Pierre Alliez, Clément Jamin, Laurent Saboret, Nader Salman, Shihao Wu

This CGAL component implements methods to analyze and process point sets. The input is an unorganized point set, possibly with normal attributes (unoriented or oriented). The point set can be analyzed to measure its average spacing, and processed through functions devised for simplification, regularization, upsampling, outlier removal, smoothing, normal estimation and normal orientation.

Introduced in: CGAL 3.5
Depends on: Eigen
BibTeX: cgal:ass-psp-15a
License: GPL
Windows Demo: See Polyhedral Surface
Common Demo Dlls: dlls

Classified Reference Pages

Functions

Classes
struct	CGAL::value_type_traits< T >
	Class providing the value type of an iterator, and in the case of an output iterator, a type of objects that can be put in it. More...

Functions
template<typename Concurrency_tag , typename ForwardIterator , typename PointPMap , typename NormalPMap , typename Kernel >
double	CGAL::bilateral_smooth_point_set (ForwardIterator first, ForwardIterator beyond, PointPMap point_pmap, NormalPMap normal_pmap, unsigned int k, typename Kernel::FT sharpness_angle, const Kernel &)
	This function smooths an input point set by iteratively projecting each point onto the implicit surface patch fitted over its k nearest neighbors. More...

template<typename InputIterator , typename PointPMap , typename Kernel >
Kernel::FT	CGAL::compute_average_spacing (InputIterator first, InputIterator beyond, PointPMap point_pmap, unsigned int k, const Kernel &)
	Computes average spacing from k nearest neighbors. More...

template<typename OutputIterator , typename ForwardIterator , typename PointPMap , typename NormalPMap , typename Kernel >
OutputIterator	CGAL::edge_aware_upsample_point_set (ForwardIterator first, ForwardIterator beyond, OutputIterator output, PointPMap point_pmap, NormalPMap normal_pmap, const typename Kernel::FT sharpness_angle, typename Kernel::FT edge_sensitivity, typename Kernel::FT neighbor_radius, const unsigned int number_of_output_points, const Kernel &)
	This method progressively upsamples the point set while approaching the edge singularities (detected by normal variation), which generates a denser point set from an input point set. More...

template<typename ForwardIterator , typename PointPMap , typename Kernel >
ForwardIterator	CGAL::grid_simplify_point_set (ForwardIterator first, ForwardIterator beyond, PointPMap point_pmap, double epsilon, const Kernel &)
	Merges points which belong to the same cell of a grid of cell size = `epsilon`. More...

template<typename OutputIteratorValueType , typename OutputIterator , typename PointPMap , typename NormalPMap , typename Kernel >
bool	CGAL::read_off_points_and_normals (std::istream &stream, OutputIterator output, PointPMap point_pmap, NormalPMap normal_pmap, const Kernel &)
	Reads points (positions + normals, if available) from a .off ASCII stream. More...

template<typename OutputIteratorValueType , typename OutputIterator , typename PointPMap , typename Kernel >
bool	CGAL::read_off_points (std::istream &stream, OutputIterator output, PointPMap point_pmap, const Kernel &kernel)
	Reads points (position only) from a .off ASCII stream. More...

template<typename OutputIteratorValueType , typename OutputIterator , typename PointPMap , typename NormalPMap , typename Kernel >
bool	CGAL::read_xyz_points_and_normals (std::istream &stream, OutputIterator output, PointPMap point_pmap, NormalPMap normal_pmap, const Kernel &)
	Reads points (positions + normals, if available) from a .xyz ASCII stream. More...

template<typename OutputIteratorValueType , typename OutputIterator , typename PointPMap , typename Kernel >
bool	CGAL::read_xyz_points (std::istream &stream, OutputIterator output, PointPMap point_pmap, const Kernel &kernel)
	Reads points (positions only) from a .xyz ASCII stream. More...

template<typename ForwardIterator , typename PointPMap , typename NormalPMap , typename Kernel >
bool	CGAL::write_off_points_and_normals (std::ostream &stream, ForwardIterator first, ForwardIterator beyond, PointPMap point_pmap, NormalPMap normal_pmap, const Kernel &)
	Saves the [first, beyond) range of points (positions + normals) to a .off ASCII stream. More...

template<typename ForwardIterator , typename PointPMap , typename Kernel >
bool	CGAL::write_off_points (std::ostream &stream, ForwardIterator first, ForwardIterator beyond, PointPMap point_pmap, const Kernel &)
	Saves the [first, beyond) range of points (positions only) to a .off ASCII stream. More...

template<typename ForwardIterator , typename PointPMap , typename NormalPMap , typename Kernel >
bool	CGAL::write_xyz_points_and_normals (std::ostream &stream, ForwardIterator first, ForwardIterator beyond, PointPMap point_pmap, NormalPMap normal_pmap, const Kernel &)
	Saves the [first, beyond) range of points (positions + normals) to a .xyz ASCII stream. More...

template<typename ForwardIterator , typename PointPMap , typename Kernel >
bool	CGAL::write_xyz_points (std::ostream &stream, ForwardIterator first, ForwardIterator beyond, PointPMap point_pmap, const Kernel &)
	Saves the [first, beyond) range of points (positions only) to a .xyz ASCII stream. More...

template<typename ForwardIterator , typename PointPMap , typename NormalPMap , typename Kernel , typename SvdTraits >
void	CGAL::jet_estimate_normals (ForwardIterator first, ForwardIterator beyond, PointPMap point_pmap, NormalPMap normal_pmap, unsigned int k, const Kernel &, unsigned int degree_fitting=2)
	Estimates normal directions of the `[first, beyond)` range of points using jet fitting on the k nearest neighbors. More...

template<typename InputIterator , typename PointPMap , typename Kernel , typename SvdTraits >
void	CGAL::jet_smooth_point_set (InputIterator first, InputIterator beyond, PointPMap point_pmap, unsigned int k, const Kernel &, unsigned int degree_fitting=2, unsigned int degree_monge=2)
	Smoothes the `[first, beyond)` range of points using jet fitting on the k nearest neighbors and reprojection onto the jet. More...

template<typename ForwardIterator , typename PointPMap , typename NormalPMap , typename Kernel >
ForwardIterator	CGAL::mst_orient_normals (ForwardIterator first, ForwardIterator beyond, PointPMap point_pmap, NormalPMap normal_pmap, unsigned int k, const Kernel &kernel)
	Orients the normals of the `[first, beyond)` range of points using the propagation of a seed orientation through a minimum spanning tree of the Riemannian graph [Hoppe92]. More...

template<typename ForwardIterator , typename PointPMap , typename NormalPMap , typename Kernel >
void	CGAL::pca_estimate_normals (ForwardIterator first, ForwardIterator beyond, PointPMap point_pmap, NormalPMap normal_pmap, unsigned int k, const Kernel &)
	Estimates normal directions of the `[first, beyond)` range of points by linear least squares fitting of a plane over the k nearest neighbors. More...

template<typename ForwardIterator , typename PointPMap , typename Kernel >
ForwardIterator	CGAL::random_simplify_point_set (ForwardIterator first, ForwardIterator beyond, PointPMap, double removed_percentage, const Kernel &)
	Randomly deletes a user-specified fraction of the input points. More...

template<typename InputIterator , typename PointPMap , typename Kernel >
InputIterator	CGAL::remove_outliers (InputIterator first, InputIterator beyond, PointPMap point_pmap, unsigned int k, double threshold_percent, const Kernel &)
	Removes outliers: More...

template<typename Concurrency_tag , typename OutputIterator , typename RandomAccessIterator , typename PointPMap , typename Kernel >
OutputIterator	CGAL::wlop_simplify_and_regularize_point_set (RandomAccessIterator first, RandomAccessIterator beyond, OutputIterator output, PointPMap point_pmap, double select_percentage, double radius, unsigned int iter_number, bool require_uniform_sampling, const Kernel &)
	This is an implementation of the Weighted Locally Optimal Projection (WLOP) simplification algorithm. More...

Function Documentation

template<typename Concurrency_tag , typename ForwardIterator , typename PointPMap , typename NormalPMap , typename Kernel >

double CGAL::bilateral_smooth_point_set	(	ForwardIterator	first,
		ForwardIterator	beyond,
		PointPMap	point_pmap,
		NormalPMap	normal_pmap,
		unsigned int	k,
		typename Kernel::FT	sharpness_angle,
		const Kernel &
	)

This function smooths an input point set by iteratively projecting each point onto the implicit surface patch fitted over its k nearest neighbors.

Bilateral projection preserves sharp features according to the normal (gradient) information. Both point positions and normals will be modified. For more details, please see section 4 in [3].

A parallel version of this function is provided and requires the executable to be linked against the Intel TBB library. To control the number of threads used, the user may use the tbb::task_scheduler_init class. See the TBB documentation for more details.

Precondition: Normals must be unit vectors; k >= 2

Template Parameters

Concurrency_tag	enables sequential versus parallel algorithm. Possible values are `Sequential_tag` and `Parallel_tag`.
ForwardIterator	iterator over input points.
PointPMap	is a model of `ReadWritePropertyMap` with the value type of `ForwardIterator` as key and `Kernel::Point_3` as value type. It can be omitted if the value type of `ForwardIterator` is convertible to `Kernel::Point_3`.
NormalPMap	is a model of `ReadWritePropertyMap` with the value type of `ForwardIterator` as key and `Kernel::Vector_3` as value type.
Kernel	Geometric traits class. It can be omitted and deduced automatically from the value type of `PointPMap` using `Kernel_traits`.

Returns: Average point movement error. It's a convergence criterium for the algorithm. This value can help the user to decide how many iterations are sufficient.

Parameters

first	forward iterator on the first input point.
beyond	past-the-end iterator.
point_pmap	point property map.
normal_pmap	normal property map.
k	size of the neighborhood for the implicit surface patch fitting. The larger the value is, the smoother the result will be.
sharpness_angle	controls the sharpness of the result. The larger the value is, the smoother the result will be. The range of possible value is [0, 90].

#include <CGAL/bilateral_smooth_point_set.h>

template<typename InputIterator , typename PointPMap , typename Kernel >

Kernel::FT CGAL::compute_average_spacing	(	InputIterator	first,
		InputIterator	beyond,
		PointPMap	point_pmap,
		unsigned int	k,
		const Kernel &
	)

Computes average spacing from k nearest neighbors.

Precondition: k >= 2.

Template Parameters

InputIterator	iterator over input points.
PointPMap	is a model of `ReadablePropertyMap` with value type `Point_3<Kernel>`. It can be omitted if the value type of `InputIterator` is convertible to `Point_3<Kernel>`.
Kernel	Geometric traits class. It can be omitted and deduced automatically from the value type of `PointPMap`.

Returns: average spacing (scalar).

Parameters

first	iterator over the first input point.
beyond	past-the-end iterator over the input points.
point_pmap	property map: value_type of InputIterator -> Point_3
k	number of neighbors.

#include <CGAL/compute_average_spacing.h>

Examples:: Point_set_processing_3/average_spacing_example.cpp.

template<typename OutputIterator , typename ForwardIterator , typename PointPMap , typename NormalPMap , typename Kernel >

OutputIterator CGAL::edge_aware_upsample_point_set	(	ForwardIterator	first,
		ForwardIterator	beyond,
		OutputIterator	output,
		PointPMap	point_pmap,
		NormalPMap	normal_pmap,
		const typename Kernel::FT	sharpness_angle,
		typename Kernel::FT	edge_sensitivity,
		typename Kernel::FT	neighbor_radius,
		const unsigned int	number_of_output_points,
		const Kernel &
	)

This method progressively upsamples the point set while approaching the edge singularities (detected by normal variation), which generates a denser point set from an input point set.

This has applications in point-based rendering, hole filling, and sparse surface reconstruction. Normals of points are required as input. For more details, please refer to [3].

Template Parameters

OutputIterator	Type of the output iterator. The type of the objects put in it is `std::pair<Kernel::Point_3, Kernel::Vector_3>`. Note that the user may use a function_output_iterator to match specific needs.
ForwardIterator	iterator over input points.
PointPMap	is a model of `ReadablePropertyMap` with the value type of `ForwardIterator` as key and `Kernel::Point_3` as value type. It can be omitted if the value type of `ForwardIterator` is convertible to `Kernel::Point_3`.
NormalPMap	is a model of `ReadablePropertyMap` with the value type of `ForwardIterator` as key and `Kernel::Vector_3` as value type.
Kernel	Geometric traits class. It can be omitted and deduced automatically from the value type of `PointPMap` using `Kernel_traits`.

Parameters

first	forward iterator on the first input point.
beyond	past-the-end iterator.
output	output iterator where output points and normals are put.
point_pmap	point property map.
normal_pmap	vector property map.
sharpness_angle	controls the preservation of sharp features. The larger the value is, the smoother the result will be. The range of possible value is [0, 90]. See section Parameter: sharpness_angle for an example.
edge_sensitivity	larger values of edge-sensitivity give higher priority to inserting points along sharp features. The range of possible values is `[0, 1]`. See section Parameter: edge_sensitivity for an example.
neighbor_radius	indicates the radius of the largest hole that should be filled. The default value is set to 3 times the average spacing of the point set. If the value given by user is smaller than the average spacing, the function will use the default value instead.
number_of_output_points	number of output points to generate.

#include <CGAL/edge_aware_upsample_point_set.h>

Examples:: Point_set_processing_3/edge_aware_upsample_point_set_example.cpp.

template<typename ForwardIterator , typename PointPMap , typename Kernel >

ForwardIterator CGAL::grid_simplify_point_set	(	ForwardIterator	first,
		ForwardIterator	beyond,
		PointPMap	point_pmap,
		double	epsilon,
		const Kernel &
	)

Merges points which belong to the same cell of a grid of cell size = epsilon.

This method modifies the order of input points so as to pack all remaining points first, and returns an iterator over the first point to remove (see erase-remove idiom). For this reason it should not be called on sorted containers.

Precondition: epsilon > 0

Template Parameters

ForwardIterator	iterator over input points.
PointPMap	is a model of `ReadablePropertyMap` with value type `Point_3<Kernel>`. It can be omitted if the value type of `ForwardIterator` is convertible to `Point_3<Kernel>`.
Kernel	Geometric traits class. It can be omitted and deduced automatically from the value type of `PointPMap`.

Returns: iterator over the first point to remove.

Parameters

first	iterator over the first input point.
beyond	past-the-end iterator over the input points.
point_pmap	property map: value_type of ForwardIterator -> Point_3
epsilon	tolerance value when merging 3D points.

#include <CGAL/grid_simplify_point_set.h>

Examples:: Point_set_processing_3/grid_simplification_example.cpp.

template<typename ForwardIterator , typename PointPMap , typename NormalPMap , typename Kernel , typename SvdTraits >

void CGAL::jet_estimate_normals	(	ForwardIterator	first,
		ForwardIterator	beyond,
		PointPMap	point_pmap,
		NormalPMap	normal_pmap,
		unsigned int	k,
		const Kernel &	,
		unsigned int	degree_fitting = `2`
	)

Estimates normal directions of the [first, beyond) range of points using jet fitting on the k nearest neighbors.

The output normals are randomly oriented.

Precondition: k >= 2

Template Parameters

ForwardIterator	iterator model of the concept of the same name over input points and able to store output normals.
PointPMap	is a model of `ReadablePropertyMap` with value type `Point_3<Kernel>`. It can be omitted if the value type of `ForwardIterator` is convertible to `Point_3<Kernel>`.
NormalPMap	is a model of `WritablePropertyMap` with value type `Vector_3<Kernel>`.
Kernel	Geometric traits class. It can be omitted and deduced automatically from the value type of `PointPMap`.
SvdTraits	template parameter for the class `Monge_via_jet_fitting` that can be ommited under conditions described in the documentation of `Monge_via_jet_fitting`.

Parameters

first	iterator over the first input point.
beyond	past-the-end iterator over the input points.
point_pmap	property map: value_type of ForwardIterator -> Point_3.
normal_pmap	property map: value_type of ForwardIterator -> Vector_3.
k	number of neighbors.
degree_fitting	fitting degree

#include <CGAL/jet_estimate_normals.h>

template<typename InputIterator , typename PointPMap , typename Kernel , typename SvdTraits >

void CGAL::jet_smooth_point_set	(	InputIterator	first,
		InputIterator	beyond,
		PointPMap	point_pmap,
		unsigned int	k,
		const Kernel &	,
		unsigned int	degree_fitting = `2`,
		unsigned int	degree_monge = `2`
	)

Smoothes the [first, beyond) range of points using jet fitting on the k nearest neighbors and reprojection onto the jet.

As this method relocates the points, it should not be called on containers sorted w.r.t. point locations.

Precondition: k >= 2

Template Parameters

InputIterator	iterator over input points.
PointPMap	is a model of `ReadWritePropertyMap` with value type `Point_3<Kernel>`. It can be omitted if the value type of `InputIterator` is convertible to `Point_3<Kernel>`.
Kernel	Geometric traits class. It can be omitted and deduced automatically from the value type of `PointPMap`.
SvdTraits	template parameter for the class `Monge_via_jet_fitting` that can be ommited under conditions described in the documentation of `Monge_via_jet_fitting`.

Parameters

first	iterator over the first input point.
beyond	past-the-end iterator over the input points.
point_pmap	property map: value_type of InputIterator -> Point_3.
k	number of neighbors.
degree_fitting	fitting degree
degree_monge	Monge degree

#include <CGAL/jet_smooth_point_set.h>

Examples:: Point_set_processing_3/jet_smoothing_example.cpp.

template<typename ForwardIterator , typename PointPMap , typename NormalPMap , typename Kernel >

ForwardIterator CGAL::mst_orient_normals	(	ForwardIterator	first,
		ForwardIterator	beyond,
		PointPMap	point_pmap,
		NormalPMap	normal_pmap,
		unsigned int	k,
		const Kernel &	kernel
	)

Orients the normals of the [first, beyond) range of points using the propagation of a seed orientation through a minimum spanning tree of the Riemannian graph [Hoppe92].

This method modifies the order of input points so as to pack all sucessfully oriented points first, and returns an iterator over the first point with an unoriented normal (see erase-remove idiom). For this reason it should not be called on sorted containers.

Warning: This function may fail when Boost version 1.54 is used, because of the following bug: https://svn.boost.org/trac/boost/ticket/9012

Precondition: Normals must be unit vectors; k >= 2

Template Parameters

ForwardIterator	iterator over input points.
PointPMap	is a model of `ReadablePropertyMap` with value type `Point_3<Kernel>`. It can be omitted if the value type of `ForwardIterator` is convertible to `Point_3<Kernel>`.
NormalPMap	is a model of `ReadWritePropertyMap` with value type `Vector_3<Kernel>` .
Kernel	Geometric traits class. It can be omitted and deduced automatically from the value type of `PointPMap`.

Returns: iterator over the first point with an unoriented normal.

Parameters

first	iterator over the first input point.
beyond	past-the-end iterator over the input points.
point_pmap	property map: value_type of ForwardIterator -> Point_3.
normal_pmap	property map: value_type of ForwardIterator -> Vector_3.
k	number of neighbors
kernel	geometric traits.

#include <CGAL/mst_orient_normals.h>

Examples:: Point_set_processing_3/normals_example.cpp.

template<typename ForwardIterator , typename PointPMap , typename NormalPMap , typename Kernel >

void CGAL::pca_estimate_normals	(	ForwardIterator	first,
		ForwardIterator	beyond,
		PointPMap	point_pmap,
		NormalPMap	normal_pmap,
		unsigned int	k,
		const Kernel &
	)

Estimates normal directions of the [first, beyond) range of points by linear least squares fitting of a plane over the k nearest neighbors.

The output normals are randomly oriented.

Precondition: k >= 2

Template Parameters

ForwardIterator	iterator over input points.
PointPMap	is a model of `ReadablePropertyMap` with value type `Point_3<Kernel>`. It can be omitted if the value type of `ForwardIterator` is convertible to `Point_3<Kernel>`.
NormalPMap	is a model of `WritablePropertyMap` with value type `Vector_3<Kernel>`.
Kernel	Geometric traits class. It can be omitted and deduced automatically from the value type of `PointPMap`.

Parameters

first	iterator over the first input point.
beyond	past-the-end iterator over the input points.
point_pmap	property map: value_type of ForwardIterator -> Point_3.
normal_pmap	property map: value_type of ForwardIterator -> Vector_3.
k	number of neighbors.

#include <CGAL/pca_estimate_normals.h>

Examples:: Point_set_processing_3/normals_example.cpp.

template<typename ForwardIterator , typename PointPMap , typename Kernel >

ForwardIterator CGAL::random_simplify_point_set	(	ForwardIterator	first,
		ForwardIterator	beyond,
		PointPMap	,
		double	removed_percentage,
		const Kernel &
	)

Randomly deletes a user-specified fraction of the input points.

This method modifies the order of input points so as to pack all remaining points first, and returns an iterator over the first point to remove (see erase-remove idiom). For this reason it should not be called on sorted containers.

Template Parameters

ForwardIterator	iterator over input points.
PointPMap	is a model of `ReadablePropertyMap` with value type `Point_3<Kernel>`. It can be omitted if the value type of `ForwardIterator` is convertible to `Point_3<Kernel>`.
Kernel	Geometric traits class. It can be omitted and deduced automatically from the value type of `PointPMap`.

Returns: iterator over the first point to remove.

Parameters

first	iterator over the first input point.
beyond	past-the-end iterator over the input points.
removed_percentage	percentage of points to remove.

#include <CGAL/random_simplify_point_set.h>

template<typename OutputIteratorValueType , typename OutputIterator , typename PointPMap , typename Kernel >

bool CGAL::read_off_points	(	std::istream &	stream,
		OutputIterator	output,
		PointPMap	point_pmap,
		const Kernel &	kernel
	)

Reads points (position only) from a .off ASCII stream.

The function expects for each point a line with the x y z position. If the position is followed by the nx ny nz normal, then the normal will be ignored. Faces are ignored.

Template Parameters

OutputIteratorValueType	type of objects that can be put in `OutputIterator`. It is default to `value_type_traits<OutputIterator>::type` and can be omitted when the default is fine.
OutputIterator	iterator over output points.
PointPMap	is a model of `WritablePropertyMap` with value_type `Point_3<Kernel>`. It can be omitted if the value type of `OutputIterator` is convertible to `Point_3<Kernel>`.
Kernel	Geometric traits class. It can be omitted and deduced automatically from the value type of `PointPMap`.

Returns: true on success.

Parameters

stream	input stream.
output	output iterator over points.
point_pmap	property map: value_type of OutputIterator -> Point_3.
kernel	geometric traits.

#include <CGAL/IO/read_off_points.h>

template<typename OutputIteratorValueType , typename OutputIterator , typename PointPMap , typename NormalPMap , typename Kernel >

bool CGAL::read_off_points_and_normals	(	std::istream &	stream,
		OutputIterator	output,
		PointPMap	point_pmap,
		NormalPMap	normal_pmap,
		const Kernel &
	)

Reads points (positions + normals, if available) from a .off ASCII stream.

The function expects for each point a line with the x y z position, optionally followed by the nx ny nz normal. Faces are ignored.

Template Parameters

OutputIteratorValueType	type of objects that can be put in `OutputIterator`. It is default to `value_type_traits<OutputIterator>::type` and can be omitted when the default is fine.
OutputIterator	iterator over output points.
PointPMap	is a model of `WritablePropertyMap` with value type `Point_3<Kernel>`. It can be omitted if the value type of `OutputIterator` is convertible to `Point_3<Kernel>`.
NormalPMap	is a model of `WritablePropertyMap` with value type `Vector_3<Kernel>`.
Kernel	Geometric traits class. It can be omitted and deduced automatically from the value type of `PointPMap`.

Returns: true on success.

Parameters

stream	input stream.
output	output iterator over points.
point_pmap	property map: value_type of OutputIterator -> Point_3.
normal_pmap	property map: value_type of OutputIterator -> Vector_3.

#include <CGAL/IO/read_off_points.h>

template<typename OutputIteratorValueType , typename OutputIterator , typename PointPMap , typename Kernel >

bool CGAL::read_xyz_points	(	std::istream &	stream,
		OutputIterator	output,
		PointPMap	point_pmap,
		const Kernel &	kernel
	)

Reads points (positions only) from a .xyz ASCII stream.

The function expects for each point a line with the x y z position. If the position is followed by the nx ny nz normal, then the normal will be ignored. The first line may contain the number of points in the file. Empty lines and comments starting by # character are allowed.

Template Parameters

OutputIteratorValueType	type of objects that can be put in `OutputIterator`. It is default to `value_type_traits<OutputIterator>::type` and can be omitted when the default is fine.
OutputIterator	iterator over output points.
PointPMap	is a model of `WritablePropertyMap` with value type `Point_3<Kernel>`. It can be omitted if the value type of `OutputIterator` is convertible to `Point_3<Kernel>`.
Kernel	Geometric traits class. It can be omitted and deduced automatically from the value type of `PointPMap`.

Returns: true on success.

Parameters

stream	input stream.
output	output iterator over points.
point_pmap	property map: value_type of OutputIterator -> Point_3.
kernel	geometric traits.

#include <CGAL/IO/read_xyz_points.h>

Examples:: Point_set_processing_3/average_spacing_example.cpp, Point_set_processing_3/grid_simplification_example.cpp, Point_set_processing_3/normals_example.cpp, Point_set_processing_3/remove_outliers_example.cpp, and Point_set_processing_3/wlop_simplify_and_regularize_point_set_example.cpp.

template<typename OutputIteratorValueType , typename OutputIterator , typename PointPMap , typename NormalPMap , typename Kernel >

bool CGAL::read_xyz_points_and_normals	(	std::istream &	stream,
		OutputIterator	output,
		PointPMap	point_pmap,
		NormalPMap	normal_pmap,
		const Kernel &
	)

Reads points (positions + normals, if available) from a .xyz ASCII stream.

The function expects for each point a line with the x y z position, optionally followed by the nx ny nz normal. The first line may contain the number of points in the file. Empty lines and comments starting by # character are allowed.

Template Parameters

OutputIteratorValueType	type of objects that can be put in `OutputIterator`. It is default to `value_type_traits<OutputIterator>::type` and can be omitted when the default is fine.
OutputIterator	iterator over output points.
PointPMap	is a model of `WritablePropertyMap` with value type `Point_3<Kernel>`. It can be omitted if the value type of OutputIterator value_type is convertible to `Point_3<Kernel>`.
NormalPMap	is a model of `WritablePropertyMap` with value type `Vector_3<Kernel>`.
Kernel	Geometric traits class. It can be omitted and deduced automatically from the value type of `PointPMap`.

Returns: true on success.

Parameters

stream	input stream.
output	output iterator over points.
point_pmap	property map: value_type of OutputIterator -> Point_3.
normal_pmap	property map: value_type of OutputIterator -> Vector_3.

#include <CGAL/IO/read_xyz_points.h>

Examples:: Point_set_processing_3/bilateral_smooth_point_set_example.cpp, Point_set_processing_3/edge_aware_upsample_point_set_example.cpp, and Point_set_processing_3/read_write_xyz_point_set_example.cpp.

template<typename InputIterator , typename PointPMap , typename Kernel >

InputIterator CGAL::remove_outliers	(	InputIterator	first,
		InputIterator	beyond,
		PointPMap	point_pmap,
		unsigned int	k,
		double	threshold_percent,
		const Kernel &
	)

Removes outliers:

computes average squared distance to the K nearest neighbors,
and sorts the points in increasing order of average distance.

This method modifies the order of input points so as to pack all remaining points first, and returns an iterator over the first point to remove (see erase-remove idiom). For this reason it should not be called on sorted containers.

Precondition: k >= 2

Template Parameters

InputIterator	iterator over input points.
PointPMap	is a model of `ReadablePropertyMap` with value type `Point_3<Kernel>`. It can be omitted ifthe value type of `InputIterator` is convertible to `Point_3<Kernel>`.
Kernel	Geometric traits class. It can be omitted and deduced automatically from the value type of `PointPMap`.

Returns: iterator over the first point to remove.

Parameters

first	iterator over the first input point.
beyond	past-the-end iterator over the input points.
point_pmap	property map: value_type of InputIterator -> Point_3
k	number of neighbors.
threshold_percent	percentage of points to remove.

#include <CGAL/remove_outliers.h>

Examples:: Point_set_processing_3/remove_outliers_example.cpp.

template<typename Concurrency_tag , typename OutputIterator , typename RandomAccessIterator , typename PointPMap , typename Kernel >

OutputIterator CGAL::wlop_simplify_and_regularize_point_set	(	RandomAccessIterator	first,
		RandomAccessIterator	beyond,
		OutputIterator	output,
		PointPMap	point_pmap,
		double	select_percentage,
		double	radius,
		unsigned int	iter_number,
		bool	require_uniform_sampling,
		const Kernel &
	)

This is an implementation of the Weighted Locally Optimal Projection (WLOP) simplification algorithm.

The WLOP simplification algorithm can produce a set of denoised, outlier-free and evenly distributed particles over the original dense point cloud. The core of the algorithm is a Weighted Locally Optimal Projection operator with a density uniformization term. For more details, please refer to [2].

A parallel version of WLOP is provided and requires the executable to be linked against the Intel TBB library. To control the number of threads used, the user may use the tbb::task_scheduler_init class. See the TBB documentation for more details.

Template Parameters

Concurrency_tag	enables sequential versus parallel algorithm. Possible values are `Sequential_tag` and `Parallel_tag`.
OutputIterator	Type of the output iterator. It must accept objects of type `Kernel::Point_3`.
RandomAccessIterator	Iterator over input points.
PointPMap	is a model of `ReadablePropertyMap` with the value type of `ForwardIterator` as key type and `Kernel::Point_3` as value type. It can be omitted if the value type of `RandomAccessIterator` is convertible to `Kernel::Point_3`.
Kernel	Geometric traits class. It can be omitted and deduced automatically from the value type of `PointPMap` using `Kernel_traits`.

Parameters

first	random-access iterator to the first input point.
beyond	past-the-end iterator.
output	output iterator where output points are put.
point_pmap	point property map.
select_percentage	percentage of points to retain. The default value is set to 5 (%).
radius	spherical neighborhood radius. This is a key parameter that needs to be finely tuned. The result will be irregular if too small, but a larger value will impact the runtime. In practice, choosing a radius such that the neighborhood of each sample point includes at least two rings of neighboring sample points gives satisfactory result. The default value is set to 8 times the average spacing of the point set.
iter_number	number of iterations to solve the optimsation problem. The default value is 35. More iterations give a more regular result but increase the runtime.
require_uniform_sampling	an optional preprocessing, which will give better result if the distribution of the input points is highly non-uniform. The default value is `false`.

#include <CGAL/wlop_simplify_and_regularize_point_set.h>

Examples:: Point_set_processing_3/wlop_simplify_and_regularize_point_set_example.cpp.

template<typename ForwardIterator , typename PointPMap , typename Kernel >

bool CGAL::write_off_points	(	std::ostream &	stream,
		ForwardIterator	first,
		ForwardIterator	beyond,
		PointPMap	point_pmap,
		const Kernel &
	)

Saves the [first, beyond) range of points (positions only) to a .off ASCII stream.

The function writes for each point a line with the x y z position.

Template Parameters

ForwardIterator	iterator over input points.
PointPMap	is a model of `ReadablePropertyMap` with a value_type = `Point_3<Kernel>`. It can be omitted if the value type of `ForwardIterator` is convertible to `Point_3<Kernel>`.
Kernel	Geometric traits class. It can be omitted and deduced automatically from the value type of `PointPMap`.

Returns: true on success.

Parameters

stream	output stream.
first	iterator over the first input point.
beyond	past-the-end iterator over the input points.
point_pmap	property map: value_type of ForwardIterator -> Point_3.

#include <CGAL/IO/write_off_points.h>

template<typename ForwardIterator , typename PointPMap , typename NormalPMap , typename Kernel >

bool CGAL::write_off_points_and_normals	(	std::ostream &	stream,
		ForwardIterator	first,
		ForwardIterator	beyond,
		PointPMap	point_pmap,
		NormalPMap	normal_pmap,
		const Kernel &
	)

Saves the [first, beyond) range of points (positions + normals) to a .off ASCII stream.

The function writes for each point a line with the x y z position followed by the nx ny nz normal.

Precondition: normals must be unit vectors

Template Parameters

ForwardIterator	iterator over input points.
PointPMap	is a model of `ReadablePropertyMap` with value type `Point_3<Kernel>`. It can be omitted if the value type of `ForwardIterator` is convertible to `Point_3<Kernel>`.
NormalPMap	is a model of `ReadablePropertyMap` with a value type `Vector_3<Kernel>`.
Kernel	Geometric traits class. It can be omitted and deduced automatically from the value type of `PointPMap`.

Returns: true on success.

Parameters

stream	output stream.
first	iterator over the first input point.
beyond	past-the-end iterator over the input points.
point_pmap	property map: value_type of ForwardIterator -> Point_3.
normal_pmap	property map: value_type of ForwardIterator -> Vector_3.

#include <CGAL/IO/write_off_points.h>

template<typename ForwardIterator , typename PointPMap , typename Kernel >

bool CGAL::write_xyz_points	(	std::ostream &	stream,
		ForwardIterator	first,
		ForwardIterator	beyond,
		PointPMap	point_pmap,
		const Kernel &
	)

Saves the [first, beyond) range of points (positions only) to a .xyz ASCII stream.

The function writes for each point a line with the x y z position.

Template Parameters

ForwardIterator	iterator over input points.
PointPMap	is a model of `ReadablePropertyMap` with value type `Point_3<Kernel>`. It can be omitted if the value type of `ForwardIterator value_type is convertible to`Point_3<Kernel>`.
Kernel	Geometric traits class. It can be omitted and deduced automatically from the value type of `PointPMap`.

Returns: true on success.

Parameters

stream	output stream.
first	iterator over the first input point.
beyond	past-the-end iterator over the input points.
point_pmap	property map: value_type of ForwardIterator -> Point_3.

#include <CGAL/IO/write_xyz_points.h>

Examples:: Point_set_processing_3/wlop_simplify_and_regularize_point_set_example.cpp.

template<typename ForwardIterator , typename PointPMap , typename NormalPMap , typename Kernel >

bool CGAL::write_xyz_points_and_normals	(	std::ostream &	stream,
		ForwardIterator	first,
		ForwardIterator	beyond,
		PointPMap	point_pmap,
		NormalPMap	normal_pmap,
		const Kernel &
	)

Saves the [first, beyond) range of points (positions + normals) to a .xyz ASCII stream.

The function writes for each point a line with the x y z position followed by the nx ny nz normal.

Precondition: normals must be unit vectors

Template Parameters

ForwardIterator	iterator over input points.
PointPMap	is a model of `ReadablePropertyMap` with a value type = `Point_3<Kernel>`. It can be omitted if the value type of `ForwardIterator value_type is convertible to`Point_3<Kernel>`. @tparam NormalPMap is a model of`ReadablePropertyMap`with a value type`Vector_3<Kernel>`.
Kernel	Geometric traits class. It can be omitted and deduced automatically from the value type of `PointPMap`.

Returns: true on success.

Parameters

stream	output stream.
first	iterator over the first input point.
beyond	past-the-end iterator over the input points.
point_pmap	property map: value_type of ForwardIterator -> Point_3.
normal_pmap	property map: value_type of ForwardIterator -> Vector_3.

#include <CGAL/IO/write_xyz_points.h>

Examples:: Point_set_processing_3/bilateral_smooth_point_set_example.cpp, Point_set_processing_3/edge_aware_upsample_point_set_example.cpp, and Point_set_processing_3/read_write_xyz_point_set_example.cpp.

Classified Reference Pages

Functions

Classes

Functions

Function Documentation