|
Point Cloud Library (PCL)
1.6.0
|
|
Point Cloud Library (PCL)
1.6.0
|
GeneralizedIterativeClosestPoint is an ICP variant that implements the generalized iterative closest point algorithm as described by Alex Segal et al. More...
#include <pcl/registration/gicp.h>
Classes | |
| struct | OptimizationFunctorWithIndices |
| optimization functor structure | |
Public Types | |
| typedef boost::shared_ptr < Registration< PointSource, PointTarget > > | Ptr |
| typedef boost::shared_ptr < const Registration < PointSource, PointTarget > > | ConstPtr |
| typedef pcl::KdTree< PointTarget > | KdTree |
| typedef pcl::KdTree < PointTarget >::Ptr | KdTreePtr |
| typedef KdTree::PointRepresentationConstPtr | PointRepresentationConstPtr |
| typedef pcl::registration::TransformationEstimation < PointSource, PointTarget > | TransformationEstimation |
| typedef TransformationEstimation::Ptr | TransformationEstimationPtr |
| typedef TransformationEstimation::ConstPtr | TransformationEstimationConstPtr |
| typedef pcl::PointCloud < PointSource > | PointCloud |
| typedef PointCloud::Ptr | PointCloudPtr |
| typedef PointCloud::ConstPtr | PointCloudConstPtr |
Public Member Functions | |
| GeneralizedIterativeClosestPoint () | |
| Empty constructor. | |
| void | setInputCloud (const PointCloudSourceConstPtr &cloud) |
| Provide a pointer to the input dataset. | |
| void | setInputTarget (const PointCloudTargetConstPtr &target) |
| Provide a pointer to the input target (e.g., the point cloud that we want to align the input source to) | |
| void | estimateRigidTransformationBFGS (const PointCloudSource &cloud_src, const std::vector< int > &indices_src, const PointCloudTarget &cloud_tgt, const std::vector< int > &indices_tgt, Eigen::Matrix4f &transformation_matrix) |
| Estimate a rigid rotation transformation between a source and a target point cloud using an iterative non-linear Levenberg-Marquardt approach. | |
| const Eigen::Matrix3d & | mahalanobis (size_t index) const |
| void | computeRDerivative (const Vector6d &x, const Eigen::Matrix3d &R, Vector6d &g) const |
| Computes rotation matrix derivative. | |
| void | setRotationEpsilon (double epsilon) |
| Set the rotation epsilon (maximum allowable difference between two consecutive rotations) in order for an optimization to be considered as having converged to the final solution. | |
| double | getRotationEpsilon () |
| Get the rotation epsilon (maximum allowable difference between two consecutive rotations) as set by the user. | |
| void | setCorrespondenceRandomness (int k) |
| Set the number of neighbors used when selecting a point neighbourhood to compute covariances. | |
| int | getCorrespondenceRandomness () |
| Get the number of neighbors used when computing covariances as set by the user. | |
| void | setMaximumOptimizerIterations (int max) |
| set maximum number of iterations at the optimization step | |
| int | getMaximumOptimizerIterations () |
| void | setTransformationEstimation (const TransformationEstimationPtr &te) |
| Provide a pointer to the transformation estimation object. | |
| virtual void | setInputTarget (const PointCloudTargetConstPtr &cloud) |
| Provide a pointer to the input target (e.g., the point cloud that we want to align the input source to) | |
| PointCloudTargetConstPtr const | getInputTarget () |
| Get a pointer to the input point cloud dataset target. | |
| Eigen::Matrix4f | getFinalTransformation () |
| Get the final transformation matrix estimated by the registration method. | |
| Eigen::Matrix4f | getLastIncrementalTransformation () |
| Get the last incremental transformation matrix estimated by the registration method. | |
| void | setMaximumIterations (int nr_iterations) |
| Set the maximum number of iterations the internal optimization should run for. | |
| int | getMaximumIterations () |
| Get the maximum number of iterations the internal optimization should run for, as set by the user. | |
| void | setRANSACIterations (int ransac_iterations) |
| Set the number of iterations RANSAC should run for. | |
| double | getRANSACIterations () |
| Get the number of iterations RANSAC should run for, as set by the user. | |
| void | setRANSACOutlierRejectionThreshold (double inlier_threshold) |
| Set the inlier distance threshold for the internal RANSAC outlier rejection loop. | |
| double | getRANSACOutlierRejectionThreshold () |
| Get the inlier distance threshold for the internal outlier rejection loop as set by the user. | |
| void | setMaxCorrespondenceDistance (double distance_threshold) |
| Set the maximum distance threshold between two correspondent points in source <-> target. | |
| double | getMaxCorrespondenceDistance () |
| Get the maximum distance threshold between two correspondent points in source <-> target. | |
| void | setTransformationEpsilon (double epsilon) |
| Set the transformation epsilon (maximum allowable difference between two consecutive transformations) in order for an optimization to be considered as having converged to the final solution. | |
| double | getTransformationEpsilon () |
| Get the transformation epsilon (maximum allowable difference between two consecutive transformations) as set by the user. | |
| void | setEuclideanFitnessEpsilon (double epsilon) |
| Set the maximum allowed Euclidean error between two consecutive steps in the ICP loop, before the algorithm is considered to have converged. | |
| double | getEuclideanFitnessEpsilon () |
| Get the maximum allowed distance error before the algorithm will be considered to have converged, as set by the user. | |
| void | setPointRepresentation (const PointRepresentationConstPtr &point_representation) |
| Provide a boost shared pointer to the PointRepresentation to be used when comparing points. | |
| template<typename FunctionSignature > | |
| bool | registerVisualizationCallback (boost::function< FunctionSignature > &visualizerCallback) |
| Register the user callback function which will be called from registration thread in order to update point cloud obtained after each iteration. | |
| double | getFitnessScore (double max_range=std::numeric_limits< double >::max()) |
| Obtain the Euclidean fitness score (e.g., sum of squared distances from the source to the target) | |
| double | getFitnessScore (const std::vector< float > &distances_a, const std::vector< float > &distances_b) |
| Obtain the Euclidean fitness score (e.g., sum of squared distances from the source to the target) from two sets of correspondence distances (distances between source and target points) | |
| bool | hasConverged () |
| Return the state of convergence after the last align run. | |
| void | align (PointCloudSource &output) |
| Call the registration algorithm which estimates the transformation and returns the transformed source (input) as output. | |
| void | align (PointCloudSource &output, const Eigen::Matrix4f &guess) |
| Call the registration algorithm which estimates the transformation and returns the transformed source (input) as output. | |
| const std::string & | getClassName () const |
| Abstract class get name method. | |
| virtual void | setInputCloud (const PointCloudConstPtr &cloud) |
| Provide a pointer to the input dataset. | |
| PointCloudConstPtr const | getInputCloud () |
| Get a pointer to the input point cloud dataset. | |
| void | setIndices (const IndicesPtr &indices) |
| Provide a pointer to the vector of indices that represents the input data. | |
| void | setIndices (const IndicesConstPtr &indices) |
| Provide a pointer to the vector of indices that represents the input data. | |
| void | setIndices (const PointIndicesConstPtr &indices) |
| Provide a pointer to the vector of indices that represents the input data. | |
| void | setIndices (size_t row_start, size_t col_start, size_t nb_rows, size_t nb_cols) |
| Set the indices for the points laying within an interest region of the point cloud. | |
| IndicesPtr const | getIndices () |
| Get a pointer to the vector of indices used. | |
| const PointSource & | operator[] (size_t pos) |
| Override PointCloud operator[] to shorten code. | |
GeneralizedIterativeClosestPoint is an ICP variant that implements the generalized iterative closest point algorithm as described by Alex Segal et al.
in http://www.stanford.edu/~avsegal/resources/papers/Generalized_ICP.pdf The approach is based on using anistropic cost functions to optimize the alignment after closest point assignments have been made. The original code uses GSL and ANN while in ours we use an eigen mapped BFGS and FLANN.
typedef boost::shared_ptr< const Registration<PointSource, PointTarget> > pcl::Registration< PointSource, PointTarget >::ConstPtr [inherited] |
Definition at line 70 of file registration.h.
typedef pcl::KdTree<PointTarget> pcl::Registration< PointSource, PointTarget >::KdTree [inherited] |
Definition at line 72 of file registration.h.
typedef pcl::KdTree<PointTarget>::Ptr pcl::Registration< PointSource, PointTarget >::KdTreePtr [inherited] |
Definition at line 73 of file registration.h.
typedef pcl::PointCloud<PointSource > pcl::PCLBase< PointSource >::PointCloud [inherited] |
Definition at line 74 of file pcl_base.h.
typedef PointCloud::ConstPtr pcl::PCLBase< PointSource >::PointCloudConstPtr [inherited] |
Definition at line 76 of file pcl_base.h.
typedef PointCloud::Ptr pcl::PCLBase< PointSource >::PointCloudPtr [inherited] |
Definition at line 75 of file pcl_base.h.
typedef KdTree::PointRepresentationConstPtr pcl::Registration< PointSource, PointTarget >::PointRepresentationConstPtr [inherited] |
Definition at line 83 of file registration.h.
typedef boost::shared_ptr< Registration<PointSource, PointTarget> > pcl::Registration< PointSource, PointTarget >::Ptr [inherited] |
Definition at line 69 of file registration.h.
typedef pcl::registration::TransformationEstimation<PointSource, PointTarget> pcl::Registration< PointSource, PointTarget >::TransformationEstimation [inherited] |
Definition at line 85 of file registration.h.
typedef TransformationEstimation::ConstPtr pcl::Registration< PointSource, PointTarget >::TransformationEstimationConstPtr [inherited] |
Definition at line 87 of file registration.h.
typedef TransformationEstimation::Ptr pcl::Registration< PointSource, PointTarget >::TransformationEstimationPtr [inherited] |
Definition at line 86 of file registration.h.
| pcl::GeneralizedIterativeClosestPoint< PointSource, PointTarget >::GeneralizedIterativeClosestPoint | ( | ) | [inline] |
| void pcl::Registration< PointSource, PointTarget >::align | ( | PointCloudSource & | output | ) | [inline, inherited] |
Call the registration algorithm which estimates the transformation and returns the transformed source (input) as output.
| output | the resultant input transfomed point cloud dataset |
Definition at line 113 of file registration.hpp.
| void pcl::Registration< PointSource, PointTarget >::align | ( | PointCloudSource & | output, |
| const Eigen::Matrix4f & | guess | ||
| ) | [inline, inherited] |
Call the registration algorithm which estimates the transformation and returns the transformed source (input) as output.
| output | the resultant input transfomed point cloud dataset |
| guess | the initial gross estimation of the transformation |
Definition at line 120 of file registration.hpp.
| void pcl::GeneralizedIterativeClosestPoint< PointSource, PointTarget >::computeRDerivative | ( | const Vector6d & | x, |
| const Eigen::Matrix3d & | R, | ||
| Vector6d & | g | ||
| ) | const |
Computes rotation matrix derivative.
rotation matrix is obtainded from rotation angles x[3], x[4] and x[5]
| void pcl::GeneralizedIterativeClosestPoint< PointSource, PointTarget >::estimateRigidTransformationBFGS | ( | const PointCloudSource & | cloud_src, |
| const std::vector< int > & | indices_src, | ||
| const PointCloudTarget & | cloud_tgt, | ||
| const std::vector< int > & | indices_tgt, | ||
| Eigen::Matrix4f & | transformation_matrix | ||
| ) |
Estimate a rigid rotation transformation between a source and a target point cloud using an iterative non-linear Levenberg-Marquardt approach.
| [in] | cloud_src | the source point cloud dataset |
| [in] | indices_src | the vector of indices describing the points of interest in cloud_src |
| [in] | cloud_tgt | the target point cloud dataset |
| [in] | indices_tgt | the vector of indices describing the correspondences of the interst points from indices_src |
| [out] | transformation_matrix | the resultant transformation matrix |
| const std::string& pcl::Registration< PointSource, PointTarget >::getClassName | ( | ) | const [inline, inherited] |
Abstract class get name method.
Definition at line 280 of file registration.h.
| int pcl::GeneralizedIterativeClosestPoint< PointSource, PointTarget >::getCorrespondenceRandomness | ( | ) | [inline] |
| double pcl::Registration< PointSource, PointTarget >::getEuclideanFitnessEpsilon | ( | ) | [inline, inherited] |
Get the maximum allowed distance error before the algorithm will be considered to have converged, as set by the user.
See setEuclideanFitnessEpsilon
Definition at line 217 of file registration.h.
| Eigen::Matrix4f pcl::Registration< PointSource, PointTarget >::getFinalTransformation | ( | ) | [inline, inherited] |
Get the final transformation matrix estimated by the registration method.
Definition at line 133 of file registration.h.
| double pcl::Registration< PointSource, PointTarget >::getFitnessScore | ( | double | max_range = std::numeric_limits<double>::max () | ) | [inline, inherited] |
Obtain the Euclidean fitness score (e.g., sum of squared distances from the source to the target)
| max_range | maximum allowable distance between a point and its correspondence in the target (default: double::max) |
Definition at line 72 of file registration.hpp.
| double pcl::Registration< PointSource, PointTarget >::getFitnessScore | ( | const std::vector< float > & | distances_a, |
| const std::vector< float > & | distances_b | ||
| ) | [inline, inherited] |
Obtain the Euclidean fitness score (e.g., sum of squared distances from the source to the target) from two sets of correspondence distances (distances between source and target points)
| [in] | distances_a | the first set of distances between correspondences |
| [in] | distances_b | the second set of distances between correspondences |
Definition at line 61 of file registration.hpp.
| IndicesPtr const pcl::PCLBase< PointSource >::getIndices | ( | ) | [inline, inherited] |
Get a pointer to the vector of indices used.
Definition at line 190 of file pcl_base.h.
| PointCloudConstPtr const pcl::PCLBase< PointSource >::getInputCloud | ( | ) | [inline, inherited] |
Get a pointer to the input point cloud dataset.
Definition at line 107 of file pcl_base.h.
| PointCloudTargetConstPtr const pcl::Registration< PointSource, PointTarget >::getInputTarget | ( | ) | [inline, inherited] |
Get a pointer to the input point cloud dataset target.
Definition at line 129 of file registration.h.
| Eigen::Matrix4f pcl::Registration< PointSource, PointTarget >::getLastIncrementalTransformation | ( | ) | [inline, inherited] |
Get the last incremental transformation matrix estimated by the registration method.
Definition at line 137 of file registration.h.
| double pcl::Registration< PointSource, PointTarget >::getMaxCorrespondenceDistance | ( | ) | [inline, inherited] |
Get the maximum distance threshold between two correspondent points in source <-> target.
If the distance is larger than this threshold, the points will be ignored in the alignment process.
Definition at line 185 of file registration.h.
| int pcl::Registration< PointSource, PointTarget >::getMaximumIterations | ( | ) | [inline, inherited] |
Get the maximum number of iterations the internal optimization should run for, as set by the user.
Definition at line 147 of file registration.h.
| int pcl::GeneralizedIterativeClosestPoint< PointSource, PointTarget >::getMaximumOptimizerIterations | ( | ) | [inline] |
| double pcl::Registration< PointSource, PointTarget >::getRANSACIterations | ( | ) | [inline, inherited] |
Get the number of iterations RANSAC should run for, as set by the user.
Definition at line 157 of file registration.h.
| double pcl::Registration< PointSource, PointTarget >::getRANSACOutlierRejectionThreshold | ( | ) | [inline, inherited] |
Get the inlier distance threshold for the internal outlier rejection loop as set by the user.
Definition at line 171 of file registration.h.
| double pcl::GeneralizedIterativeClosestPoint< PointSource, PointTarget >::getRotationEpsilon | ( | ) | [inline] |
| double pcl::Registration< PointSource, PointTarget >::getTransformationEpsilon | ( | ) | [inline, inherited] |
Get the transformation epsilon (maximum allowable difference between two consecutive transformations) as set by the user.
Definition at line 200 of file registration.h.
| bool pcl::Registration< PointSource, PointTarget >::hasConverged | ( | ) | [inline, inherited] |
Return the state of convergence after the last align run.
Definition at line 261 of file registration.h.
| const Eigen::Matrix3d& pcl::GeneralizedIterativeClosestPoint< PointSource, PointTarget >::mahalanobis | ( | size_t | index | ) | const [inline] |
| const PointSource & pcl::PCLBase< PointSource >::operator[] | ( | size_t | pos | ) | [inline, inherited] |
Override PointCloud operator[] to shorten code.
| pos | position in indices_ vector |
Definition at line 197 of file pcl_base.h.
| bool pcl::Registration< PointSource, PointTarget >::registerVisualizationCallback | ( | boost::function< FunctionSignature > & | visualizerCallback | ) | [inline, inherited] |
Register the user callback function which will be called from registration thread in order to update point cloud obtained after each iteration.
| [in] | visualizerCallback | reference of the user callback function |
Definition at line 233 of file registration.h.
| void pcl::GeneralizedIterativeClosestPoint< PointSource, PointTarget >::setCorrespondenceRandomness | ( | int | k | ) | [inline] |
Set the number of neighbors used when selecting a point neighbourhood to compute covariances.
A higher value will bring more accurate covariance matrix but will make covariances computation slower.
| k | the number of neighbors to use when computing covariances |
| void pcl::Registration< PointSource, PointTarget >::setEuclideanFitnessEpsilon | ( | double | epsilon | ) | [inline, inherited] |
Set the maximum allowed Euclidean error between two consecutive steps in the ICP loop, before the algorithm is considered to have converged.
The error is estimated as the sum of the differences between correspondences in an Euclidean sense, divided by the number of correspondences.
| epsilon | the maximum allowed distance error before the algorithm will be considered to have converged |
Definition at line 211 of file registration.h.
| void pcl::PCLBase< PointSource >::setIndices | ( | const IndicesPtr & | indices | ) | [inline, inherited] |
Provide a pointer to the vector of indices that represents the input data.
| indices | a pointer to the vector of indices that represents the input data. |
Definition at line 113 of file pcl_base.h.
| void pcl::PCLBase< PointSource >::setIndices | ( | const IndicesConstPtr & | indices | ) | [inline, inherited] |
Provide a pointer to the vector of indices that represents the input data.
| indices | a pointer to the vector of indices that represents the input data. |
Definition at line 124 of file pcl_base.h.
| void pcl::PCLBase< PointSource >::setIndices | ( | const PointIndicesConstPtr & | indices | ) | [inline, inherited] |
Provide a pointer to the vector of indices that represents the input data.
| indices | a pointer to the vector of indices that represents the input data. |
Definition at line 135 of file pcl_base.h.
| void pcl::PCLBase< PointSource >::setIndices | ( | size_t | row_start, |
| size_t | col_start, | ||
| size_t | nb_rows, | ||
| size_t | nb_cols | ||
| ) | [inline, inherited] |
Set the indices for the points laying within an interest region of the point cloud.
| row_start | the offset on rows |
| col_start | the offset on columns |
| nb_rows | the number of rows to be considered row_start included |
| nb_cols | the number of columns to be considered col_start included |
Definition at line 151 of file pcl_base.h.
| virtual void pcl::PCLBase< PointSource >::setInputCloud | ( | const PointCloudConstPtr & | cloud | ) | [inline, virtual, inherited] |
Provide a pointer to the input dataset.
| cloud | the const boost shared pointer to a PointCloud message |
Definition at line 103 of file pcl_base.h.
| void pcl::GeneralizedIterativeClosestPoint< PointSource, PointTarget >::setInputCloud | ( | const PointCloudSourceConstPtr & | cloud | ) | [inline] |
Provide a pointer to the input dataset.
| cloud | the const boost shared pointer to a PointCloud message |
| void pcl::Registration< PointSource, PointTarget >::setInputTarget | ( | const PointCloudTargetConstPtr & | cloud | ) | [inline, virtual, inherited] |
Provide a pointer to the input target (e.g., the point cloud that we want to align the input source to)
| cloud | the input point cloud target |
Definition at line 42 of file registration.hpp.
| void pcl::GeneralizedIterativeClosestPoint< PointSource, PointTarget >::setInputTarget | ( | const PointCloudTargetConstPtr & | target | ) | [inline] |
| void pcl::Registration< PointSource, PointTarget >::setMaxCorrespondenceDistance | ( | double | distance_threshold | ) | [inline, inherited] |
Set the maximum distance threshold between two correspondent points in source <-> target.
If the distance is larger than this threshold, the points will be ignored in the alignment process.
| distance_threshold | the maximum distance threshold between a point and its nearest neighbor correspondent in order to be considered in the alignment process |
Definition at line 179 of file registration.h.
| void pcl::Registration< PointSource, PointTarget >::setMaximumIterations | ( | int | nr_iterations | ) | [inline, inherited] |
Set the maximum number of iterations the internal optimization should run for.
| nr_iterations | the maximum number of iterations the internal optimization should run for |
Definition at line 143 of file registration.h.
| void pcl::GeneralizedIterativeClosestPoint< PointSource, PointTarget >::setMaximumOptimizerIterations | ( | int | max | ) | [inline] |
| void pcl::Registration< PointSource, PointTarget >::setPointRepresentation | ( | const PointRepresentationConstPtr & | point_representation | ) | [inline, inherited] |
Provide a boost shared pointer to the PointRepresentation to be used when comparing points.
| point_representation | the PointRepresentation to be used by the k-D tree |
Definition at line 223 of file registration.h.
| void pcl::Registration< PointSource, PointTarget >::setRANSACIterations | ( | int | ransac_iterations | ) | [inline, inherited] |
Set the number of iterations RANSAC should run for.
| ransac_iterations | is the number of iterations RANSAC should run for |
Definition at line 153 of file registration.h.
| void pcl::Registration< PointSource, PointTarget >::setRANSACOutlierRejectionThreshold | ( | double | inlier_threshold | ) | [inline, inherited] |
Set the inlier distance threshold for the internal RANSAC outlier rejection loop.
The method considers a point to be an inlier, if the distance between the target data index and the transformed source index is smaller than the given inlier distance threshold. The value is set by default to 0.05m.
| inlier_threshold | the inlier distance threshold for the internal RANSAC outlier rejection loop |
Definition at line 167 of file registration.h.
| void pcl::GeneralizedIterativeClosestPoint< PointSource, PointTarget >::setRotationEpsilon | ( | double | epsilon | ) | [inline] |
| void pcl::Registration< PointSource, PointTarget >::setTransformationEpsilon | ( | double | epsilon | ) | [inline, inherited] |
Set the transformation epsilon (maximum allowable difference between two consecutive transformations) in order for an optimization to be considered as having converged to the final solution.
| epsilon | the transformation epsilon in order for an optimization to be considered as having converged to the final solution. |
Definition at line 194 of file registration.h.
| void pcl::Registration< PointSource, PointTarget >::setTransformationEstimation | ( | const TransformationEstimationPtr & | te | ) | [inline, inherited] |
Provide a pointer to the transformation estimation object.
(e.g., SVD, point to plane etc.)
| te | is the pointer to the corresponding transformation estimation object |
Definition at line 119 of file registration.h.
1.7.6.1
