This class builds a point cloud graph representing an input dataset by using nearest neighbor search. More...

#include <VisionX/libraries/PointCloudGraph/nearest_neighbors_graph_builder.h>

Inheritance diagram for NearestNeighborsGraphBuilder< PointT, GraphT >:

Public Types
typedef pcl::search::Search< PointOutT >	Search

typedef Search::Ptr	SearchPtr

Public Types inherited from GraphBuilder< PointT, GraphT >
typedef PointT	PointInT
	Type of points in the input cloud. More...

typedef point_cloud_graph_traits< GraphT >::point_type	PointOutT
	Type of points in the output graph. More...

typedef boost::shared_ptr< GraphBuilder< PointT, GraphT > >	Ptr

typedef boost::graph_traits< GraphT >::vertex_descriptor	VertexId

Public Member Functions
virtual void	compute (GraphT &graph)
	Build a graph based on the provided input data. More...

size_t	getNumberOfNeighbors () const
	Returns the number of neighbors to find when building a graph. More...

SearchPtr	getSearchMethod () const
	Get the search method used for finding nearest neighbors when building a graph. More...

	NearestNeighborsGraphBuilder (size_t num_neighbors=14)
	Constructor. More...

void	setNumberOfNeighbors (size_t num_neighbors)
	Set the number of neighbors to find when building a graph. More...

void	setSearchMethod (const SearchPtr &search)
	Set search method that will be used for finding K nearest neighbors when building a graph. More...

Public Member Functions inherited from GraphBuilder< PointT, GraphT >
const std::vector< VertexId > &	getPointToVertexMap () const
	Get a mapping between points in the input cloud and the vertices in the output graph. More...

Additional Inherited Members
Protected Attributes inherited from GraphBuilder< PointT, GraphT >
std::vector< VertexId >	point_to_vertex_map_

Detailed Description

template<typename PointT, typename GraphT>
class pcl::graph::NearestNeighborsGraphBuilder< PointT, GraphT >

This class builds a point cloud graph representing an input dataset by using nearest neighbor search.

The points from the input cloud become vertices and the edges are established between each point and its neighbors (as found by the search object provided with setSearchMethod()). The user may choose to use the default search method, which will be either KdTree or OrganizedNeighbor depending on whether the input point cloud is organized or not.

The data contained in the points of the input cloud will be copied inside the vertices of the newly created graph. Note that the points in the input cloud and the output graph may have different types, in which case only the intersecting fields will be copied over.

For additional information see documentation for GraphBuilder.