eos::fitting::KDTreeVectorOfVectorsAdaptor< VectorOfVectorsType, num_t, DIM, Distance, IndexType > Struct Template Reference

#include <closest_edge_fitting.hpp>

Public Types
typedef KDTreeVectorOfVectorsAdaptor< VectorOfVectorsType, num_t, DIM, Distance >	self_t
typedef Distance::template traits< num_t, self_t >::distance_t	metric_t
typedef nanoflann::KDTreeSingleIndexAdaptor< metric_t, self_t, DIM, IndexType >	index_t

Public Member Functions
	KDTreeVectorOfVectorsAdaptor (const int dimensionality, const VectorOfVectorsType &mat, const int leaf_max_size=10)
	The kd-tree index for the user to call its methods as usual with any other FLANN index.
void	query (const num_t *query_point, const size_t num_closest, IndexType *out_indices, num_t *out_distances_sq, const int nChecks_IGNORED=10) const
Interface expected by KDTreeSingleIndexAdaptor
const self_t &	derived () const
self_t &	derived ()
size_t	kdtree_get_point_count () const
num_t	kdtree_distance (const num_t *p1, const size_t idx_p2, size_t size) const
num_t	kdtree_get_pt (const size_t idx, int dim) const
template<class BBOX >
bool	kdtree_get_bbox (BBOX &) const

Public Attributes
index_t *	index
const VectorOfVectorsType &	m_data

Detailed Description

template<class VectorOfVectorsType, typename num_t = double, int DIM = -1, class Distance = nanoflann::metric_L2, typename IndexType = size_t>
struct eos::fitting::KDTreeVectorOfVectorsAdaptor< VectorOfVectorsType, num_t, DIM, Distance, IndexType >

A simple vector-of-vectors adaptor for nanoflann, without duplicating the storage. The i'th vector represents a point in the state space.

This adaptor is from the nanoflann examples and shows how to use it with these types of containers: typedef std::vector<std::vector<double> > my_vector_of_vectors_t; typedef std::vector<Eigen::VectorXd> my_vector_of_vectors_t;

It works with any type inside the vector that has operator[] defined to access its elements, as well as a ::size() operator to return its number of dimensions. Eigen::VectorX is one of them. cv::Point is not (no [] and no size()), glm is also not (no size()).

Template Parameters

DIM	If set to >0, it specifies a compile-time fixed dimensionality for the points in the data set, allowing more compiler optimizations.
num_t	The type of the point coordinates (typically, double or float).
Distance	The distance metric to use: nanoflann::metric_L1, nanoflann::metric_L2, nanoflann::metric_L2_Simple, etc.
IndexType	The type for indices in the KD-tree index (typically, size_t of int)

Constructor & Destructor Documentation

KDTreeVectorOfVectorsAdaptor()

template<class VectorOfVectorsType , typename num_t = double, int DIM = -1, class Distance = nanoflann::metric_L2, typename IndexType = size_t>

eos::fitting::KDTreeVectorOfVectorsAdaptor< VectorOfVectorsType, num_t, DIM, Distance, IndexType >::KDTreeVectorOfVectorsAdaptor ( const int  dimensionality, const VectorOfVectorsType &  mat, const int  leaf_max_size = 10  )

inline

The kd-tree index for the user to call its methods as usual with any other FLANN index.

Constructor: takes a const ref to the vector of vectors object with the data points

Member Function Documentation

query()

template<class VectorOfVectorsType , typename num_t = double, int DIM = -1, class Distance = nanoflann::metric_L2, typename IndexType = size_t>

void eos::fitting::KDTreeVectorOfVectorsAdaptor< VectorOfVectorsType, num_t, DIM, Distance, IndexType >::query ( const num_t *  query_point, const size_t  num_closest, IndexType *  out_indices, num_t *  out_distances_sq, const int  nChecks_IGNORED = 10  ) const

inline

Query for the num_closest closest points to a given point (entered as query_point[0:dim-1]). Note that this is a short-cut method for index->findNeighbors(). The user can also call index->... methods as desired.

Note

nChecks_IGNORED is ignored but kept for compatibility with the original FLANN interface.

---

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

• include/eos/fitting/closest_edge_fitting.hpp