File Exchange

image thumbnail

Fast Gaussian Transform mex implementation

version 1.2.0.0 (80.5 KB) by Sebastien PARIS
Mex inplementation of the FGT transform

3 Downloads

Updated 28 Jan 2020

View License

Here is another fast implementation of the FGT transform.

Reference :
http://www.cs.ubc.ca/~awll/nbody_methods.html

Cite As

Sebastien PARIS (2020). Fast Gaussian Transform mex implementation (https://www.mathworks.com/matlabcentral/fileexchange/17438-fast-gaussian-transform-mex-implementation), MATLAB Central File Exchange. Retrieved .

Comments and Ratings (5)

James Heald

In example 2 that you give, your data is drawn from a multivariate gaussian with covariance R. You specify the bandwidth as h = 1, yet you pass the data in it's raw form. Do you not need to whiten the data first if you are going to specify the bandwidth as 1?

Abhi..

Hi, can you please tell me how to use this code to compute the information potential of an image.

H.Z Dai

Sebastien PARIS

If you want, you can cite this URL :
http://www.lsis.org/fiche.php?nom=sebastien_paris&afficher=&presentation_publis=&processus=ok

I'll plan to upload lot of my matlab works inside (maybe more than 200 functions/toolboxes)

Regards

Andriy M

Excellent job! The code is nicely written and commented. FGT is very useful in many application and having its mex implementation in Matlab helps a lot. Do you have a code web page or something to cite (in case somebody use you code)?

Updates

1.2.0.0

Fixed for modern Matlab & OS64

1.1.0.0

-Add mexme_fgt
-Minor changes

MATLAB Release Compatibility
Created with R2016b
Compatible with any release
Platform Compatibility
Windows macOS Linux