face levers in Maya (thoughts)

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
8 messages Options
Reply | Threaded
Open this post in threaded view
|

face levers in Maya (thoughts)

Paul Sheldon-2
It illustrated a mesh that you tweak point positions on to deform.

You could paint a face on as a texture and have it follow the mesh.

I'd have to go look for the example Maya file I downloaded.

I've made a usb controller in mac os x
that would send inter application communicate
through applescript melscript messages inside Maya.

What I did was much simpler than a facial sensor control.

I switched camera position with generated melscript
and, with help of Juri Munki on circuitry and pre os x software in os 9,
liquid crystal shutterglasses' black eyeing  to get 3D (1 bit
which eye control to usb device
not vice versa = control from usb device).

So, with abstract facial model with painted on texture,
you'd need an a to d converter (more than a bit).
I believe such hardware is currently expensive as its marketbase
is in cinema. As marketbase increases, the cost goes down.

Reply | Threaded
Open this post in threaded view
|

Re: face levers in Maya (thoughts)

Les Howell
On Tue, 2007-04-03 at 15:32 -0800, [hidden email] wrote:

> It illustrated a mesh that you tweak point positions on to deform.
>
> You could paint a face on as a texture and have it follow the mesh.
>
> I'd have to go look for the example Maya file I downloaded.
>
> I've made a usb controller in mac os x
> that would send inter application communicate
> through applescript melscript messages inside Maya.
>
> What I did was much simpler than a facial sensor control.
>
> I switched camera position with generated melscript
> and, with help of Juri Munki on circuitry and pre os x software in os 9,
> liquid crystal shutterglasses' black eyeing  to get 3D (1 bit
> which eye control to usb device
> not vice versa = control from usb device).
>
> So, with abstract facial model with painted on texture,
> you'd need an a to d converter (more than a bit).
> I believe such hardware is currently expensive as its marketbase
> is in cinema. As marketbase increases, the cost goes down.
>
Screen synchronized glasses were marketed for a while, although I
haven't seen them lately.  There are some issues with them, one of which
is epilepitics have issues with flashing images, and another being
eyestrain.  Also the 50/60 hz update rate is too slow, and when scanned
alternately the eye looses some image memory.  This causes headaches in
many people and probably is the prime reason that the goggles were not
accepted.  Projection systems use dual cameras with polarized lenses so
that the image is either vertically or horizontally polarized, and the
corresponding lense admits the correct image.  The polarizing also
accommodates color, which the red/green images of the 50's did not.

        Yet another recent outgrowth is goggles with individual screens along
with a position and accelerometer on board so the images can be properly
modified to match viewer expectations.  (Can you tell I've been
interested in this for a long time?)  There is currently some form of
rotational device being used for medical diagnostics but I haven't
actually seen one of those.  A free air projection system has been
demonstrated at some trade shows, but I think it is still 2d, but it is
certainly interesting technology.

        There is a field of work done in stereo perception, with some still
images available from MIT (at one time anyway), and some interesting
work with various primitive biological eyes, for instance a frog's eye
uses a bar/dot system so that an image crossing the dot and then the bar
will feed back some kind of vector information that is directly coupled
to its tongue.  A sort of automatic targeting device that allows it to
snatch bugs right out of the air without conscious thought.  Not quite
related to 3d, but an interesting approach to calculating a velocity and
trajectory vector that could prove useful to someone.
       
        I read a lot of the early papers on vision and AI, because that is a
perceptual problem that will certainly require resolution for some
interesting robotics, human aids, and other functions.

        Stereo vision can produce a bas-relief of data just from our two eyes,
yielding perspective, surface recognition (smooth, rough, textured,
etc.)  One of the amazing things is that an error of only .0003 inches
is detectable by eyesight alone (for  you metric types that is about 76
micrometers) at a distance of about 3'.  This is of course in the center
of precise vision.  outside the middle of the eye it is somewhat less
precise, I have heard 1mm as the resolution beyond 40degrees off center.
And our normal range of vision is somewhere near 220 degrees.  

        All of these factors are part of the issues surrounding immersive
vision technologies.  But is it soooo interesting, don't you think?

Regards,
Les H

Reply | Threaded
Open this post in threaded view
|

Re: face levers in Maya (thoughts)

Les Howell
In reply to this post by Paul Sheldon-2
On Tue, 2007-04-03 at 15:32 -0800, [hidden email] wrote:

> It illustrated a mesh that you tweak point positions on to deform.
>
> You could paint a face on as a texture and have it follow the mesh.
>
> I'd have to go look for the example Maya file I downloaded.
>
> I've made a usb controller in mac os x
> that would send inter application communicate
> through applescript melscript messages inside Maya.
>
> What I did was much simpler than a facial sensor control.
>
> I switched camera position with generated melscript
> and, with help of Juri Munki on circuitry and pre os x software in os 9,
> liquid crystal shutterglasses' black eyeing  to get 3D (1 bit
> which eye control to usb device
> not vice versa = control from usb device).
>
> So, with abstract facial model with painted on texture,
> you'd need an a to d converter (more than a bit).
> I believe such hardware is currently expensive as its marketbase
> is in cinema. As marketbase increases, the cost goes down.
>
I don't understand the need for an A/D converter.  Today all cameras
have the light to image (A/D converter and pixel addressing) built in.
Capturing the image in real time is a bit of a hassel if you want
sufficient definition, but with firewire or newer usb technology it is
certainly possible to get good images, not eye quality, nor even 35mm
(about 12Megapixel with 32bit/color and gamma correction), but surely
good enough for our task here.

        If you were to attempt something like a flying spot scanner for
positional input, then an A/D would be required, but I think a dual
camera setup with the correct software could accomplish much of the
task.

Regards,
Les H
Regards,
Les H

Reply | Threaded
Open this post in threaded view
|

Re: face levers in Maya (thoughts)

Paul Sheldon-2
In reply to this post by Paul Sheldon-2
OK. I was reading about Merce Cunningham around 1999
when digital cameras, if they even existed, were a new word
and I probably confounded words :
confounded A/D converter with digital camera.

Sorry.

This stuff is soooo interesting, indeed.

Reply | Threaded
Open this post in threaded view
|

Re: [OT] face levers in Maya (thoughts)

Howard Stearns
off topic: Digital CCD cameras have been around since 1970.
http://en.wikipedia.org/wiki/Charge-coupled_device
In 1984 I helped create a robotic telescope that analyzed it's own data to move
itself around.

[hidden email] wrote:
> OK. I was reading about Merce Cunningham around 1999
> when digital cameras, if they even existed, were a new word
> and I probably confounded words :
> confounded A/D converter with digital camera.
>
> Sorry.
>
> This stuff is soooo interesting, indeed.
>

--
Howard Stearns
University of Wisconsin - Madison
Division of Information Technology
mailto:[hidden email]
jabber:[hidden email]
office:+1-608-262-3724
mobile:+1-608-658-2419
Reply | Threaded
Open this post in threaded view
|

Re: Re: [OT] face levers in Maya (thoughts)

Ric Moore
On Wed, 2007-04-04 at 10:00 -0500, Howard Stearns wrote:
> off topic: Digital CCD cameras have been around since 1970.
> http://en.wikipedia.org/wiki/Charge-coupled_device
> In 1984 I helped create a robotic telescope that analyzed it's own data to move
> itself around.

You DO go back aways! Just a couple of years before that, when I worked
for BASF, there were only two employees in the US that had home
computers. I had an original garage built Apple ][ and the other guy had
a Commodore Pet. I used to catch merry Hell from my District Manager for
"screwing around" with "that" computer using Visacalc to calculate my
territory's sales and my piece of the financial pie. I had a Teletype 44
for my printer and an internal DC Hayes 300 baud modem. Oh yeah, two
floppy drives @ $500 apiece. I don't miss those days at all, although I
did get my 15 minutes of fame by being interviewed in Leo Gomes Esquire
article on Computer Piracy in 1982. Yo Ho! I got a whole paragraph.
:) Ric

--
================================================
My father, Victor Moore (Vic) used to say:
"There are two Great Sins in the world...
..the Sin of Ignorance, and the Sin of Stupidity.
Only the former may be overcome." R.I.P. Dad.
Linux user# 44256 Sign up at: http://counter.li.org/
http://www.sourceforge.net/projects/oar
http://www.wayward4now.net
================================================


Reply | Threaded
Open this post in threaded view
|

Re: Re: [OT] face levers in Maya (thoughts)

Les Howell
In reply to this post by Howard Stearns
Does this use some form of point location with trigonometry to arrive at
the delta in position?  If so is the code for selecting tracking points
available for generic use?  Could it be used to garner points from two
slightly different scenes (say the eyes, corner of the mouth, chin or
ears) so a correlation algorithm could compute the physical offsets and
arrive at a location?

And is it in the Public Domain?

Regards,
Les H
On Wed, 2007-04-04 at 10:00 -0500, Howard Stearns wrote:

> off topic: Digital CCD cameras have been around since 1970.
> http://en.wikipedia.org/wiki/Charge-coupled_device
> In 1984 I helped create a robotic telescope that analyzed it's own data to move
> itself around.
>
> [hidden email] wrote:
> > OK. I was reading about Merce Cunningham around 1999
> > when digital cameras, if they even existed, were a new word
> > and I probably confounded words :
> > confounded A/D converter with digital camera.
> >
> > Sorry.
> >
> > This stuff is soooo interesting, indeed.
> >
>

Reply | Threaded
Open this post in threaded view
|

Re: Re: [OT] face levers in Maya (thoughts)

Les Howell
In reply to this post by Ric Moore
On Wed, 2007-04-04 at 11:38 -0400, Ric Moore wrote:

> On Wed, 2007-04-04 at 10:00 -0500, Howard Stearns wrote:
> > off topic: Digital CCD cameras have been around since 1970.
> > http://en.wikipedia.org/wiki/Charge-coupled_device
> > In 1984 I helped create a robotic telescope that analyzed it's own data to move
> > itself around.
>
> You DO go back aways! Just a couple of years before that, when I worked
> for BASF, there were only two employees in the US that had home
> computers. I had an original garage built Apple ][ and the other guy had
> a Commodore Pet. I used to catch merry Hell from my District Manager for
> "screwing around" with "that" computer using Visacalc to calculate my
> territory's sales and my piece of the financial pie. I had a Teletype 44
> for my printer and an internal DC Hayes 300 baud modem. Oh yeah, two
> floppy drives @ $500 apiece. I don't miss those days at all, although I
> did get my 15 minutes of fame by being interviewed in Leo Gomes Esquire
> article on Computer Piracy in 1982. Yo Ho! I got a whole paragraph.
> :) Ric
>
I have my original Altair 8800B (which I purchased used in 1979) in my
garage.  Does this mean I'm a member of the "oldtimers club"?  How about
a self-synchronizing bit of software to read data at 8kbits/second from
audio tape, written in assembly language, hand compiled to machine
language, entered by toggle switches, then used to record itself on
audio tape?  A bit of swithing magic to get a single bit to toggle
through a speaker to read hex (because I couldn't afford a monitor, and
reading the digital seven segments was too time consuming) while I was
checking my own machine code.  Desperate times called for desperate
measures.

Regards,
Les H