The Frankencamera

Ron K. Jeffries rjeffries at gmail.com
Sun Jul 25 20:12:55 EDT 2010


I question the feasibilty of recovering a high
quality image if no lens is used at all.

The Hubble case was a lens system that was
not focused correctly.

In any case, even zoom lenses are not very expensive. To achieve a usable
digital imaging
system in finite time, the plan of record would
logically include a commercial lens and image
sensor. The computational power behind those
minimal parts can do a lot of magic, witness the
cool results from the Frankencamera project,
which to my eye appears to be very open, and also
has some heavy muscles, e.g. Stanford & MIT.

What is the application goal of this project?
One assumes industrial grade security imaging?

It's certainly not a Copyleft Point and Shoot. ;)
---
Ron K. Jeffries
http://ronkjeffries.posterous.com








2010/7/25 Andrés Calderón <andres.calderon at emqbit.com>

> On Sun, Jul 25, 2010 at 4:30 PM, Werner Almesberger <werner at openmoko.org>
> wrote:
> > Marco Castorina wrote:
> >> I know a little about digital photography: yes you can use digital
> >> zoom but what it does is using just a part of the CMOS and
> >> calculate the pixels from there for the full resolution image.
> >
> > Ah, zoom was a bad example. You're right, digital zoom just throws
> > away information and interpolates the rest.
> >
> > What I meant was, instead of using optics to create an undistorted
> > small image on the sensor, one could use a distorted image and
> > algorithmically reconstruct the undistorted image. Like they did
> > it with the Hubble :-)
> >
> > For zoom, you'd still have to change the angle of view, but you
> > wouldn't need the complex optics that make sure the image doesn't
> > get distorted when changing the focal length.
> >
>
> Noise reduction it's another great example of computational photography:
> http://graphics.stanford.edu/papers/gkdtrees/
>
> -   Andrés Calderón
>
>
> > - Werner
> >
> > _______________________________________________
> > Qi Developer Mailing List
> > Mail to list (members only): developer at lists.qi-hardware.com
> > Subscribe or Unsubscribe:
> http://en.qi-hardware.com/mailman/listinfo/developer
> >
>
> _______________________________________________
> Qi Developer Mailing List
> Mail to list (members only): developer at lists.qi-hardware.com
> Subscribe or Unsubscribe:
> http://en.qi-hardware.com/mailman/listinfo/developer
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.en.qi-hardware.com/pipermail/discussion/attachments/20100725/a5182504/attachment.htm>


More information about the discussion mailing list


interactive