[sci.virtual-worlds] Tracking of Targets

mccool@dgp.toronto.edu (Michael McCool) (06/20/91)

This is somewhat related to the Japanese Stereo TV thread, in particular the
tracking of objects.  Does anyone know of an inexpensive signal for 
tracking with a video camera?  IE the object being shot has some kind of
target attached to it and the camera is on a moterized platform.  I seem
to have seen something of the sort (with two sensors on the end of boom
projecting from the base of the platform) but have not been able to track it
down again.

Application not: this is so people can take videotapes of themselves 
riding a horse and then self-critique their posture.  Is this virtual
worlds?

Michael McCool@dgp.utoronto.ca

mgobbi@cs.ubc.ca (Mike Gobbi) (06/20/91)

In article <1991Jun19.193222.20056@milton.u.washington.edu> mccool@dgp.toronto.e
du (Michael McCool) writes:
>>
>This is somewhat related to the Japanese Stereo TV thread, in particular the
>tracking of objects.  Does anyone know of an inexpensive signal for 
> ...
>
>Application note: this is so people can take videotapes of themselves 
>riding a horse and then self-critique their posture.  Is this virtual
>worlds?
>
>Michael McCool@dgp.utoronto.ca

I would suggest that the target wear a small radio/sound transmitter,
and the camera have 5 directional pickups (one aimed "straight", the others
"up", "down", "left", and "right" by about 5 degrees).  The camera would be
aimed by swinging it in the direction of the stronger signal in the
vertical and horizontal planes.  If the "straight" pickup has the best
signal, the camera is held steady.

This has the problem that it must start out aimed at the target, and if
the target moves faster than the device can rotate it will lose it.  There
are probably numerous technical problems as well, since I'm not a radio
expert, but I think it might work.

--
    __     In Quest of Knowledge...
   /..\
--mm--mm--     Mike Gobbi
               (mgobbi@cs.ubc.ca)