GPU-accelerated Voxelized GICP matching cost factor Koide et al., "Voxelized GICP for Fast and Accurate 3D Point Cloud Registration", ICRA2021 Koide et al., "Globally Consistent 3D LiDAR Mapping with GPU-accelerated GICP Matching Cost Factors", RA-L2021. More...

#include <integrated_vgicp_factor_gpu.hpp>

Inheritance diagram for gtsam_points::IntegratedVGICPFactorGPU:

[legend]

Collaboration diagram for gtsam_points::IntegratedVGICPFactorGPU:

[legend]

Public Types
using	shared_ptr = gtsam_points::shared_ptr< IntegratedVGICPFactorGPU >

Public Types inherited from gtsam_points::NonlinearFactorGPU
using	shared_ptr = gtsam_points::shared_ptr< NonlinearFactorGPU >

Public Member Functions
	IntegratedVGICPFactorGPU (gtsam::Key target_key, gtsam::Key source_key, const GaussianVoxelMap::ConstPtr &target, const PointCloud::ConstPtr &source, CUstream_st *stream=nullptr, std::shared_ptr< TempBufferManager > temp_buffer=nullptr)
	Create a binary VGICP_GPU factor between target and source poses.

	IntegratedVGICPFactorGPU (const gtsam::Pose3 &fixed_target_pose, gtsam::Key source_key, const GaussianVoxelMap::ConstPtr &target, const PointCloud::ConstPtr &source, CUstream_st *stream=nullptr, std::shared_ptr< TempBufferManager > temp_buffer=nullptr)
	Create a unary VGICP_GPU factor between a fixed target pose and an active source pose.

virtual void	print (const std::string &s="", const gtsam::KeyFormatter &keyFormatter=gtsam::DefaultKeyFormatter) const override
	Print the factor information.

size_t	memory_usage () const
	Calculate the CPU memory usage of this factor.

size_t	memory_usage_gpu () const
	Calculate the GPU memory usage of this factor.

void	set_enable_offloading (bool enable)
	Enable or disable GPU memory offloading.

void	set_enable_surface_validation (bool enable)
	Enable or disable surface orientation validation for correspondence search.

void	set_inlier_update_thresh (double trans, double angle)
	Set the threshold values to trigger inlier points update. Setting larger values reduces GPU sync but may affect the registration accuracy.

int	num_inliers () const
	Get the number of inlier points.

double	inlier_fraction () const
	Get the fraction of inlier points.

GaussianVoxelMapGPU::ConstPtr	get_target () const
	Get the target voxelmap.

Eigen::Isometry3f	get_fixed_target_pose () const
	Get the pose of the fixed target. This function is only valid for unary factors.

	IntegratedVGICPFactorGPU (const IntegratedVGICPFactorGPU &)=delete

IntegratedVGICPFactorGPU &	operator= (const IntegratedVGICPFactorGPU &)=delete

virtual gtsam::NonlinearFactor::shared_ptr	clone () const override

virtual size_t	dim () const override

virtual double	error (const gtsam::Values &values) const override

virtual gtsam::GaussianFactor::shared_ptr	linearize (const gtsam::Values &values) const override

virtual size_t	linearization_input_size () const override
	Size of data to be uploaded to the GPU before linearization.

virtual size_t	linearization_output_size () const override
	Size of data to be downloaded from the GPU after linearization.

virtual size_t	evaluation_input_size () const override
	Size of data to be uploaded to the GPU before cost evaluation.

virtual size_t	evaluation_output_size () const override
	Size of data to be downloaded from the GPU after cost evaluation.

virtual void	set_linearization_point (const gtsam::Values &values, void *lin_input_cpu) override
	Write linearization input data to the upload buffer.

virtual void	issue_linearize (const void lin_input_cpu, const void lin_input_gpu, void *lin_output_gpu) override
	Issue linearization task.

virtual void	store_linearized (const void *lin_output_cpu) override
	Read linearization output data from the download buffer.

virtual void	set_evaluation_point (const gtsam::Values &values, void *eval_input_cpu) override
	Write cost evaluation input data to the upload buffer.

virtual void	issue_compute_error (const void lin_input_cpu, const void eval_input_cpu, const void lin_input_gpu, const void eval_input_gpu, void *eval_output_gpu) override
	Issue cost evaluation task.

virtual void	store_computed_error (const void *eval_output_cpu) override
	Read cost evaluation output data from the download buffer.

virtual void	sync () override
	Perform CPU-GPU synchronization and wait for the task.

Public Member Functions inherited from gtsam_points::NonlinearFactorGPU
template<typename CONTAINER >
	NonlinearFactorGPU (const CONTAINER &keys)

Detailed Description

GPU-accelerated Voxelized GICP matching cost factor Koide et al., "Voxelized GICP for Fast and Accurate 3D Point Cloud Registration", ICRA2021 Koide et al., "Globally Consistent 3D LiDAR Mapping with GPU-accelerated GICP Matching Cost Factors", RA-L2021.

Constructor & Destructor Documentation

◆ IntegratedVGICPFactorGPU() [1/2]

gtsam_points::IntegratedVGICPFactorGPU::IntegratedVGICPFactorGPU	(	gtsam::Key	target_key,
		gtsam::Key	source_key,
		const GaussianVoxelMap::ConstPtr &	target,
		const PointCloud::ConstPtr &	source,
		CUstream_st *	stream = `nullptr`,
		std::shared_ptr< TempBufferManager >	temp_buffer = `nullptr`
	)

Create a binary VGICP_GPU factor between target and source poses.

Parameters

target_key	Target key
source_key	Source key
target	Target voxelmap
source	Source frame
stream	CUDA stream
temp_buffer	CUDA temporary buffer manager

◆ IntegratedVGICPFactorGPU() [2/2]

gtsam_points::IntegratedVGICPFactorGPU::IntegratedVGICPFactorGPU	(	const gtsam::Pose3 &	fixed_target_pose,
		gtsam::Key	source_key,
		const GaussianVoxelMap::ConstPtr &	target,
		const PointCloud::ConstPtr &	source,
		CUstream_st *	stream = `nullptr`,
		std::shared_ptr< TempBufferManager >	temp_buffer = `nullptr`
	)

Create a unary VGICP_GPU factor between a fixed target pose and an active source pose.

Parameters

targfixed_target_pose	Fixed target pose
source_key	Source key
target	Target voxelmap
source	Source frame
stream	CUDA stream
temp_buffer	CUDA temporary buffer manager

Member Function Documentation

◆ evaluation_input_size()

virtual size_t gtsam_points::IntegratedVGICPFactorGPU::evaluation_input_size ( ) const

overridevirtual

Size of data to be uploaded to the GPU before cost evaluation.

Returns: size_t

Implements gtsam_points::NonlinearFactorGPU.

◆ evaluation_output_size()

virtual size_t gtsam_points::IntegratedVGICPFactorGPU::evaluation_output_size ( ) const

overridevirtual

Size of data to be downloaded from the GPU after cost evaluation.

Returns: size_t

Implements gtsam_points::NonlinearFactorGPU.

◆ inlier_fraction()

double gtsam_points::IntegratedVGICPFactorGPU::inlier_fraction ( ) const

Get the fraction of inlier points.

Note: This function must be called after the factor is linearized.

◆ issue_compute_error()

virtual void gtsam_points::IntegratedVGICPFactorGPU::issue_compute_error	(	const void *	lin_input_cpu,
		const void *	eval_input_cpu,
		const void *	lin_input_gpu,
		const void *	eval_input_gpu,
		void *	eval_output_gpu
	)

overridevirtual

Issue cost evaluation task.

Parameters

lin_input_cpu	Linearization input data on the CPU memory (size == linearization_input_size)
eval_input_cpu	Cost evaluation input data on the CPU memory (size == evaluation_input_size)
lin_input_gpu	Linearization input data on the GPU memory (size == linearization_input_size)
eval_input_gpu	Cost evaluation input data on the GPU memory (size == evaluation_input_size)
eval_output_gpu	Output data destination on the GPU memory (size == evaluation_output_size)

Implements gtsam_points::NonlinearFactorGPU.

◆ issue_linearize()

virtual void gtsam_points::IntegratedVGICPFactorGPU::issue_linearize	(	const void *	lin_input_cpu,
		const void *	lin_input_gpu,
		void *	lin_output_gpu
	)

overridevirtual

Issue linearization task.

Parameters

lin_input_cpu	Linearization input data on the CPU memory (size == linearization_input_size)
lin_input_gpu	Linearization input data on the GPU memory (size == linearization_input_size)
lin_output_gpu	Output data destination on the GPU memory (size == linearization_output_size)

Implements gtsam_points::NonlinearFactorGPU.

◆ linearization_input_size()

virtual size_t gtsam_points::IntegratedVGICPFactorGPU::linearization_input_size ( ) const

overridevirtual

Size of data to be uploaded to the GPU before linearization.

Returns: size_t

Implements gtsam_points::NonlinearFactorGPU.

◆ linearization_output_size()

virtual size_t gtsam_points::IntegratedVGICPFactorGPU::linearization_output_size ( ) const

overridevirtual

Size of data to be downloaded from the GPU after linearization.

Returns: size_t

Implements gtsam_points::NonlinearFactorGPU.

◆ memory_usage()

size_t gtsam_points::IntegratedVGICPFactorGPU::memory_usage ( ) const

Calculate the CPU memory usage of this factor.

Note: The result is approximate and does not account for objects not owned by this factor (e.g., point clouds)

Returns: Approximate CPU memory usage in bytes

◆ memory_usage_gpu()

size_t gtsam_points::IntegratedVGICPFactorGPU::memory_usage_gpu ( ) const

Calculate the GPU memory usage of this factor.

Note: The result is approximate and does not account for objects not owned by this factor (e.g., point clouds)

Returns: Approximate GPU memory usage in bytes

◆ num_inliers()

int gtsam_points::IntegratedVGICPFactorGPU::num_inliers ( ) const

Get the number of inlier points.

Note: This function must be called after the factor is linearized.

◆ set_enable_surface_validation()

void gtsam_points::IntegratedVGICPFactorGPU::set_enable_surface_validation ( bool enable )

Enable or disable surface orientation validation for correspondence search.

Note: To enable surface orientation validation, source frame must have point normals

◆ set_evaluation_point()

virtual void gtsam_points::IntegratedVGICPFactorGPU::set_evaluation_point	(	const gtsam::Values &	values,
		void *	eval_input_cpu
	)

overridevirtual

Write cost evaluation input data to the upload buffer.

Parameters

values	Current estimate
eval_input_cpu	Data upload buffer (size == evaluation_input_size)

Implements gtsam_points::NonlinearFactorGPU.

◆ set_linearization_point()

virtual void gtsam_points::IntegratedVGICPFactorGPU::set_linearization_point	(	const gtsam::Values &	values,
		void *	lin_input_cpu
	)

overridevirtual

Write linearization input data to the upload buffer.

Parameters

values	Current estimate
lin_input_cpu	Data upload buffer (size == linearization_input_size)

Implements gtsam_points::NonlinearFactorGPU.

◆ store_computed_error()

virtual void gtsam_points::IntegratedVGICPFactorGPU::store_computed_error ( const void * eval_output_cpu )

overridevirtual

Read cost evaluation output data from the download buffer.

Parameters

eval_output_cpu Data down load buffer (size == evaluation_output_size)

Implements gtsam_points::NonlinearFactorGPU.

◆ store_linearized()

virtual void gtsam_points::IntegratedVGICPFactorGPU::store_linearized ( const void * lin_output_cpu )

overridevirtual

Read linearization output data from the download buffer.

Parameters

lin_output_cpu Data download buffer (size == linearization_output_size)

Implements gtsam_points::NonlinearFactorGPU.

◆ sync()

virtual void gtsam_points::IntegratedVGICPFactorGPU::sync ( )

overridevirtual

Perform CPU-GPU synchronization and wait for the task.

Implements gtsam_points::NonlinearFactorGPU.

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

/home/runner/work/gtsam_points/gtsam_points/include/gtsam_points/factors/integrated_vgicp_factor_gpu.hpp

Public Types

Public Member Functions

Detailed Description

Constructor & Destructor Documentation

◆ IntegratedVGICPFactorGPU() [1/2]

◆ IntegratedVGICPFactorGPU() [2/2]

Member Function Documentation

◆ evaluation_input_size()

◆ evaluation_output_size()

◆ inlier_fraction()

◆ issue_compute_error()

◆ issue_linearize()

◆ linearization_input_size()

◆ linearization_output_size()

◆ memory_usage()

◆ memory_usage_gpu()

◆ num_inliers()

◆ set_enable_surface_validation()

◆ set_evaluation_point()

◆ set_linearization_point()

◆ store_computed_error()

◆ store_linearized()

◆ sync()