistence.h.

template<typename PointSource, typename PointFeature>

void pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::setPointRepresentation ( const FeatureRepresentationConstPtr & feature_representation ) [inline]

Provide a pointer to the feature representation to use to convert features to k-D vectors.

Parameters:

feature_representation the const boost shared pointer to a PointRepresentation

Definition at line 121 of file multiscale_feature_persistence.h.

template<typename PointSource, typename PointFeature>

void pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::setScalesVector ( std::vector< float > & scale_values ) [inline]

Method for setting the scale parameters for the algorithm.

Parameters:

scale_values vector of scales to determine the characteristic of each scaling step

Definition at line 99 of file multiscale_feature_persistence.h.

Member Data Documentation

template<typename PointSource, typename PointFeature>

float pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::alpha_ [private]

Parameter that determines if a feature is to be considered unique or not.

Definition at line 184 of file multiscale_feature_persistence.h.

template<typename PointSource, typename PointFeature>

NormType pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::distance_metric_ [private]

Parameter that determines which distance metric is to be usedto calculate the difference between feature vectors.

Definition at line 187 of file multiscale_feature_persistence.h.

template<typename PointSource, typename PointFeature>

FeatureEstimatorPtr pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::feature_estimator_ [private]

the feature estimator that will be used to determine the feature set at each scale level

Definition at line 190 of file multiscale_feature_persistence.h.

template<typename PointSource, typename PointFeature>

FeatureRepresentationConstPtr pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::feature_representation_ [private]

Definition at line 195 of file multiscale_feature_persistence.h.

template<typename PointSource, typename PointFeature>

std::vector<FeatureCloudPtr> pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::features_at_scale_ [private]

Definition at line 192 of file multiscale_feature_persistence.h.

template<typename PointSource, typename PointFeature>

std::vector<std::vector<std::vector<float> > > pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::features_at_scale_vectorized_ [private]

Definition at line 193 of file multiscale_feature_persistence.h.

template<typename PointSource, typename PointFeature>

std::vector<float> pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::mean_feature_ [private]

Definition at line 194 of file multiscale_feature_persistence.h.

template<typename PointSource, typename PointFeature>

std::vector<float> pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::scale_values_ [private]

The general parameter for determining each scale level.

Definition at line 181 of file multiscale_feature_persistence.h.

template<typename PointSource, typename PointFeature>

std::vector<std::list<size_t> > pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::unique_features_indices_ [private]

Two structures in which to hold the results of the unique feature extraction process. They are superfluous with respect to each other, but improve the time performance of the algorithm.

Definition at line 200 of file multiscale_feature_persistence.h.

template<typename PointSource, typename PointFeature>

std::vector<std::vector<bool> > pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >::unique_features_table_ [private]

Definition at line 201 of file multiscale_feature_persistence.h.

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

Public Types

Public Member Functions

Private Member Functions

Private Attributes

Detailed Description

template<typename PointSource, typename PointFeature> class pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >

Member Typedef Documentation

Constructor & Destructor Documentation

Member Function Documentation

Member Data Documentation

template<typename PointSource, typename PointFeature>
class pcl::MultiscaleFeaturePersistence< PointSource, PointFeature >