Does anybody know what algorithm is used here?
I want to implement this function to do detection's windows grouping.
Thank you.
2
-
1OpenCV is open source. Why not have a look?Roger Rowland– Roger Rowland2015-07-08 10:31:20 +00:00Commented Jul 8, 2015 at 10:31
-
@RogerRowland yes, I know :) But it would be great to know what algorithm do they use first. What I want to do is to implement this thing in a different language and first I want to know the name of the algorithm (maybe there are libraries that might have it implemented) before porting it to another language.warmspringwinds– warmspringwinds2015-07-08 10:34:35 +00:00Commented Jul 8, 2015 at 10:34
Add a comment
|
1 Answer
If you look at the OpenCV source code for the partition function, you will see the following comments:
// This function splits the input sequence or set into one or more equivalence classes and
// returns the vector of labels - 0-based class indexes for each element.
// predicate(a,b) returns true if the two sequence elements certainly belong to the same class.
//
// The algorithm is described in "Introduction to Algorithms"
// by Cormen, Leiserson and Rivest, the chapter "Data structures for disjoint sets"
template<typename _Tp, class _EqPredicate> int partition( const vector<_Tp>& _vec, vector<int>& labels, _EqPredicate predicate=_EqPredicate())
{
// ... etc.
}
This gives you both the source code, and the reference for the algorithm.
So, that's Chapter 21 in this book.
3 Comments
warmspringwinds
Great! Thank you, Roger. Could you, please, refer me where did you find this lines? Because I found only this github.com/Itseez/opencv/blob/…
Roger Rowland
@warmspringwinds Maybe they removed the comment in later sources - I have 2.4.8 and the body of that function looks the same to me. For example, you can see it here
warmspringwinds
Thank you :) Now it's much easier for me with this information.