Simulation of human body in movement

[Rick Johnson]

Nick Mellor <thebalancepro <at> gmail.com> writes:

> I'm looking for a fairly undetailed simulation of the human body walking and
standing. Has anyone had a go at
> this in cgkit or similar?

Hi nick, your question contains too many variables to provide a proper response.
It would equivalent to asking:

 """I need a fairly ubiquitous automobile that can drive and park"""

Okay, but what do you intent to use the automobile /for/? 

If you will be transporting many people on a sleazy tour of overpriced Hollywood
stars' extravagant homes you would probably need a large bus, or if you were
going to help your bum of a brother-in-law move out of the apartment he cannot
afford anymore (and unfortunately into your spare bedroom!) then you may want a
pick-up truck, if however you are going to see your grandma on her deathbed and
the hospital is 2000 miles away, and your absolutely phobic about flying, and
you're on a crusade to save the environment, then you might want a prius! 

The point is: just as there are many types of specialized automobiles for
specialized purposes there are also many "cgkits" for specialized purposes. Are
you wanting to write code that is "close to the metal" like OpenGL; which offers
more control and almost unlimited extensibility, or are you looking for
something very high level, like VPython; which is easy to use but quite limiting.

You also were vague about the Level-Of-Detail. You mentioned "fairly undetailed"
-- does that mean stick figures (composed of simple lines) or "marionettes";
composed of geometric primitives (say cylinders)? Or something more?

And what about the LOD of the hinged joints? Are we talking only articulating
hips (goose stepping)? What about shoulders, elbows and wrist; knees and ankles
and toes? Perhaps something more, perhaps less?

But then again you used the word "simulation" so i am beginning to think you
don't want to actually code the walk loop for specific purposed, you merely wish
to attain a pre-coded simulation? Then why not use a motion-capture like Juhani
suggested? 

You supply the "What Where When and Why" and we'll supply the "How".