Motion Capture is the process of tracking real life movement in a
3-dimensional space, and recording that movement as a series of XYZ
coordinates. Those 3-dimensional coordinates are then used to move a digital
character or other model, so it exactly follows the original movement.
Usually, humans are the subjects (performers) in motion capture (although
virtually anything could be captured). The resulting movement on an
animated character is extremely lifelike, and reflects all the subtleties
of the captured performance.

_ Motion capture is a tool, its use generates a stream of data which can be exploited as one whishes (as a painter can use his hands, a hammer or even a paintbrush to realize his canvas) .It is all about connections : What are these data attributed to ? (Volumes, images, sounds.)
In summary, this type of tools allows to inject real data (physical constraints) in a virtual universe (as the painter assimilates his vision of the reality to retranscribe it in his universe).

_ The fact that generated data is exploitable in the moment when it is created, allows to show the work in progress. Fact that does not allow the cinema for example (contrary to the 6 traditional arts: painting, sculpture, literature, dance, etc. where creation takes effectively life under the eye of the creator). Furthermore, this technique is called opened in the sense that anybody will be able to generate a coherent quadridimensional data (x, y, z, time).

_ "Virtual" graphic representation in 3 dimensions is a domain of creation which allows to gather various as foresaid arts : modelling, animation and stage setting, visual depiction…
Besides, it is necessary to clarify that the 3D should, generally, be called 4D because, as we, volumes evolve in time. The essential point of this remark is the analogy among "us" and the "3D".

IAs regards the E-GaLab project, 3 main lines of exploitation of data appeared:

_ "Classic" exploitation allowing animation of 3D avatars.
The principal point of this project being distribution and sharing of information in real time on the www, this type of exploitation entails a certain originality : a manipulator generates a movement according to a shape, the data collected by various users will be able to be connected to various avatars (volumes, images, sounds) having no ressemblance with the initial shape. The space of creation remains free of any constraint of interpretation.

_ Exploitation of the magnetic field : the distortions of the universe and the magnetic data being semi-unpredictable (the insertion of metal objects stresses these deformations), these can be connected to elements (volumes, images, sounds) which, by means of constraints and mathematical expressions, will be able to move and evolve in a autonomous way, or more simply will allow to show a magnetic space defines.
A manipulator is not necessary any more for the genesis of the stream of data, only the users who "get back" this stream parameter this exploitation.

_ Exploitation being said " unknown up today ": The collaboration between developers and 3D technicians has allowed to work out one paradigm of a "organico-intuitive" interface (morpho-sensitive, morpho-intuitive) of the software " cls " :
The manipulation of tools and objects, as well as navigation, are made with the movements of the body of the user (equipped with sensors), so without keyboard or mouse.
So arouse the concept of "natural" interface man - machine, creating an actual bridge between the reality and the virtuality.
To control by "natural" behavior a synthetic universe ( natural implying nature receipts or innate, voluntary or involuntary movements, logical or reflexological behavior) develops in the user's an anthropomorphic mode of manipulation of this universe.
This type of concept opens fields to numerous applications as behavioral recognition, assistance of production, scientific research, entertainment, etc.

How does Motion Capture work?

In the case of an “optical” motion capture, a series of high resolution
cameras with special strobe lights are set up around an area to be
“captured.” Small spheres (markers) covered with a retro-reflective substance
are placed in strategic locations on the subject (performer). Through
the use of sophisticated software and precise calibration, the software
can locate these markers by triangulating the various camera views.
The marker locations are recorded as 3-dimensional XYZ coordinates. The
locations are recorded as often as 240 times every second.

The recorded location of these markers in 3D space is often called
“Translational” data. This “translational” information is used directly to
move a computer character when doing face capture. The markers on the
face of the performer correspond to “bones” on a digital character’s
face, which emulate the performer’s actions.

For full body motion capture, another step is necessary. Through the
use of simple measurements, the distance from the markers to the
performer’s actual joint centers are determined. The software then creates a
hierarchy of links between these joint centers. By noting the position
of various markers, the software can determine the angle of these links.
These angles are commonly called “Rotational” data. By combining
translational and rotational data, full human (or animal) movement can be
represented by this hierarchical “skeleton.” In a computer animation
package, this skeleton can be used to drive a 3D character. The result is a
digital character which moves exactly like the performer during the
capture.

Is Motion Capture better than keyframing?

Yes and no. Motion capture is best used in situations where you need
“real” looking motion, and/or where many moves will be applied to a
single character. For multiple moves, the cost advantage of motion capture
can become quite significant.
If you are animating a cartoonish character, and you want your
movements purposely exaggerated, keyframing would be the desired method. We
understand this quite well, because we are also computer animators. We use
both methods where appropriate.