[ptx] autopano: first attempt at semi-automatic panorama stitching

[Pablo d'Angelo]

Hallo Sebastian,

Sebastian Nowozin schrieb am Sonntag, den 28. Dezember 2003:

> > Your first correlation step you could use your rotation invariant, radial
> > sum descriptor, but additionally you have additional information that can be
> > exploited, like it is done with the RANSAC in Browns approach.
>  
> Have to read up on that ;) (but programming and trying out things is more fun
> ;)

Yep. If I had some more time I'd like to implement some stuff in that
direction as well. 

> Just for the part (1) I now completely dropped my idea of feature pixel
> identification. I read up a bit on standard ways to do image feature indexing
> and found an interesting PhD thesis:
> http://www.cat.csiro.au/cmst/staff/jmr/thesis/

Interesting, I know Jonathan, but never read his thesis ;)

> If I understand Harris's algorithm I will implement it, but I will try to put
> on a faster and more accurate version of my small program tomorrow using the
> Smith algorithm.

The VIGRA library, used in hugin/nona, already contains a harris corner
detector, so I'll just use that in a c++ implementation.

I'm already interested to see how your program performs with the new feature
detector. Unfortunately, I don't like to download mono over a normal
modem.

> I also found an interesting paper today that might be an alternative to the
> [iccv2003] panorama paper you posted me:
> http://citeseer.nj.nec.com/mikolajczyk02affine.html

Sounds very interesting, but also a bit overkill the top for our purposes, since
we probably do not need fully affine invariant features.

> > Alexandre hasn't posted it yet. we have to think if we want to introduce
> > patented algorithms into hugin. hmm maybe its ok for us germans, at least for
> > now ;)
>  
> If there is a good description of whats patented exactly one can surely work
> around that.

Yep, it would be interesting, but I haven't found that patent yet, it
doesn't seem to be in the german patent database and I just made a 5 minute
search with the US patent database, but didn't find it.

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