A K-Means++ Clustering Implementation for VTK

Please use this identifier to cite or link to this publication: http://hdl.handle.net/10380/3220
K-Means clustering is an excellent technique for clustering points when the number of clusters is known. We present a implementation (vtkKMeanClustering) of the algorithm written in a VTK context. We also implement the K-Means++ initialization method which finds the global optimum much more frequently than a naive/random initialization.

The code is currently hosted at http://github.com/daviddoria/KMeansClustering .

minus Excellent Contribution! by Arnaud Gelas on 2010-09-28 13:39:32 for revision #1
starstarstarstarstar expertise: 5 sensitivity: 5

A Must Have!!!

Free comment :
Once again: excellent work!

I have to admit that I have not read the paper yet, but I had a quick look to the code...

Using a kd-tree to find closest points will significantly speed up the implementation 

whenever the number of clusters is quite large.

I would also recommend that you abstract the metric and the way to compute the centroid, 

like that if someone wants to use L_1 or another metric (for example which makes use of normals) 

he won't need to duplicate too much code.




Add a new review
Quick Comments

Download All
Download Paper , View Paper

Statistics more
Global rating: starstarstarstarstar
Review rating: starstarstarstarstar [review]
Paper Quality: plus minus

Information more
Categories: Iterative clustering, PointSet
Keywords: clustering, Kmeans, kmeans++
Export citation:


Linked Publications more
Importing Contours from DICOM-RT Structure Sets Importing Contours from DICOM-RT Structure Sets
by Dowling J., Malaterre M., Greer P.B., Salvado O.
Providing values of adjacent voxel with vtkDiscreteMarchingCubes Providing values of adjacent voxel with vtkDiscreteMarchingCubes
by Grothausmann R.

View license
Loading license...

Send a message to the author
ISSN 2328-3459
Powered by Midas