SIFTKeypoint detects the Scale Invariant Feature Transform keypoints for a given point cloud dataset containing points and intensity. This implementation adapts the original algorithm from images to point clouds. More...

#include <sift_keypoint.h>

Inheritance diagram for pcl::SIFTKeypoint< PointInT, PointOutT >:

[legend]

List of all members.

Public Types
typedef boost::shared_ptr < const SIFTKeypoint< PointInT, PointOutT > >	ConstPtr
typedef Keypoint< PointInT, PointOutT >::KdTree	KdTree
typedef Keypoint< PointInT, PointOutT >::PointCloudIn	PointCloudIn
typedef Keypoint< PointInT, PointOutT >::PointCloudOut	PointCloudOut
typedef boost::shared_ptr < SIFTKeypoint< PointInT, PointOutT > >	Ptr
Public Member Functions
void	setMinimumContrast (float min_contrast)
	Provide a threshold to limit detection of keypoints without sufficient contrast.
void	setScales (float min_scale, int nr_octaves, int nr_scales_per_octave)
	Specify the range of scales over which to search for keypoints.
	SIFTKeypoint ()
	Empty constructor.
Protected Member Functions
void	detectKeypoints (PointCloudOut &output)
	Detect the SIFT keypoints for a set of points given in setInputCloud () using the spatial locator in setSearchMethod ().
bool	initCompute ()
	This method should get called before starting the actual computation.
Private Member Functions
void	computeScaleSpace (const PointCloudIn &input, KdTree &tree, const std::vector< float > &scales, Eigen::MatrixXf &diff_of_gauss)
	Compute the difference-of-Gaussian (DoG) scale space for the given input and scales.
void	detectKeypointsForOctave (const PointCloudIn &input, KdTree &tree, float base_scale, int nr_scales_per_octave, PointCloudOut &output)
	Detect the SIFT keypoints for a given point cloud for a single octave.
void	findScaleSpaceExtrema (const PointCloudIn &input, KdTree &tree, const Eigen::MatrixXf &diff_of_gauss, std::vector< int > &extrema_indices, std::vector< int > &extrema_scales)
	Find the local minima and maxima in the provided difference-of-Gaussian (DoG) scale space.
Private Attributes
SIFTKeypointFieldSelector < PointInT >	getFieldValue_
float	min_contrast_
	The minimum contrast required for detection.
float	min_scale_
	The standard deviation of the smallest scale in the scale space.
int	nr_octaves_
	The number of octaves (i.e. doublings of scale) over which to search for keypoints.
int	nr_scales_per_octave_
	The number of scales to be computed for each octave.
std::vector< pcl::PCLPointField >	out_fields_
	The list of fields present in the output point cloud data.
int	scale_idx_
	Set to a value different than -1 if the output cloud has a "scale" field and we have to save the keypoints scales.

Detailed Description

template<typename PointInT, typename PointOutT>
class pcl::SIFTKeypoint< PointInT, PointOutT >

SIFTKeypoint detects the Scale Invariant Feature Transform keypoints for a given point cloud dataset containing points and intensity. This implementation adapts the original algorithm from images to point clouds.

For more information about the image-based SIFT interest operator, see:

David G. Lowe, "Distinctive image features from scale-invariant keypoints," International Journal of Computer Vision, 60, 2 (2004), pp. 91-110.

Author:: Michael Dixon

Definition at line 94 of file sift_keypoint.h.

Member Typedef Documentation

template<typename PointInT, typename PointOutT>

typedef boost::shared_ptr<const SIFTKeypoint<PointInT, PointOutT> > pcl::SIFTKeypoint< PointInT, PointOutT >::ConstPtr

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

Definition at line 98 of file sift_keypoint.h.

template<typename PointInT, typename PointOutT>

typedef Keypoint<PointInT, PointOutT>::KdTree pcl::SIFTKeypoint< PointInT, PointOutT >::KdTree

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

Definition at line 102 of file sift_keypoint.h.

template<typename PointInT, typename PointOutT>

typedef Keypoint<PointInT, PointOutT>::PointCloudIn pcl::SIFTKeypoint< PointInT, PointOutT >::PointCloudIn

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

Definition at line 100 of file sift_keypoint.h.

template<typename PointInT, typename PointOutT>

typedef Keypoint<PointInT, PointOutT>::PointCloudOut pcl::SIFTKeypoint< PointInT, PointOutT >::PointCloudOut

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

Definition at line 101 of file sift_keypoint.h.

template<typename PointInT, typename PointOutT>

typedef boost::shared_ptr<SIFTKeypoint<PointInT, PointOutT> > pcl::SIFTKeypoint< PointInT, PointOutT >::Ptr

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

Definition at line 97 of file sift_keypoint.h.

Constructor & Destructor Documentation

template<typename PointInT, typename PointOutT>

pcl::SIFTKeypoint< PointInT, PointOutT >::SIFTKeypoint ( ) [inline]

Empty constructor.

Definition at line 112 of file sift_keypoint.h.

Member Function Documentation

template<typename PointInT , typename PointOutT >

void pcl::SIFTKeypoint< PointInT, PointOutT >::computeScaleSpace	(	const PointCloudIn &	input,
		KdTree &	tree,
		const std::vector< float > &	scales,
		Eigen::MatrixXf &	diff_of_gauss
	)		`[private]`

Compute the difference-of-Gaussian (DoG) scale space for the given input and scales.

Parameters:

input	the point cloud for which the DoG scale space will be computed
tree	a k-D tree of the points in input
scales	a vector containing the scales over which to compute the DoG scale space
diff_of_gauss	the resultant DoG scale space (in a number-of-points by number-of-scales matrix)

Definition at line 207 of file sift_keypoint.hpp.

template<typename PointInT , typename PointOutT >

void pcl::SIFTKeypoint< PointInT, PointOutT >::detectKeypoints ( PointCloudOut & output ) [protected, virtual]

Detect the SIFT keypoints for a set of points given in setInputCloud () using the spatial locator in setSearchMethod ().

Parameters:

output the resultant cloud of keypoints

Implements pcl::Keypoint< PointInT, PointOutT >.

Definition at line 98 of file sift_keypoint.hpp.

template<typename PointInT , typename PointOutT >

void pcl::SIFTKeypoint< PointInT, PointOutT >::detectKeypointsForOctave	(	const PointCloudIn &	input,
		KdTree &	tree,
		float	base_scale,
		int	nr_scales_per_octave,
		PointCloudOut &	output
	)		`[private]`

Detect the SIFT keypoints for a given point cloud for a single octave.

Parameters:

input	the point cloud to detect keypoints in
tree	a k-D tree of the points in input
base_scale	the first (smallest) scale in the octave
nr_scales_per_octave	the number of scales to to compute
output	the resultant point cloud containing the SIFT keypoints

Definition at line 152 of file sift_keypoint.hpp.

template<typename PointInT , typename PointOutT >

void pcl::SIFTKeypoint< PointInT, PointOutT >::findScaleSpaceExtrema	(	const PointCloudIn &	input,
		KdTree &	tree,
		const Eigen::MatrixXf &	diff_of_gauss,
		std::vector< int > &	extrema_indices,
		std::vector< int > &	extrema_scales
	)		`[private]`

Find the local minima and maxima in the provided difference-of-Gaussian (DoG) scale space.

Parameters:

input	the input point cloud
tree	a k-D tree of the points in input
diff_of_gauss	the DoG scale space (in a number-of-points by number-of-scales matrix)
extrema_indices	the resultant vector containing the point indices of each keypoint
extrema_scales	the resultant vector containing the scale indices of each keypoint

Definition at line 258 of file sift_keypoint.hpp.

template<typename PointInT , typename PointOutT >

bool pcl::SIFTKeypoint< PointInT, PointOutT >::initCompute ( ) [protected, virtual]

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 whether a set of input indices has been given. Returns true if yes.
if no input indices have been given, a fake set is created, which will be used until:
- either a new set is given via setIndices(), or
- a new cloud is given that has a different set of points. This will trigger an update on the set of fake indices

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

Definition at line 64 of file sift_keypoint.hpp.

template<typename PointInT , typename PointOutT >

void pcl::SIFTKeypoint< PointInT, PointOutT >::setMinimumContrast ( float min_contrast )

Provide a threshold to limit detection of keypoints without sufficient contrast.

Parameters:

min_contrast the minimum contrast required for detection

Definition at line 57 of file sift_keypoint.hpp.

template<typename PointInT , typename PointOutT >

void pcl::SIFTKeypoint< PointInT, PointOutT >::setScales	(	float	min_scale,
		int	nr_octaves,
		int	nr_scales_per_octave
	)

Specify the range of scales over which to search for keypoints.

Parameters:

min_scale	the standard deviation of the smallest scale in the scale space
nr_octaves	the number of octaves (i.e. doublings of scale) to compute
nr_scales_per_octave	the number of scales to compute within each octave

Definition at line 47 of file sift_keypoint.hpp.