Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
CA 02615406 2007-12-19
1
SYSTEM AND METHOD FOR OBTAINING A LIVE PERFORMANCE IN
A VIDEO GAME OR MOVIE INVOLVING MASSIVELY 3D DIGITIZED
HUMAN FACE AND OBJECT
FIELrI OF THE INVENTI(lN
The present invention generally relates to computer animation. More
particularly, it relates to color non-contact optical 3D digitizers, avatars,
immersive environments, and live performances in video games and
movies.
BACKGROUND OF THE INVENTION
3D computer animation became a common tool in creation of video games
and movies postproduction in the last few years. A texture mapped
computer model of almost any object can be created and animated by
using commercially available software applications. A number of input
devices, such as 3D scanner, 3D digitizer and motion capture systems are
used to help to achieve a highly realistic animation. With the dramatic
increase in computing power of easily available computers, high quality
computer animation can now be produced by using an affordable
computer.
On the other hand, the creation of a highly realistic human model is still
carried out by professional computer modelers and animators because of
the technical complexity of such production. Some techniques have been
developed to create an avatar of a real human being. But in general, these
techniques map simply a 2D image (photo) onto a generic model and the
graphic quality of the avatar cannot match the quality of the computer
animation in video games and movies.
More and more immersive media environments are created to involve
gamers and spectators in a new show experience. The gamers and
CA 02615406 2007-12-19
2
spectators become part of the performance although the integration of the
live performers into a computer animation is still a very time consuming
process and with very limited functions.
Recently, a new device, an automatic 3D Photo Crystal Booth, has begun
to be introduced to the consumer market. Some of the booths are already
implanted in theme parks, family entertainment centers and shopping
malls. This 3D photo crystal booth can capture a 3D model of a human
portrait by using a real 3D digitizer and engraves the 3D portrait in a glass
cube. So far, the 3D model of a human face captured in this automatic
booth is not ready to be animated because its mesh is not structured based
on facial features. In addition, the captured 3D data are still stored in a
local computer although most of these booths have an internet connection
for remote maintenance purposes.
SUMMARY OF THE INVENTION
Statement of the Objects of the Invention:
This invention consists of a whole process to massively bring a digitized
3D human face in the form of an avatar along with 3D objects into an
immersive environment for video games and movies. It covers a concept to
bring video gamers and spectators to a real time live performance in video
games or movies that are produced by using both traditional filming
techniques and computer animation.
Summary of the Invention
The first part of this invention consists of obtaining a virtual 3D model of
some objects (i.e. in computer format) that represents some part of the
human body (usually, the face). These virtual 3D models (including among
other information the 3D geometry and color overlay information they
CA 02615406 2007-12-19
3
possess) are acquired by using a 3D optical imaging device. This device
can be integrated in a fully automatic 3D Photo Crystal Booth or it can be
an automatic, semi-automatic or manually operated stand-alone 3D
digitizing system. The captured 3D model will be saved in a commonly
accepted computer format. In order to make these data accessible from
any future locations where a live performance of a video game or a movie
will take place, it is important to archive these captured 3D data in a
centralized server through an internet connection already available at the
3D photo Crystal Booth or the data processing center.
The second part of this invention is a process to convert the captured 3D
model in a model form that can be animated. A software tool will help to
perform the conversion, which might consist of morphing of a generic pre-
arranged mesh structure to a captured 3D model and mapping the
captured color texture to a pre-arranged UV mapped texture, or some other
process that can achieve a similar result. This process needs to be done in
a very easy way so that an untrained operator can perform it within a very
short time.
Once the model is ready to be animated, it will be brought to a pre-
animated scene where an avatar will be integrated in a video game or a
movie either by real time rendering (controlled by an interaction of the
gamer) or a pre-rendered process.
A non-restrictive description of a preferred embodiment of the invention will
now be given with reference to the appended drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
Fig. 1 is a graphic representation of the 3D capture of a human face using
a stand-alone 3D digitizing device;
Fig. 2; is a graphic representation of the 3D capture of a human face in a
fully automatic 3D photo crystal booth;
CA 02615406 2007-12-19
4
Fig. 3 illustrates a typical conversion of the captured model into an avatar
that can be animated; and
Fig. 4; is a diagram showing the process to bring a gamer or a spectator
in~~ ~n imm~rcii~ nrmri+rr ~nnr~i-.r~~
iiiw aii nliiiicioivc ciivliviriimi~iciii
DESi.KiP T iON OF PREFERRED EMEODiMEN T S
The process of involving the gamer or the spectator into a performance in a
video game or a movie consists of the following elements and steps:
First, the face, or another part of the body or an object, of a video gamer or
a spectator needs to be captured by a 3D imaging device prior to the
performance or, at least, prior to the moment in the performance when the
computer processed 3D model is deemed necessary. The 3D capture can
be carried out in a fully automatic 3D photo crystal booth as shown in fig.1,
or can be completed by using a stand-alone 3D digitizing device (fig.2), as
desired by, or depending on the method available to, the gamer or
spectator.
The captured 3D data are either saved in a local computer or archived into
a centralized server through an internet connection. These data can be
retrieved some time later at a location where an internet connection is
available, through a process that will depend, among other things but not
restricted to, on the type of connection, the location of the server or the
type of data storage.
A 3D model of the acquired object, be it the face, another body part or an
object, is then created from the captured 3D raw data by a computer
process designed for this purpose, and is then saved. This 3D model is
converted to a model that can be animated by applying a pre-arranged and
pre-mapped generic model, as shown in Fig3. The software used for this
CA 02615406 2007-12-19
processing can be a standalone software, a web based software, or part of
an existing software compatible with the data format and capable of
carrying out the remapping.
Finully the ?rl mndel thut hueheem m~ulifieuJ in ho eacy fn unimu4o i~
5 integrated in the desired computer animation scenes, which could consist,
among other things but not limited to, in a video game or a performance.
Tha animatinn cranac ;yhirh rnntain tha a;iatar r)f tha %;iclen gamer or
spectator could be rendered, preferably but not necessarily, in real-time
with the gamer or the spectator interaction, or pre-rendered to show a
performance which includes animation of the avatar.
Fig. 4 illustrates the whole process.
Although the present invention has been explained hereinabove by way of
a preferred embodiment thereof, it should be pointed out that any
modifications to this preferred embodiment is not deemed to alter or
change the nature and scope of the present invention.
