SIFT and HOG features descriptors

Question

Heyyy,,, Im going to do classification. And for the descriptor I planning to use the HOG and SIFT descriptors from LOWE..

 1. For HOG, is that true that we need to compute the gradient of all image's pixels?? 
 For example we have image with size 10x10 pixels. And we compute the HOG to seek the orientation        
 for every pixels of image. And eventually we will get 100 orientations and generate histogram   
 (represent all of those image's pixels orientation). This Histogram is going to be used for the      
 classification??
 And to get the scale invariant we need to perform this descriptor of different size (scale)     
 images??

SIFT is Scale Invariant Features Transform. So it's scale and rotation invariant.
I read from here that in SIFT we need to do smoothing using Gaussian to make our image's
resolution become low..

   2. Why we have to do that?
   And for scale invariant features, how to obtain that in SIFT?? Do we need to rescale our image   
   in every octave and then apply Gaussian filter in this new scaled images?? Or it's enough to   
   get scale invariant only in 1 octave by applying 3 times gaussian filter??
   How about the histogram, Is it same with HOG that we have to compute all the pixels???

Thankss

Hi, welcome to SO. Your question, apart from being poorly formatted, is not about programming. Figure out the theory and come back with any implementation problems you might encounter. — Cape Code, Apr 03 '14 at 17:41

score 2 · Answer 1 · answered Apr 05 '14 at 15:07

If you plan to implement SIFT at least you should read the paper of Lowe since is one of the most cited one in CV history!

In SIFT gaussian smoothing is applied in order to compute the DOG (difference of gaussian). Then performing Scale Extrema Detection you will detect the feature points.

Once you have this feature points you will need to compute the HOG for each feature. You don't need to compute it for the whole image! Since we take a 16x16 neighbourhood the result will be a 128 length descriptor.

The scale invariance comes because the gradients are sampled in a window around each keypoint with respect to the scale of the keypoing.

But again, you should go through Lowe's paper because it is very clear there. You should also have some concepts clear such as feature points, HOG, DOG, etc. to really understand SIFT

SIFT and HOG features descriptors

1 Answers1