songsenand
/
cgal
mirror of https://github.com/CGAL/cgal
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

AABB tree: more on internal KD-tree used to accelerate the distance queries.

This commit is contained in:
Pierre Alliez 2009-05-07 10:51:01 +00:00
parent 0f225c01cf
commit db287f0e56
2 changed files with 47 additions and 24 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
AABB_tree/doc_tex/AABB_tree_ref/AABB_tree.tex
View File

@ -140,14 +140,14 @@ squared_distance(const Point& query,
\ccHeading{Accelerating the distance queries}
\ccMethod{void construct_search_tree();}
{ Constructs the internal search KD-tree used to accelerate the distance queries. The points in the search tree are taken from the primitives by calling the member function \ccc{reference_point} from the primitive.}
\ccMethod{bool construct_search_tree();}
{ Constructs the internal search KD-tree used to accelerate the distance queries. The points in the search tree are taken from the primitives by calling the member function \ccc{reference_point} from the primitive. Returns \ccc{true}, iff the memory allocation is successful.}
\begin{ccAdvanced}
\ccMethod{template <class InputIterator>
void construct_search_tree(InputIterator begin,
bool construct_search_tree(InputIterator begin,
InputIterator beyond);}
{ Constructs the internal search KD-tree used to accelerate the distance queries from a specified point set. Iterator \ccc{InputIterator} must have \ccc{Point_and_primitive} as value type. Each point from the specified \ccc{Point_and_primitive} set must be located on the corresponding input primitive. For a triangle surface mesh this point set can be provided, e.g., as the vertex points or as the triangle centroids. It is not required to provide one point per primitive, such that, e.g., large input primitive can be sampled with several sample points or conversely such that a single point is provided for clusters of small input primitives. }
{ Constructs the internal search KD-tree used to accelerate the distance queries from a specified point set. Iterator \ccc{InputIterator} must have \ccc{Point_and_primitive} as value type. Each point from the specified \ccc{Point_and_primitive} set must be located on the corresponding input primitive. For a triangle surface mesh this point set can be provided, e.g., as the vertex points or as the triangle centroids. It is not required to provide one point per primitive, such that, e.g., large input primitive can be sampled with several sample points or conversely such that a single point is provided for clusters of small input primitives. Returns \ccc{true}, iff the memory allocation is successful.}
\end{ccAdvanced}
\ccSeeAlso

63
AABB_tree/include/CGAL/AABB_tree.h
View File

@ -70,33 +70,37 @@ namespace CGAL {
bool rebuild(ConstPrimitiveIterator first, ConstPrimitiveIterator beyond);
/// Non virtual destructor
~AABB_tree() {
~AABB_tree()
{
clear();
if(m_search_tree_constructed)
delete(m_search_tree);
}
/// Clears the tree
void clear(void)
void clear()
{
// clear AABB tree
m_data.clear();
delete[] m_p_root;
delete [] m_p_root;
m_p_root = NULL;
m_search_tree_constructed = false;
clear_search_tree();
}
// bbox and size
Bounding_box bbox() const { return m_p_root->bbox(); }
Size_type size() const { return m_data.size(); }
bool empty() const { return m_data.empty(); }
/// Construct internal search tree with a given point set
// returns true iff successful memory allocation
template<typename ConstPointIterator>
void construct_search_tree(ConstPointIterator first, ConstPointIterator beyond);
bool construct_search_tree(ConstPointIterator first, ConstPointIterator beyond);
/// Construct internal search tree from
/// a point set taken on the internal primitives
void construct_search_tree(void);
// returns true iff successful memory allocation
bool construct_search_tree();
template<typename Query>
bool do_intersect(const Query& query) const;
@ -126,6 +130,17 @@ namespace CGAL {
Point_and_primitive closest_point_and_primitive(const Point& query, const Point& hint) const;
Point_and_primitive closest_point_and_primitive(const Point& query) const;
private:
void clear_search_tree()
{
// clear KD tree
delete m_p_search_tree;
m_p_search_tree = NULL;
m_search_tree_constructed = false;
}
/// generic traversal of the tree
template <class Query, class Traversal_traits>
void traversal(const Query& query, Traversal_traits& traits) const
@ -353,7 +368,7 @@ namespace CGAL {
// single root node
Node* m_p_root;
// search KD-tree
Search_tree* m_search_tree;
Search_tree* m_p_search_tree;
bool m_search_tree_constructed;
private:
@ -370,6 +385,7 @@ namespace CGAL {
ConstPrimitiveIterator beyond)
: m_data()
, m_p_root(NULL)
, m_p_search_tree(NULL)
, m_search_tree_constructed(false)
{
// Insert each primitive into tree
@ -398,7 +414,7 @@ namespace CGAL {
bool AABB_tree<Tr>::rebuild(ConstPrimitiveIterator first,
ConstPrimitiveIterator beyond)
{
// cleanup current tree
// cleanup current tree and internal KD tree
clear();
// inserts primitives
@ -425,17 +441,25 @@ namespace CGAL {
// constructs the search KD tree from given points
template<typename Tr>
template<typename ConstPointIterator>
void
AABB_tree<Tr>::construct_search_tree(ConstPointIterator first,
ConstPointIterator beyond)
bool AABB_tree<Tr>::construct_search_tree(ConstPointIterator first,
ConstPointIterator beyond)
{
m_search_tree = new Search_tree(first, beyond);
m_search_tree_constructed = true;
// clears current KD tree
clear_search_tree();
m_p_search_tree = new Search_tree(first, beyond);
if(m_p_search_tree != NULL)
{
m_search_tree_constructed = true;
return true;
}
else
return false;
}
// constructs the search KD tree from interal primitives
template<typename Tr>
void AABB_tree<Tr>::construct_search_tree(void)
bool AABB_tree<Tr>::construct_search_tree()
{
CGAL_assertion(!m_data.empty());
@ -443,10 +467,9 @@ namespace CGAL {
std::vector<Point> points;
typename std::vector<Primitive>::const_iterator it;
for(it = m_data.begin(); it != m_data.end(); ++it)
{
points.push_back(it->reference_point());
}
construct_search_tree(points.begin(), points.end());
return construct_search_tree(points.begin(), points.end());
}
template<typename Tr>
@ -550,7 +573,7 @@ namespace CGAL {
{
Point hint;
if(m_search_tree_constructed)
hint = m_search_tree->closest_point(query); // pick closest neighbor point as hint (fast)
hint = m_p_search_tree->closest_point(query); // pick closest neighbor point as hint (fast)
else
hint = m_data[0].reference_point(); // pick first primitive reference point as hint (slow)
return closest_point(query,hint);