Ask Your Question

HOGDescriptor derivAperture parameter

asked 2012-08-10 13:01:55 -0500

TRiBi gravatar image

Hi there!

I have search in the OpenCV documentation and I haven't found any clue about derivAperture parameter in the HOGDescriptor function.

Anyone knows what is this parameter for?

Thank you in advance.

edit retag flag offensive close merge delete

2 answers

Sort by » oldest newest most voted

answered 2012-08-10 18:29:14 -0500

TRiBi gravatar image

updated 2012-08-10 18:50:55 -0500

Thank you for your answer I will try to discover if it is used for the Sobel operator. It,s defined in this function in objectdetect.cpp

CV_WRAP HOGDescriptor() : winSize(64,128), blockSize(16,16), blockStride(8,8), cellSize(8,8), nbins(9), derivAperture(1), winSigma(-1), histogramNormType(HOGDescriptor::L2Hys), L2HysThreshold(0.2), gammaCorrection(true), nlevels(HOGDescriptor::DEFAULT_NLEVELS)

If anyone had the same problem I think I have found the solution in:

aperture_size Size of the extended Sobel kernel, must be 1, 3, 5 or 7. In all cases except 1, aperture_size ×aperture_size separable kernel will be used to calculate the derivative. For aperture_size=1 3x1 or 1x3 kernel is used (Gaussian smoothing is not done).

But I'm not totally sure.

edit flag offensive delete link more

answered 2012-08-10 14:35:44 -0500

updated 2012-08-10 14:38:00 -0500


It seems weird but I grepped HOG source code and I'm not able to find a place where this parameter is actually used. I expected it to be aperture size of the Sobel operator to compute derivatives (like Canny has apertureSize parameter). But it seems like this parameter is not used in HOGDescriptor.

edit flag offensive delete link more
Login/Signup to Answer

Question Tools


Asked: 2012-08-10 13:01:55 -0500

Seen: 882 times

Last updated: Aug 10 '12