[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#652103: ITP: nanoflann -- C++ header-only fork of the FLANN library for KD-trees



also sprach Jose Luis Blanco (University of Malaga) <joseluisblancoc@gmail.com> [2011.12.14.2059 +0100]:
> nanoflann is a C++ header-only library for building KD-Trees,
> mostly optimized for 2D or 3D point clouds.

I won't ask what "clouds" are, but please tell us about how the
2D/3D optimisation happens.

> This implementation is roughly one order of magnitude faster than
> some other popular KD-tree libraries.

What are other popular KD-tree libraries? And does this apply to
2D/3D only?

How does nanoflann compare to libkdtree++? Why would you want to use
one over the other?

Note: I am the author of libkdtree++, and I think choice is good.
However, I simply don't believe some of the claims made in the long
description, and I question the usefulness of 2D/3D optimisation wrt
KD-Trees: if you are dealing with only two or three dimensions,
KD-Trees are usually not what you want. Then again, that was just my
experience, my KD-Trees usually had dimensions in the low thousands…

-- 
 .''`.   martin f. krafft <madduck@d.o>      Related projects:
: :'  :  proud Debian developer               http://debiansystem.info
`. `'`   http://people.debian.org/~madduck    http://vcs-pkg.org
  `-  Debian - when you have better things to do than fixing systems
 
philosophy: unintelligible answers to insoluble problems.

Attachment: digital_signature_gpg.asc
Description: Digital signature (see http://martin-krafft.net/gpg/sig-policy/999bbcc4/current)


Reply to: