[ptx] Registration testcases

[Pablo d'Angelo]

On Tue, 20 Jan 2004, Alexandre Jenny wrote:

> 
> > > By having that, you didn't need anymore to worry about 
> > overexposure, 
> > > underexposure, etc. Just take many pictures with different 
> > exposures 
> > > and let have the algorithm solves that. Moreover, blending 
> > is really 
> > > simplified. ... It's just an idea for future ...
> > 
> > Why stop there...?  Once you go down that route you can aim 
> > for super-resolution.
> > 
> > At the limit you could imagine holding down the trigger on 
> > your SLR, waving it around, and then having the pictures all 
> > aligned, overexposed and underexposed shots used to fill in 
> > fully dynamic range, and the overlapping shots used to create 
> > super-resolution images...  
> 
> That's exactly what I mean. 
> After having solved all problems of stitching picture, that is extending a
> picture spacially, 
> the future is extending picture in dynamic by using HDR technics.

I've read the paper by debevec, and saw that they even have a C software for
the calibration purpose, that can write out the calibration mapping G, given
some registered images, apperature and exposure times, a gain factor and
neutral density:

http://www.debevec.org/FiatLux/mkhdr/

Seems like it should be quite easy to read the calibration, and apply the
transform during remapping :)

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