From: Angus Dorbie (dorbie++at++sgi.com)
Date: 05/23/2000 10:54:27
I have no code to offer, but you should try breaking the problem down.
First make sure your tracking is calibrated properly to the cave.
Once you know this then the problem is either a scaling mismatch of
virtual to real coordinates, an error with your frustum calculations or
some other gross translation error.
Perhaps you are missing out a complete stage of the process.
testing the individual stages should allow you to sort things out.
The stages are:
1) track the eye
2) recompute the frusta with new eye w.r.t. cave walls
3) apply scaled viewing translation based on tracked eye position
4) make sure there is no viewing rotation (OK this isn't a stage but
it's important enough to mention)
You can calibrate 1 by simply checking the numbers from the tracker and
scaling them to the cave.
Check 2 by eliminating stage 3 and ensuring that nothing changes w.r.t.
the viewer as the eye moves around the cave, i.e. virtual colimation.
3 requires that the virtual units match world units. You can check this
by moving around the cave, even if 2 is wrong or if you disable it
completely you should be able to place an virtual object in the corners
of the cave and move the tracker to the object and it will bove right up
to the viewer. You're just checking the units of the virtual world, a
prerequisite is that stage 1 is correct.
If you independently check these stages and get them right it sould all
come together easily.
Cheers,Angus.
Yohan Baillot wrote:
>
> Hi,
>
> does anybody would be willing to share a piece of code that has been used
> successfully to set the viewing frustum in a CAVE-like system from an
> head-tracker.
> As a verification, we displayed a model of the cave (a cube done with 12
> cylinders)
> in the cave itself and we move the tracker around. As the tracker moves, the
> projection
> of the cave on the walls should stay the same since the projection of the
> walls of the
> cave on themselves are the walls themselves. However, the cylinder
> representing the
> cave are moving all around the place, in a way that suggest that the frustum
> is not
> changed but still have the same field of view.
>
> Any body could give some insight on this problem?
>
> thanks
>
> Yohan
>
> _______________________________________________________________________
>
> Yohan BAILLOT
> Electrical and Computer Engineer
>
> Virtual Reality Laboratory,
> Advanced Information Technology (Code 5580),
> Naval Research Laboratory,
> 4555 Overlook Avenue SW,
> Washington, DC 20375-5337
> baillot++at++ait.nrl.navy.mil
> (202) 404 7801
>
> Home : (703) 566 3684
> Cell : (407) 222 6359
> Fax : (202) 767 1122
> Web : www.ait.nrl.navy.mil/people/baillot
>
> _______________________________________________________________________
>
> -----------------------------------------------------------------------
> List Archives, FAQ, FTP: http://www.sgi.com/software/performer/
> Submissions: info-performer++at++sgi.com
> Admin. requests: info-performer-request++at++sgi.com
-- For Performer+OpenGL tutorials http://www.dorbie.com/"In the middle of difficulty lies opportunity." --Albert Einstein
This archive was generated by hypermail 2b29 : Tue May 23 2000 - 10:54:50 PDT