FREAK descriptor: scaleNormalized parameter [closed]

asked 2012-08-08 09:17:03 -0500

Arno gravatar image

Hi all,

I've been playing around with the FREAK keypoint descriptor. One of the parameters used by the FREAK constructor, is scaleNormalized.

When using the default scaleNormalized of 22.0f, many of the keypoints don't get any descriptor assigned. When I use a smaller value for this parameter, e.g. 8.0f, I get much more descriptors.

I have looked at the FREAK source code from the OpenCV repository, but honestly, I have no idea what this parameter means...

Does anybody has any idea about the purpose of the scaleNormalized parameter?

Kind regards, Arno

edit retag flag offensive reopen merge delete

Closed for the following reason the question is answered, right answer was accepted by sturkmen
close date 2017-09-04 16:10:12.192268

Comments

1

Here is the original paper on FREAK http://infoscience.epfl.ch/record/175537/files/2069.pdf I think it will help you understand how it works, and what each parameter does.

sammy gravatar imagesammy ( 2012-08-08 09:24:40 -0500 )edit

I have the same problem as you and I am ging to mail the author of the codes to see if I can learn more. I have been playing with different scales of the training image (with cv::pyrDown(source, dest)) and the scaleNormalized parameter. I think it is very sensitive and dependent on the training image, its size and the number of keypoints.

jav_rock gravatar imagejav_rock ( 2012-12-17 13:30:36 -0500 )edit