Is it common to make models follow motion data of specific points? To
me it seems a rather large amount of error is introduced, because of
the marker placements may not be in the exact position on the person
as attached in the model, and the obvious difference of scale between
the person and model, even assuming the model was scaled the
proportions may be different. Or is it more common to merely drive
the positions by angles and get the desired general motion? (What I am
currently doing)
Basically I am wondering what is the general practice for modeling a
desired movement which would follow a recorded set of movement data?
To all members: As a curtesy to readers
please identify yourselves when asking a question.
Yes it is a very common to impose measured movements on a model. You
are also quite riht that it mus be done carefully to avoid the types of
errors you mention. It is typically a good idea to place markers on
anatomical landmarks that can easily be identified in the experiment
and also in the model, and the size of the model must be scaled to the
test subject.
When this is done you can obtain very good results. There will be a new
tutorial avalable on precisely these issues in a few days.
Best regards,
John
AnyBody Support.
— In anyscript@yahoogroups.com, “smitswil” <smitswil@…> wrote:
>
> Is it common to make models follow motion data of specific points?
To
> me it seems a rather large amount of error is introduced, because of
> the marker placements may not be in the exact position on the person
> as attached in the model, and the obvious difference of scale between
> the person and model, even assuming the model was scaled the
> proportions may be different. Or is it more common to merely drive
> the positions by angles and get the desired general motion? (What I
am
> currently doing)
>
> Basically I am wondering what is the general practice for modeling a
> desired movement which would follow a recorded set of movement data?
>