Rank: 466 based on 224 downloads (last 30 days) and 2 files submitted
photo

Leo

E-mail

Personal Profile:
Professional Interests:

 

Watch this Author's files

 

Files Posted by Leo View all
Updated   File Tags Downloads
(last 30 days)
Comments Rating
15 Aug 2012 Histograms of Oriented Gradients mex function for calculating histograms of (oriented) gradients (Dalal & Triggs CVPR 2005). Author: Leo computer vision, image processing, gradients, histograms, object detection, histogram of oriented... 119 41
  • 4.83333
4.8 | 6 ratings
11 Apr 2012 Random Forest Creates an ensemble of cart trees similar to the matlab TreeBagger class. Author: Leo tree ensembles, cart trees, treebagger, machine learning, decision tree, random forest 105 58
  • 4.54545
4.5 | 15 ratings
Comments and Ratings by Leo View all
Updated File Comments Rating
18 Mar 2013 Histograms of Oriented Gradients mex function for calculating histograms of (oriented) gradients (Dalal & Triggs CVPR 2005). Author: Leo

Hi Victor,

It is a bit difficult to comment on the VL code as it isnt mine, but I think the difference comes from how the normalization of the cells on the boundary is treated.

The cells on the left and bottom boundary only belong to two blocks (or in some cases one). For example there is no block starting at a cell on the left boundary and extending to the left as that would take it outside the image. Thus a 256x256 image only has 63*63 blocks of size 2x2.

If I understand the VL code correctly ,hog.c line 927 and onward, it assumes that every cell belongs to 4 blocks. For the case of boudary cells, where this assumption does not hold, certain block normalization schemes are simply repeated to make up the four blocks, see lines 932-935. So if I am correct, for boundary cells, there should be repetitions in the final descriptor of the VL code. Though obviously you shouldnt cite me on this :)

Hope this helps.

Leo

29 Jan 2013 Histograms of Oriented Gradients mex function for calculating histograms of (oriented) gradients (Dalal & Triggs CVPR 2005). Author: Leo

Hi Anas,

I am afraid tracking isnt my field soI dont really know. It would depend a lot on the resolution of the camera andhow many fps you want to process. In general HoGs are computationally expensive so I wouldn't be suprised if simpler descriptors are used in practice.

Leo

22 Dec 2012 Random Forest Creates an ensemble of cart trees similar to the matlab TreeBagger class. Author: Leo

Hi fairy,

lcc is not a cpp compiler. Using the Visual Studio compiler I think the following should do the trick.

in GBCC.cpp change line 24 to

saved_logs[j] = log(j+1)/log(2);

line 115 to

if (diff_labels[nl]>0) bh-=diff_labels[nl]*(log(diff_labels[nl])/log(2)-log(sum_W)/log(2));

line 151 to

if(diff_labels_l[nl]>0) ch-=(diff_labels_l[nl])*(log(diff_labels_l[nl])/log(2)-log(sum_l)/log(2));

and line 152 to

if(diff_labels_r[nl]>0) ch-=(diff_labels_r[nl])*(log(diff_labels_r[nl])/log(2)-log(sum_W-sum_l)/log(2));

Hope this solves it.

Leo

22 Dec 2012 Random Forest Creates an ensemble of cart trees similar to the matlab TreeBagger class. Author: Leo

Hi fairy,

Could you copy paste the exact error message you get when running mx_compile_cartree.m ?

Leo

18 Dec 2012 Random Forest Creates an ensemble of cart trees similar to the matlab TreeBagger class. Author: Leo

Hi fairy,

It would seem that the function is not in matlab's search path. You can run

addpath(genpath(cd))

Leo

Comments and Ratings on Leo's Files View all
Updated File Comment by Comments Rating
02 May 2014 Histograms of Oriented Gradients mex function for calculating histograms of (oriented) gradients (Dalal & Triggs CVPR 2005). Author: Leo faten, san

Hi,I'd like to repeat Raul's unanswered question.
"I want to use HOG with SVM. Your code returned different size's of vector's for different images. But, to train the features with SVM, the vector's must be in the same size."
is there any suggestion how can we change the code so that we can use it with the SVM? thx in advance

30 Mar 2014 Histograms of Oriented Gradients mex function for calculating histograms of (oriented) gradients (Dalal & Triggs CVPR 2005). Author: Leo Aviel, Yuval

the first option is the correct one:
histImg = reshape(hog, [Params(1) nCells nCells]);

24 Mar 2014 Histograms of Oriented Gradients mex function for calculating histograms of (oriented) gradients (Dalal & Triggs CVPR 2005). Author: Leo Aviel, Yuval

Hi Leo, I'd like to repeat victorp's unanswered question:
"could you tell me how to fold the output vector back (to 3D)?"

in particular, considering the following code:

Params = [9 4 1 0 0.2]; hogs = HoG(Image,Params);
nCells = sqrt(length(hogs)/Params[1]) % (assuming square Image)

should it be:
histImg = reshape(hog, [Params(1) nCells nCells]);

or

histImg = reshape(hog, [nCells nCells Params(1)]);

??
Thanks,
yuval

28 Feb 2014 Histograms of Oriented Gradients mex function for calculating histograms of (oriented) gradients (Dalal & Triggs CVPR 2005). Author: Leo Sanny, A.

Hi Leo. I'm new and not well-known about this topic thus I need some help. Can you please explain the part about interpolation? In the paper it says "votes are interpolated bilinearly between the neighbouring bin centres in both orientation and position" and the code says "trilinear interpolation". Are there any improvement you made on Dalal-Triggs paper? Thank you :)

31 Jan 2014 Random Forest Creates an ensemble of cart trees similar to the matlab TreeBagger class. Author: Leo Hussein

Could anyone give an example of how to use this function, I mean the input parameters,.... I successfully built it, so if anyone could please advise. Thanks

Contact us