Public Types | Public Member Functions | Protected Member Functions

pcl::RIFTEstimation< PointInT, GradientT, PointOutT > Class Template Reference
[Module features]

RIFTEstimation estimates the Rotation Invariant Feature Transform descriptors for a given point cloud dataset containing points and intensity. More...

#include <pcl/features/rift.h>

Inheritance diagram for pcl::RIFTEstimation< PointInT, GradientT, PointOutT >:
Inheritance graph
[legend]
Collaboration diagram for pcl::RIFTEstimation< PointInT, GradientT, PointOutT >:
Collaboration graph
[legend]

List of all members.

Public Types

typedef pcl::PointCloud< PointInT > PointCloudIn
typedef Feature< PointInT,
PointOutT >::PointCloudOut 
PointCloudOut
typedef pcl::PointCloud
< GradientT > 
PointCloudGradient
typedef PointCloudGradient::Ptr PointCloudGradientPtr
typedef
PointCloudGradient::ConstPtr 
PointCloudGradientConstPtr

Public Member Functions

 RIFTEstimation ()
 Empty constructor.
void setInputGradient (const PointCloudGradientConstPtr &gradient)
 Provide a pointer to the input gradient data.
PointCloudGradientConstPtr getInputGradient ()
 Returns a shared pointer to the input gradient data.
void setNrDistanceBins (size_t nr_distance_bins)
 Set the number of bins to use in the distance dimension of the RIFT descriptor.
int getNrDistanceBins ()
 Returns the number of bins in the distance dimension of the RIFT descriptor.
void setNrGradientBins (size_t nr_gradient_bins)
 Set the number of bins to use in the gradient orientation dimension of the RIFT descriptor.
int getNrGradientBins ()
 Returns the number of bins in the gradient orientation dimension of the RIFT descriptor.
void computeRIFT (const PointCloudIn &cloud, const PointCloudGradient &gradient, int p_idx, float radius, const std::vector< int > &indices, const std::vector< float > &squared_distances, Eigen::MatrixXf &rift_descriptor)
 Estimate the Rotation Invariant Feature Transform (RIFT) descriptor for a given point based on its spatial neighborhood of 3D points and the corresponding intensity gradient vector field.

Protected Member Functions

void computeFeature (PointCloudOut &output)
 Estimate the Rotation Invariant Feature Transform (RIFT) descriptors at a set of points given by <setInputCloud (), setIndices ()> using the surface in setSearchSurface (), the gradient in setInputGradient (), and the spatial locator in setSearchMethod ().

Detailed Description

template<typename PointInT, typename GradientT, typename PointOutT>
class pcl::RIFTEstimation< PointInT, GradientT, PointOutT >

RIFTEstimation estimates the Rotation Invariant Feature Transform descriptors for a given point cloud dataset containing points and intensity.

For more information about the RIFT descriptor, see:

Svetlana Lazebnik, Cordelia Schmid, and Jean Ponce. A sparse texture representation using local affine regions. In IEEE Transactions on Pattern Analysis and Machine Intelligence, volume 27, pages 1265-1278, August 2005.

Author:
Michael Dixon

Definition at line 57 of file rift.h.


Member Typedef Documentation

template<typename PointInT , typename GradientT , typename PointOutT >
typedef pcl::PointCloud<GradientT> pcl::RIFTEstimation< PointInT, GradientT, PointOutT >::PointCloudGradient

Definition at line 72 of file rift.h.

template<typename PointInT , typename GradientT , typename PointOutT >
typedef PointCloudGradient::ConstPtr pcl::RIFTEstimation< PointInT, GradientT, PointOutT >::PointCloudGradientConstPtr

Definition at line 74 of file rift.h.

template<typename PointInT , typename GradientT , typename PointOutT >
typedef PointCloudGradient::Ptr pcl::RIFTEstimation< PointInT, GradientT, PointOutT >::PointCloudGradientPtr

Definition at line 73 of file rift.h.

template<typename PointInT , typename GradientT , typename PointOutT >
typedef pcl::PointCloud<PointInT> pcl::RIFTEstimation< PointInT, GradientT, PointOutT >::PointCloudIn

Reimplemented from pcl::Feature< PointInT, PointOutT >.

Definition at line 69 of file rift.h.

template<typename PointInT , typename GradientT , typename PointOutT >
typedef Feature<PointInT, PointOutT>::PointCloudOut pcl::RIFTEstimation< PointInT, GradientT, PointOutT >::PointCloudOut

Reimplemented from pcl::Feature< PointInT, PointOutT >.

Definition at line 70 of file rift.h.


Constructor & Destructor Documentation

template<typename PointInT , typename GradientT , typename PointOutT >
pcl::RIFTEstimation< PointInT, GradientT, PointOutT >::RIFTEstimation (  )  [inline]

Empty constructor.

Definition at line 77 of file rift.h.


Member Function Documentation

template<typename PointInT , typename GradientT , typename PointOutT >
void pcl::RIFTEstimation< PointInT, GradientT, PointOutT >::computeFeature ( PointCloudOut output  )  [protected, virtual]

Estimate the Rotation Invariant Feature Transform (RIFT) descriptors at a set of points given by <setInputCloud (), setIndices ()> using the surface in setSearchSurface (), the gradient in setInputGradient (), and the spatial locator in setSearchMethod ().

Parameters:
output the resultant point cloud model dataset that contains the RIFT feature estimates

Implements pcl::Feature< PointInT, PointOutT >.

Definition at line 113 of file rift.hpp.

template<typename PointInT , typename GradientT , typename PointOutT >
void pcl::RIFTEstimation< PointInT, GradientT, PointOutT >::computeRIFT ( const PointCloudIn cloud,
const PointCloudGradient gradient,
int  p_idx,
float  radius,
const std::vector< int > &  indices,
const std::vector< float > &  squared_distances,
Eigen::MatrixXf &  rift_descriptor 
)

Estimate the Rotation Invariant Feature Transform (RIFT) descriptor for a given point based on its spatial neighborhood of 3D points and the corresponding intensity gradient vector field.

Parameters:
cloud the dataset containing the Cartesian coordinates of the points
gradient the dataset containing the intensity gradient at each point in cloud
p_idx the index of the query point in cloud (i.e. the center of the neighborhood)
radius the radius of the RIFT feature
indices the indices of the points that comprise p_idx's neighborhood in cloud
squared_distances the squared distances from the query point to each point in the neighborhood
rift_descriptor the resultant RIFT descriptor

Eigen::Map<Eigen::Vector3f> point (& (cloud.points[indices[idx]].x));

Definition at line 45 of file rift.hpp.

template<typename PointInT , typename GradientT , typename PointOutT >
PointCloudGradientConstPtr pcl::RIFTEstimation< PointInT, GradientT, PointOutT >::getInputGradient (  )  [inline]

Returns a shared pointer to the input gradient data.

Definition at line 90 of file rift.h.

template<typename PointInT , typename GradientT , typename PointOutT >
int pcl::RIFTEstimation< PointInT, GradientT, PointOutT >::getNrDistanceBins (  )  [inline]

Returns the number of bins in the distance dimension of the RIFT descriptor.

Definition at line 101 of file rift.h.

template<typename PointInT , typename GradientT , typename PointOutT >
int pcl::RIFTEstimation< PointInT, GradientT, PointOutT >::getNrGradientBins (  )  [inline]

Returns the number of bins in the gradient orientation dimension of the RIFT descriptor.

Definition at line 112 of file rift.h.

template<typename PointInT , typename GradientT , typename PointOutT >
void pcl::RIFTEstimation< PointInT, GradientT, PointOutT >::setInputGradient ( const PointCloudGradientConstPtr gradient  )  [inline]

Provide a pointer to the input gradient data.

Parameters:
gradient a pointer to the input gradient data

Definition at line 86 of file rift.h.

template<typename PointInT , typename GradientT , typename PointOutT >
void pcl::RIFTEstimation< PointInT, GradientT, PointOutT >::setNrDistanceBins ( size_t  nr_distance_bins  )  [inline]

Set the number of bins to use in the distance dimension of the RIFT descriptor.

Parameters:
nr_distance_bins the number of bins to use in the distance dimension of the RIFT descriptor

Definition at line 97 of file rift.h.

template<typename PointInT , typename GradientT , typename PointOutT >
void pcl::RIFTEstimation< PointInT, GradientT, PointOutT >::setNrGradientBins ( size_t  nr_gradient_bins  )  [inline]

Set the number of bins to use in the gradient orientation dimension of the RIFT descriptor.

Parameters:
nr_gradient_bins the number of bins to use in the gradient orientation dimension of the RIFT descriptor

Definition at line 108 of file rift.h.


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