[ptx] Even more SIFT successes

Pablo d'Angelo pablo at mathematik.uni-ulm.de
Mon Jan 19 22:43:22 GMT 2004


On Mon, 19 Jan 2004, Ed Halley wrote:

> Is there any need to rethink the feature markers situation?  SIFT just
> finds prominent items in one image, right?  So do we need to visualize
> new markers that have not been associated/tied to other images?

The point of the SIFT features its that they are easy to match, so we
will only provide functions which create complete pairs. I don't see the use
for creating a lonely point. do you think the automatically created pairs
should be marked in a different way?

ciao
  Pablo
-- 
http://wurm.wohnheim.uni-ulm.de/~redman/
Please use PGP


More information about the ptX mailing list