Which feature detection algorithm is the easiest to learn?

Question

I am wrapping around function detector algorithms. I studied the options that I have: SIFT, SURF, BRISK, FREAK, etc. All of them seem rather complicated in terms of basic mathematics. On the contrary, I want to take one step at a time, so I am looking for a simple method that should not be as good as SURF, for example. What algorithm would you recommend to study and implement?

Answer 1

First of all, the difference between the detector and the descriptor should be borne in mind. A detector is an algorithm for detecting points of interest in an image, which are usually either corners or centers of structures like blobs. Then, if you need to map these points to images, you compute descriptors, which are some kind of value vectors that represent patches around points of interest.

This should help eliminate some confusion. For example, “good tracking features,” for example, a mini-angle detector, is a percentage point detector. FREAK - function descriptor. SIFT, SURF, and BRISK include both a detector and a descriptor. In general, however, you can mix and match detectors and descriptors.

, , GFTT Harris, . .

SIFT. , , , .

Answer 2

, , , - NxN . , , , .., .

, " ", , (http://cvlabwww.epfl.ch/~lepetit/papers/calonder_eccv10.pdf), . , : , AKAZE, BRISK, FREAK ORB.

Answer 3

, GFTT: tracj, -. , , .

Answer 4

-, SIFT. . Vlfeat.org , , , .