Hot answers tagged

9 votes

What are some free alternatives to SIFT/ SURF that can be used in commercial applications?

I would rather look into KAZE / AKAZE, which perform equally good with significant speed-up. The deformation cases are also tolerated. OpenCV has recently obtained an implementation through GSoC 2014. ...
user avatar
  • 5,225
6 votes
Accepted

Why Does the Odd Multiple of $ \frac{\pi}{4} $ on Gaussian Cause Loss in Repeatability Under Image Rotations?

The answer boils down to 2 issues with the practical approximations of the Gaussian Kernel: Though the Gaussian Kernel is radially symmetric its discrete approximation has a rectangle support. ...
user avatar
  • 41.4k
6 votes
Accepted

What Is "Description Vector" in Image Processing?

I think you have a matrix. Each Row / Column is a descriptor vector of a point in the image. Just like having features, let's say M features, and each point has M values corresponding to M features. ...
user avatar
  • 41.4k
6 votes

Why optical flow ? why not tracking

There are two versions of optical flow(OF): Feature based (sparse) or dense. In the dense version OF is applied to all the image pixels, while in the sparse one, only certain characteristic feature ...
user avatar
  • 5,225
4 votes
Accepted

Laplacian of Gaussian Approximation and Gaussian Blur as the Solution of Heat Equation

I'm not sure I fully understood what's the issue you're having. Yet I will show a simple property of the Gaussian filter which might make things clearer. For simplicity, I will use 1D Signal. Yet it ...
user avatar
  • 41.4k
3 votes

Why do we need multiple layers in each octave and multiple octaves in SIFT?

In music theory, an octave is an interval in frequency, from a frequency $f$ to frequency $2f$. For example "an octave higher" means "twice the frequency". Expressed as wavelength ...
user avatar
3 votes

Why do we need to construct Gaussian pyramid using SIFT detector

Actually, the purpose of all this is to approximate a Laplacian of Gaussian! This computation is part of the corner detection of SIFT. You can find corners by examining extrema of the Laplacian of ...
user avatar
  • 3,887
3 votes

Would the locations of SIFT features generally agree with features detected by Shi-Tomasi method?

Probably not. The SIFT detector finds centers of blob-like features. Shi-Tomasi detector finds corners. Furthermore, SIFT detector operates at multiple scales, while the classic Shi-Tomasi does not.
user avatar
  • 4,941
2 votes
Accepted

Using ROC curves for comparing the performance of SIFT and SURF

The ROC: ROC curves are popularly used as performance metric for classification tasks. If the images in your dataset has class labels, then you can employ supervised learning to train a classifier (...
user avatar
2 votes

Intuitive understanding of scale-space extrema detection

One of the most important characteristics of the key points is its repeatability under different geometric transformations and also lighting. Repeatability ensures that if, for example, you have two ...
user avatar
2 votes

Intuitive understanding of scale-space extrema detection

LoG and DoG (an approximation of LoG) masks can serve as blob detectors. A blob can exist in an image at a number of locations $(x,y)$-coordinates and scales (some parameter; $t$). In some situation ...
user avatar
  • 21
2 votes

Why resizing an image smoothed by Gaussian by factor of 2 also increase sigma by factor of 2

A first discrete is pixel wise: imagine an image with only one active pixel. If you upsample it by two in both directions, you get a $2\times 2$ pixel block. Below, I did not extend the size of the ...
user avatar
1 vote

What's the difference between SIFT and general stereo matching algorithm (eg, sgbm)?

After weeks of reading, researching and experiments, now I have more knowledge to answer my own question. Both of SIFT and SGBM can be used to find matching points but they are very different in the ...
user avatar
  • 131
1 vote

Invariances of FFT-based Image-Registration vs. SIFT-Features

I don't understand whether these processes are also invariant to object-alterations! They are not. How would the extracted fft-features look, if I alter the object (scratches, marks, dents etc.)? ...
user avatar
  • 10.1k
1 vote
Accepted

SIFT About Difference-of-Gaussian function extrema?

You have to take the derivative with respect to the vector $x$ and set it equal to zero. For a constant matrix $A$, the derivative of $A^Tx$ is $A$, and the derivative of $\frac12 x^TA^Tx=Ax$. So ...
user avatar
  • 80.4k
1 vote

Why is scale space (DoG) needed to detect scale invariant features?

As you know, you find interest point (SIFT points) by finding the local maxima in scale-space, it mean the response of the detector must be maximum regarding the coordinates and also the scale. So ...
user avatar
  • 1,306
1 vote

How to recognize an object from a small training set of images?

Few reasons i could think of are: Size of your training set is very small. Larger training sets have always been the key for accuracy. Each algorithm will have some drawback like SURF is not good at ...
user avatar
1 vote
Accepted

Hessian Matrix. Second partial derivative test

The function is fully approximated if one uses all the derivatives (see Taylor expansion). With using the Hessian only, we can only make a second degree approximation (because it is second derivative ...
user avatar
  • 5,225
1 vote

Would the locations of SIFT features generally agree with features detected by Shi-Tomasi method?

Depends. If you use two separate pre-canned libraries to compute them, likely not. However, note that when people talk about "SIFT features" they refer to two things: Point locations on the images ...
user avatar
1 vote

What is the story behind the story about SIFT descriptor?

The descriptor obtained from a $64\times 64$ neighborhood of interest point at the obtained scale. It will divide this $64\times 64$ region to $16\times 16$ patches which lead to 16 patches. For ...
user avatar
  • 1,306
1 vote

Why is it necessary to implement octaves in sift

Yes, only scale space is sufficient, but at some point when you are scaling it down, instead of creating new Gaussian filters, it's more efficient to just resize the image and use the same/old filters ...
user avatar
1 vote

How to proceed for BOW creation From sift features

First of all, read this. Then: Due to the nature of this question, I will only give you some hints on pre-processing to improve your retrieval task. Don't use Sift. Use RootSift. This is a ...
user avatar
  • 5,225
1 vote

How to proceed for BOW creation From sift features

There is now support for the bag-of-words model in the Computer Vision System Toolbox for MATLAB.
user avatar
  • 4,941
1 vote

Scale and Rotation invariant feature descriptors

If you remap a local patch around a feature point to log–polar coordinates (with the origin in the point of interest), scale changes correspond to a translation along the log–radial axis, while ...
user avatar
1 vote

Scale and Rotation invariant feature descriptors

I would rather look into KAZE / AKAZE, which perform equally good with significant speed-up. The deformation cases are also tolerated. OpenCV has recently obtained an implementation through GSoC 2014. ...
user avatar
  • 5,225
1 vote

Is sift a good way to extract features from an image?

Building on previous responses: (1) You can use SIFT (or another improved variant of this local-patch descriptor) with dense sampling, instead of the inbuilt detector. You can choose the size of the ...
user avatar

Only top scored, non community-wiki answers of a minimum length are eligible