Factory object for creating point cloud objects. More...

#include <cvb/point_cloud_factory.hpp>

Static Public Member Functions
template<class T >
static std::shared_ptr< T >	Load (const String &fileName, PointCloudFlags flags)
	Loads a 3D point, polygon or mesh file (typed). More...

static PointCloudPtr	Load (const String &fileName, PointCloudFlags flags)
	Loads a 3D point, polygon or mesh file. More...

static PointCloudPtr	Create (const ImagePlane &rangeMap, const Calibrator3D &calibrator, PointCloudFlags flags)
	Creates a new Cartesian 3D point cloud from the given 2.5D range map image. More...

template<class T >
static std::shared_ptr< T >	Create (const ImagePlane &rangeMap, const Calibrator3D &calibrator, PointCloudFlags flags)
	Creates a new Cartesian 3D point cloud from the given 2.5D range map image (typed). More...

static PointCloudPtr	CreateWithSettings (const ImagePlane &rangeMap, const Calibrator3D &calibrator, PointCloudFlags flags, const SensorSettings &settings, double encoderStep)
	Creates a new Cartesian 3D point cloud from the given 2.5D range map image with modified camera settings. More...

template<class T >
static std::shared_ptr< T >	CreateWithSettings (const ImagePlane &rangeMap, const Calibrator3D &calibrator, PointCloudFlags flags, const SensorSettings &settings, double encoderStep)
	Creates a new Cartesian 3D point cloud from the given 2.5D range map image with modified camera settings (typed). More...

static SparsePointCloudPtr	CreateSparse (std::size_t numPoints, PointCloudFlags flags)
	Creates a new sparse Cartesian 3D point cloud. More...

static SparsePointCloudPtr	CreateSparse (const std::vector< void * > &planeBasePtrs, int numPlanes, Cvb::DataType dataType, std::size_t numPoints, const std::vector< std::intptr_t > &planeIncs)
	Creates a point cloud from the given memory without copying the data. More...

static DensePointCloudPtr	CreateDense (Size2D< int > size, PointCloudFlags flags)
	Creates a new dense Cartesian 3D point cloud. More...

static DensePointCloudPtr	CreateDense (const Cvb::ImagePlane &rangeMap, Factors3D factors, PointCloudFlags flags)
	Creates a new Cartesian 3D point cloud from the given 2.5D range map image. More...

Detailed Description

Factory object for creating point cloud objects.

Member Function Documentation

◆ Create() [1/2]

static PointCloudPtr Create	(	const ImagePlane &	rangeMap,
		const Calibrator3D &	calibrator,
		PointCloudFlags	flags
	)

inlinestatic

Creates a new Cartesian 3D point cloud from the given 2.5D range map image.

Depending on the calibration method the point cloud resulting from the untyped function PointCloudFactory::Create is a DensePointCloud or a SparsePointCloud. Note, that if this output point cloud is dense, the confidence layer is always available. Background range map pixels are set non-confident. In case of a sparse point cloud, points representing background values and also the confidence layer are removed (except the user explicitly wants to keep the confidence layer which can be requested via the point cloud flags).

Only range maps with floating point and unsigned integer DataType values are supported. For unsigned integer data types, the following DataType::BytesPerPixel values are supported: 1, 2 and 4. Note, that only linear access is supported (check with function ImagePlane::TryLinearAccess).

Note: Creating a calibrated point cloud, the camera settings applied during the acquisition of the range map have to be known. Default settings are read from the calibrator loaded from a calibration file. If the actual settings differ from the default ones, the user has to set them and create a calibrated point cloud with function PointCloudFactory::CreateWithSettings.

Parameters

[in]	rangeMap	Handle of 2.5D image plane.
[in]	calibrator	Calibration object to apply.
[in]	flags	Flags specifying the kind of point cloud to be created.

Returns: A pointer to the point cloud.

Exceptions

Any	exception derived from std::exception including CvbException.

Examples: Cvb/CppPointCloudCreateCalibrated, Foundation/CppMetricCalibration, Foundation/CppMetricCalibrationInclinationLaserPlane, and Foundation/CppMetricCalibrationRigidBodyTrafo.

◆ Create() [2/2]

static std::shared_ptr< T > Create	(	const ImagePlane &	rangeMap,
		const Calibrator3D &	calibrator,
		PointCloudFlags	flags
	)

inlinestatic

Creates a new Cartesian 3D point cloud from the given 2.5D range map image (typed).

The point cloud will automatically be converted to the given type.

Attention: If the the given type is a DensePointCloud, but the created point cloud actually would be a SparsePointCloud (applies to calibrators containing an affine matrix, which can be checked with LaserPlaneCalibrator3D::CorrectionOfLaserPlaneInclination and Calibrator3D::ExtrinsicMatrix), the point cloud will be converted with function SparsePointCloud::ToDensePointCloud. Be aware, that point clouds created with SparsePointCloud::ToDensePointCloud might have gaps! In general type DensePointCloud is not recommended for calibrators containing an affine matrix!

Depending on the calibration method the point cloud resulting from the untyped function PointCloudFactory::Create is a DensePointCloud or a SparsePointCloud. Note, that if this output point cloud is dense, the confidence layer is always available. Background range map pixels are set non-confident. In case of a sparse point cloud, points representing background values and also the confidence layer are removed (except the user explicitly wants to keep the confidence layer which can be requested via the point cloud flags).

Only range maps with floating point and unsigned integer DataType values are supported. For unsigned integer data types, the following DataType::BytesPerPixel values are supported: 1, 2 and 4. Note, that only linear access is supported (check with function ImagePlane::TryLinearAccess).

Note: Creating a calibrated point cloud, the camera settings applied during the acquisition of the range map have to be known. Default settings are read from the calibrator loaded from a calibration file. If the actual settings differ from the default ones, the user has to set them and create a calibrated point cloud with function PointCloudFactory::CreateWithSettings.

Parameters

[in]	rangeMap	Handle of 2.5D image plane.
[in]	calibrator	Calibration object to apply.
[in]	flags	Flags specifying the kind of point cloud to be created.

Returns: A pointer to the point cloud.

Exceptions

Any	exception derived from std::exception including CvbException.

◆ CreateDense() [1/2]

static DensePointCloudPtr CreateDense	(	const Cvb::ImagePlane &	rangeMap,
		Factors3D	factors,
		PointCloudFlags	flags
	)

inlinestatic

Creates a new Cartesian 3D point cloud from the given 2.5D range map image.

Only range maps with floating point and unsigned integer DataType values are supported. For unsigned integer data types, the following DataType::BytesPerPixel values are supported: 1, 2 and 4. Note, that only linear access is supported (check with function ImagePlane::TryLinearAccess).

Parameters

[in]	rangeMap	2.5D image plane.
[in]	factors	Factors to apply.
[in]	flags	Specifying the kind of point cloud to be created.

Returns: The size as number of columns and number of rows.

Exceptions

Any	exception derived from std::exception including CvbException.

◆ CreateDense() [2/2]

static DensePointCloudPtr CreateDense	(	Size2D< int >	size,
		PointCloudFlags	flags
	)

inlinestatic

Creates a new dense Cartesian 3D point cloud.

The resulting point cloud will always have a contiguous point array. Creating the point cloud with with PointCloudFlags::XYZConfidence will provide interleaved XYZConfidence (PointCloudLayout will be Linear), whereas creating it with PointCloudFlags::WithConfidence, only XYZ will be interleaved but not the confidence plane (PointCloudLayout will be Interleaved).

Parameters

[in]	size	Number of Cartesian X, Y, Z(, W) points on the x/y-component of the lattice.
[in]	flags	Specifying the kind of point cloud to be created.

Returns: The variable to receive the handle of the point cloud object.

Exceptions

Any	exception derived from std::exception including CvbException.

Examples: Cvb/CppPointCloudCreateAndHandle.

◆ CreateSparse() [1/2]

static SparsePointCloudPtr CreateSparse	(	const std::vector< void * > &	planeBasePtrs,
		int	numPlanes,
		Cvb::DataType	dataType,
		std::size_t	numPoints,
		const std::vector< std::intptr_t > &	planeIncs
	)

inlinestatic

Creates a point cloud from the given memory without copying the data.

Parameters

[in]	planeBasePtrs	Base pointers of the X, Y, Z and optionally W component planes.
[in]	numPlanes	Number of pPlaneBasePtrs.
[in]	dataType	Data type of the pPlaneBasePtrs.
[in]	numPoints	Number of elements in each plane.
[in]	planeIncs	Offset to the next component in the respective plane in bytes.

Returns: The variable to receive the handle of the point cloud object.

Exceptions

Any	exception derived from std::exception including CvbException.

Currently only single precision (float) data type is supported. Also currently number of planes must be 3.

◆ CreateSparse() [2/2]

static SparsePointCloudPtr CreateSparse	(	std::size_t	numPoints,
		PointCloudFlags	flags
	)

inlinestatic

Creates a new sparse Cartesian 3D point cloud.

The resulting point cloud will always have a contiguous point array. Creating the point cloud with with PointCloudFlags::XYZConfidence will provide interleaved XYZConfidence (PointCloudLayout will be Linear), whereas creating it with PointCloudFlags::WithConfidence, only XYZ will be interleaved but not the confidence plane (PointCloudLayout will be Interleaved).

Parameters

[in]	numPoints	Number of Cartesian X, Y, Z(, W) points.
[in]	flags	Specifying the kind of point cloud to be created.

Returns: The variable to receive the handle of the point cloud object.

Exceptions

Any	exception derived from std::exception including CvbException.

◆ CreateWithSettings() [1/2]

static PointCloudPtr CreateWithSettings	(	const ImagePlane &	rangeMap,
		const Calibrator3D &	calibrator,
		PointCloudFlags	flags,
		const SensorSettings &	settings,
		double	encoderStep
	)

inlinestatic

Creates a new Cartesian 3D point cloud from the given 2.5D range map image with modified camera settings.

Creating a calibrated point cloud, the camera settings applied during the acquisition of the range map have to be known. Default settings are read from the calibrator loaded from a calibration file. If the actual settings differ from the default ones, the user has to set them and create a calibrated point cloud via function this function. You can get and check the default settings via LaserPlaneCalibrator3D::SensorSettings. The default encoder step can also be get from the calibrator via function LaserPlaneCalibrator3D::YFactor.

Depending on the calibration method the point cloud resulting from the untyped function PointCloudFactory::CreateWithSettings is a DensePointCloud or a SparsePointCloud. Note, that if this output point cloud is dense, the confidence layer is always available. Background range map pixels are set non-confident. In case of a sparse point cloud, points representing background values and also the confidence layer are removed (except the user explicitly wants to keep the confidence layer which can be requested via the point cloud flags).

Only range maps with floating point and unsigned integer DataType values are supported. For unsigned integer data types, the following DataType::BytesPerPixel values are supported: 1, 2 and 4. Note, that only linear access is supported (check with function ImagePlane::TryLinearAccess).

Parameters

[in]	rangeMap	Handle of 2.5D image plane.
[in]	calibrator	Calibration object to apply.
[in]	flags	Flags specifying the kind of point cloud to be created.
[in]	settings	Sensor settings applied during the acquisition of the range map.
[in]	encoderStep	Encoder step.

Returns: A pointer to the point cloud.

Exceptions

Any	exception derived from std::exception including CvbException.

◆ CreateWithSettings() [2/2]

static std::shared_ptr< T > CreateWithSettings	(	const ImagePlane &	rangeMap,
		const Calibrator3D &	calibrator,
		PointCloudFlags	flags,
		const SensorSettings &	settings,
		double	encoderStep
	)

inlinestatic

Creates a new Cartesian 3D point cloud from the given 2.5D range map image with modified camera settings (typed).

The point cloud will automatically be converted to the given type.

Attention: If the the given type is a DensePointCloud, but the created point cloud actually would be a SparsePointCloud (applies to calibrators containing an affine matrix, which can be checked with LaserPlaneCalibrator3D::CorrectionOfLaserPlaneInclination and Calibrator3D::ExtrinsicMatrix), the point cloud will be converted with function SparsePointCloud::ToDensePointCloud. Be aware, that point clouds created with SparsePointCloud::ToDensePointCloud might have gaps! In general type DensePointCloud is not recommended for calibrators containing an affine matrix!

Creating a calibrated point cloud, the camera settings applied during the acquisition of the range map have to be known. Default settings are read from the calibrator loaded from a calibration file. If the actual settings differ from the default ones, the user has to set them and create a calibrated point cloud via function this function. You can get and check the default settings via LaserPlaneCalibrator3D::SensorSettings. The default encoder step can also be get from the calibrator via function LaserPlaneCalibrator3D::YFactor.

Depending on the calibration method the point cloud resulting from the untyped function PointCloudFactory::CreateWithSettings is a DensePointCloud or a SparsePointCloud. Note, that if this output point cloud is dense, the confidence layer is always available. Background range map pixels are set non-confident. In case of a sparse point cloud, points representing background values and also the confidence layer are removed (except the user explicitly wants to keep the confidence layer which can be requested via the point cloud flags).

Only range maps with floating point and unsigned integer DataType values are supported. For unsigned integer data types, the following DataType::BytesPerPixel values are supported: 1, 2 and 4. Note, that only linear access is supported (check with function ImagePlane::TryLinearAccess).

Parameters

[in]	rangeMap	Handle of 2.5D image plane.
[in]	calibrator	Calibration object to apply.
[in]	flags	Flags specifying the kind of point cloud to be created.
[in]	settings	Sensor settings applied during the acquisition of the range map.
[in]	encoderStep	Encoder step.

Returns: A pointer to the point cloud.

Exceptions

Any	exception derived from std::exception including CvbException.

◆ Load() [1/2]

static std::shared_ptr< T > Load	(	const String &	fileName,
		PointCloudFlags	flags
	)

inlinestatic

Loads a 3D point, polygon or mesh file (typed).

Parameters

[in]	fileName	Full path to the file to load.
[in]	flags	PointCloudFlags specifying the kind of point cloud to be created.

Returns: A typed pointer to the point cloud.

Exceptions

Any	exception derived from std::exception including CvbException.

◆ Load() [2/2]

static PointCloudPtr Load	(	const String &	fileName,
		PointCloudFlags	flags
	)

inlinestatic

Loads a 3D point, polygon or mesh file.

Parameters

[in]	fileName	Complete path to the file to load.
[in]	flags	Flags specifying the kind of point cloud to be created.

Returns: A pointer to the point cloud.

Exceptions

Any	exception derived from std::exception including CvbException.

Static Public Member Functions

Detailed Description

Member Function Documentation

◆ Create() [1/2]

◆ Create() [2/2]

◆ CreateDense() [1/2]

◆ CreateDense() [2/2]

◆ CreateSparse() [1/2]

◆ CreateSparse() [2/2]

◆ CreateWithSettings() [1/2]

◆ CreateWithSettings() [2/2]

◆ Load() [1/2]

◆ Load() [2/2]