Possible Duplicate:
Free alternatives to SIFT/SURF for commercial use

There are same very good algorithms for interest point detection/description, like:

  • SIFT (Scale-Invariant Feature Transform)
  • SURF (Speeded-Up Robust Features)
  • MOPs (Multi-Scale Oriented Patches)

Regrettably, at least the first two are patented and cannot be used commercially without paying huge fees to the authors. I am currently negotiating about the third one, which is patented as well.

In a search for patent-free algorithm, I found there is a multitude of patents like this one. Huh?! Is there merely anything usable then?

Maybe at least Harris-Laplace or some basic detector can be used. But how to be sure?



If you only want detectors:

Descriptors are harder...

Histograms of Oriented Gradients might be worth considering - again I've not seen any claims of patent on the original form.

  • $\begingroup$ Thanks. I will finally go for the multi-scale Harris-Laplace detector. The descriptor will consists simply of normalized gradients if it would be sufficient. If not, that affine invariant version is also proposed my Mikolajczyk & Schmidt, which I believe is also patent-free. $\endgroup$
    – Libor
    Mar 7 '12 at 15:00

Not the answer you're looking for? Browse other questions tagged or ask your own question.