Note: Descriptions are shown in the official language in which they were submitted.
CA 02373459 2001-11-07
~~' ~~ o/~~~
METHOD AND SYSTEM FOR SHARING THE VISUAL LIKENESS OF A USER AMONG
A MULTITUDE OF APPLICATIONS OR SERVICES
Technical Field
The present invention relates generally to the field of creating and
maintaining a virtual
identity. By virtual idenfiiy is meant the identity of a user which is adopted
for the purposes of
interacting in a virtual environment, such as a website, computer game, etc.
More
particularly, the present invention relates to creating and maintaining a
virtual identity across
a multitude of network centric or online applications and services, and to the
field of
graphically representing the user as part of their virtual identity.
Background of the Invention
Many online applications and services require the user create an account with
the site in
order to fully exploit the personalized features of the site. As a result, all
user specific data
and customizations are accessed and updated every time the user returns to the
site and
togins with their account name and password. The benefit to the user is
persistence of the
customizations affected during his or her visit to the site. Simply put, users
do not have to
start personalization from scratch every time they visit the site. By signing-
in, the Web
application gives the user access to personal data, which represents the
user's identity with
respect to the application.
The MicrosoftT"" (MS) Passport is probably the best-known example of a virtual
identity
system. All online applications that recognize a user's MS Passport is a
member of the MS
Passport Network. The idea of virtual identity is to provide users with a
unique account
across a multitude of Web applications and services, and across multiple
devices. With a
virtual identity account, a user does not have to remember an account name and
password
for each Web application and service he or she uses on a regular basis. This
becomes the
function of the virtual identity technology employed as the single sign-in
system.
All sites employing or compliant with a single virtual identity system
constitute the network of
applications and services associated with the specific virtual identity
system. Within that
network, users nrgister a single account name and password and then have
access to
personalized services with network applications. The virtual identity system
provider is
15539-1 US Page 1
CA 02373459 2001-11-07
responsible for maintaining and operating the infrastructure that support the
single sign-in
across all applications in the network.
The personal data contained in the MS Passport is generic personal data. This
is useful for
many environments and sites that need this basic level of identity data.
However, the MS
Passport does not address the need for specific identity information required
by
environments that make use of more detailed identity data. Some environments,
such as
those which provide the user with highly personalized environment features,
such as special
interface adaptations and account information, cannot use data from a
centralized sign-in
containing general identify data.
Many applications and services on the Web also employ a persona or avatar,
which could
be that of the user or that of a generic character. Applications such as
virtual dressing
rooms, weight loss programs, instant messaging systems and computer games may
in one
form or another includes a visual representation of the user. In some cases
the application
can allow the user to personalize the avatar with the objective of having it
resemble the
user's own body and/or face. In the context of the virtual dressing room
application, the user
is able to provide values to parameters such as weight, height, shoulder size,
waist size, and
hip size, which together establish the user's morphology, and is able to
render a virtual
model representing the user. An example is the MyVirtualModeIT"" environment
developed
by Applicants' assignee. Afterwards, the user can view how online catalog
garments fit the
model as an aid in deciding on which garments) to purchase.
This type of highly personal information is not presently shared among
different
environments. Furthermore, in the case of a visual representation of a user,
different
environments may either require different likeness data or generate a
different
representation or likeness image based on the environment-specific image
generating
engine. Thus reliance on such virtual identity data cannot be used across
different platforms
without risking offending users who would see their likeness change in a
displeasing manner
from environment to environment.
Summary of the Invention
15539-1 US Page 2
CA 02373459 2001-11-07
The present invention has as one object to provide a system and method that
allow users to
make use of a virtual identity across a number of sites or environments for
the purposes of
generating an image to be associated with the user at the site or environment
without the
drawbacks of the state of the art.
The present invention has as another object to provide a system and method
that allow
users to make use of a virtual identity across a number of sites or
environments having
special virtual identity data requirements without the drawbacks of the state
of the art.
According to a first broad aspect of the invention, there is provided a method
of providing an
image-based likeness of a user in a plurality of environments having access
via a
telecommunications nefwork to a repository of personal likeness data. The
environments
each comprise an image generating engine having rendering characteristics
which results in
a different rendered image based on the likeness data for at least some
instances of the
likeness data. For each of the plurality of environments, the method involves
the following
steps: recognizing the user and retrieving from the repository the likeness
data for the user;
rendering an image corresponding to the likeness data retrieved using the
engine according
to the rendering characteristics; providing a display of the rendered image to
the user at an
interface of the environment; and providing the user with an option at the
interface to edit the
likeness data, wherein the user edits the likeness data when the rendering
characteristics
result in the rendered image at the environment being unsatisfactory to the
user. The
rendering characteristics may comprise image characteristics attributed to
likeness data
parameters unique to only some of the environments. The step of rendering the
likeness
data may also comprise sending a request to a central remote server to render
the likeness
data and obtain the rendered image.
Preferably, the personal likeness data is categorized according to what is
referred to herein
as metaphor, and within each metaphor, profile data for the user defines the
individual
parameters defining the user's personal likeness within the metaphor. Examples
of
metaphors are facial profile, body silhouette, a style of cartoon character
drawing, fantasy
game avatar, and real-to-life form model. The virtual identity likeness data
can contain
metaphor specific data that can only be shared among environments employing
the same
metaphor, and data that can be adapted for a plurality of metaphors.
155.39-1 US Page 3
CA 02373459 2001-11-07
Preferably, the edits to the likeness data are stored in the repository. The
step of rendering
may comprise rendering the image corresponding to the likeness data as mod~ed
by the
stored edits.
The step of providing a display preferably comprises including the rendered
image as part of
an introductory portion of a session in the environment. Alternatively, the
step of providing
the user with an option may comprise providing the user with a selection
between the option
and an introductory portion of a session in the environment.
The repository may be stored at a central virtual identity server, and
alternatively the
repository may be stored in a cookie for each user.
According to a second broad aspect of the invention, there is provided a
method of providing
a virtual identity dataset to a plurality of environments. The method
comprises providing a
central virtual identity server for the environments, the central virtual
identity server
containing a virtual identity data repository, configuring a list of elements
defining a desired
virtual identity dataset for each of the environments, recognizing a user and
requesting the
central virtual identity server to retrieve the virtual identity dataset for
use with a selected one
of the environments and the user, compiling the virtual identity dataset using
data stored in
the repository and in response to the configured fist of elements and to the
selected one of
the environments and the user, and, when required, machine interpretable
instructions for
obtaining additional-information defining one or more of the list of elements
for which the
repository has no data for the user, using the machine interpretable
instructions in a user
interface to obtain the additional information from user, and to complete the
virtual identity
dataset, and providing the virtual identity dataset to the selected
environment.
Preferably, the step of compiling comprises sending the virtual identity
dataset and the
instructions to an environment server of the selected one of the environments.
The virtual
identity dataset and the instructions may be provided in an object executable
by the
environment server.
Preferably, the complete virtual identity dataset for the user is stored in
the repository. The
additional information may be stored for use locally within the selected
environment.
__ _ 15539-1 US Page 4
CA 02373459 2001-11-07
Advantageously, the user may be prompted to define whether the additional
information
provided for the selected environment may be shared by other ones of the
plurality of
environments. More specifically, the user may be prompted to define whether a
field of the
additional information may be shared by other ones of the plurality of
environments.
It will be appreciated that the virtual identity system allows the personal
likeness data of the
user to be shared among applications or environments. In each case, the
application
generates a visual rendering of the data suitable for the context and service
being offered by
the site. Due to this, users provide information about their visual likeness
once for every
instance of visual definition. Each application linked to the virtual identity
system is then able
to retrieve this likeness information and render the appropriate image of the
user. From the
user's perspective, his or her visual identity becomes mobility from
application to application.
The ability to visualize one's self may be either central to the user
experience or may
enhance it tremendously in applications that rely on presenting an image of
the user as part
of the user experience.
In addition, the user can register specific information with the virtual
identity system. This
information may be securely stored online in the virtual identity data
repository and shared
according to sharing agreements when necessary to access a user's personalized
features
of a participating environment, e.g. Web site. The virtual identity account
may be effectively
an aggregate of the users' identity with respect to each application.
Brief Description of the Drawings
The present invention will be better understood by way of the following
detailed description
of a preferred embodiment with reference to the appended drawings, in which:
Fig. 1 is a block diagram of the system according to the prefdrred embodiment.
Fig. 2 is a schematic diagram illustrating a plurality of applications
including a virtual identity
or a likeness representation of a user.
Fig. 3 is a screen from one application including a visual identity, a session
interface and an
edit option selection object, according to the preferred embodiment.
Fig. 4 is a flow chart of the steps according the method of one aspect of the
invention.
Fig. 5 is a flow chart of the steps according the method of another aspect of
the invention.
15539-1 US Page 5
CA 02373459 2001-11-07
Fig. 6 is a model diagram of the software implemented in the central virtual
identity server,
according to the preferred embodiment.
Fig. 7 is a diagram illustrating schematically the sequence and structure of
the system
according to the preferred embodiment:
Detailed Description of the Preferred Embodiment
In one embodiment, the present invention provides a computer system called the
Universal
Visual Identify System {UVIS) design to personifying virtual identity {VI)
systems and
consequently, enabling the mobility of the users' visual identities across Web
applications
and devices. In the UVIS system, a visual identity is called a Virtual Model.
In a preferred
embodiment, applications in a UVIS-enabled network register the visual
descriptor, also
termed likeness data, required by the application's rendering engine to
generate the virtual
model (VM) of the user. Subsequently, when a user signs into the application
with their
virtual identity account for the first time, UVIS queries the users to
establish an answer set of
the likeness data descriptors. This information is instantly stored
temporarily within the
application or permanently in a UVIS-managed online repository. At this point,
the
application typically :displays the resulting image of the user as generated
by the rendering
engine of the application. The user may accept the image as is, or can modify
the image by
editing the initial likeness data parameters. Once the user is satisfied with
his or her virtual
model, the likeness data is stored permanently in a UVIS-managed online
repository.
Afterwards, other applications within the UVIS-enabled network having the.
same rendering
engine will automatically display the same user image whenever the he or: she
uses the
persona-based services of those applications. At any time during a session or
upon repeat
visits to any UV1S-enabled network application the user has the opportunity to
edit their
likeness data and thus impacting the appearance of their virtual model across
the network.
Visual Identity Metaphor
The preferred embodiment includes a specification for defining visual identity
descriptors
that can be common to a visual identity metaphor, an application domain and/or
an
15539-1 US Page 6
CA 02373459 2001-11-07
application instance. As a result, Web applications and services can render
the users' visual
identities in a manner suitable and specific to the application's
requirements.
The term "visual identity metaphor" refers to a target visual concept such as
the user's body
or the user's face. The body metaphor is very important for application
domains such as
gaming and apparel shopping, while the face metaphor is predominantly
important for
makeover and instant messaging domains.
Visual Identity Profile
A Visual Identity Profile refers to a set of descriptors or questions which,
when interpreted by
a target MVM-compliant application, generates a graphical image corresponding
to the
descriptor values or answers supplied by a user. It enables the user to
interact with the
application content. Afterwards, other applications using the same profile
data will
automatically allow users to see their visual selves.
Users with existing Visual Identity Profiles will be queried for additional
descriptor values if
necessary, while new users will be asked to create a profile and provide all
question set
values necessary for the specfic applications' needs. Given the user's set of
answers, the
application will generate the visual image matching the profile.
Mobili
In the preferred embodiment, several application domains can share the same
visual identity
metaphor. This implies Visual Identity Profile descriptors associated with the
metaphor are
employed to ensure a common thread of resemblance to all applications that use
the
metaphor regardless of the domain.
For example, in the Virtual Dressing Room (VDR) application domain (see the
example of
Fig. 3), the visual identity metaphor is the user's body morphology. The same
body
metaphor can be used across VDR applications; but also in other domains such
as gaming.
A user's body parameters registered in the user's Visual Identity Profile is
interpreted by a
VDR application to render an image 18 of the user's body. The same Visual
Identity Profile
parameters can be accepted by a gaming application, allowing the user to see
herself or
himself in action. As a result, mobility can occur across.domains.
Resemblance
15539-1 US Page 7
CA 02373459 2001-11-07
Actual resemblance to the user is an abstraction controlled both by the
information given by
the user as well as by the domain applications themselves. The resulting image
for a given
profile can vary from application to application depending on the rendering
engines available
to the application as well as on the intended use.
In Figure 2; application 14e provides fashion advice based on the user's body
morphology,
while application 14a allows the user to try-on garments. Both images are
different - a
silhouette being sufficient for the purpose of application 14e, while a fully
rendered body is
necessary for application 14e. The same network centric Visual Identity
Profile is interpreted
by both applications.
An application can also generate multiple images based on the same profile and
on the
intended use. For example, for the weight loss program application, different
images can
con-espond to the same Visual Identity Profile. In this case, the images stay
local to the
application and will be re-generated the next time the user fogs in using his
or her Visual
Identity Profile. .
An instant messaging application can use this multiple image feature to
provide its users
with a complete set of personalized facial expressions, body postures and
gestures to define
a more believable visual identity.
Resemblance is thus an abstraction, which is not limited to a single visual
representation of
the user: The important characteristic of MVM VI is that the user controls his
or her visual
identity through the same Visual Identity Profile account.
Visual Identi~~ Information Server
UVIS provides a UVIS client component 24 including an application-programming
interface
(API). This enables applications to communicate with other UVIS network
centric
components in order to achieve the functionality of visual identity mobility
in a seamless
manner.
At the core of the MVM technology implementation resides the Visual Identity
Information
Server (VIIS).
As illustrated in Figure 7, an application provider registers a UVIS compliant
application with
VIIS prior to offering the service online. In this process, re-use of existing
visual identity
15539-1 US Page 8
CA 02373459 2001-11-07
descriptors are identified, while additional ones are defined in order to
account for the
application's specific needs.
VIIS assigns each application a unique ident~er, which consequently enables it
to know
exactly which visual identity metaphor to apply in what context.
The conceptual model of VIIS is descrybed as follows. The Site owner
configures the
information on the left, while the User fills in the information on the right.
The site-speck
interpretation of the VIP occurs when a user signs-in to the site at which
point the
relationships between right and left are enforced and a subset of the
information in selected
from the central repository and sent to the site for rendering. For example,
only the Answers
and AnswerSets in he user's Sign-in that match the sites Questions and
QuestionSets are
selected. All of the other Answers attached to the User's Sign-in are filtered
out.
Visual Identity Applications
The image rendering capabilities of an application can vary from domain to
domain as well
as from application to application. In the first case, a body image can be
static and needs
high quality imaging in the VDR context since the intent is to be photo-
realistic when selling
apparel online. On the other hand, gaming requires real-time 3D where image
quality is not
as important. In the second case, a specific VDR application can render bodies
for virtual
try-on as 2D images, while another VDR application can render bodies as' 3D
images in
order to create an immersed user experience. in all cases, the resulting
graphics is
' . modulated by the same set of visual identity parameters stored in the
Visual Identity Profile
account.
The Visual Identity Profile descriptor set is designed to expand as new
domains and
applications are made available o the MVM Network. The MVM VI specification
provides a
way to define these visual identity domains and parameters so that specific
Web
applications and services can render the users' visual identities in a manner
appropriate to
each and every application's business logic.
This flexibility to be both image and application agnostic is easily
understood by the typical
structure of a visual identity application. Each application includes an MVM
compliant client
that allows it to communicate directly with VIIS in retrieving and updating
Visual Identity
15539-1 US Page 9
CA 02373459 2001-11-07
Profiles. The image-rendering engine contained within the application employs
it own library
of visual assets to construct and display the visual identity image matching
the Visual
Identity Profiles of its users.
The MVM VI specification also provides a means of registering visual assets
with VIIS. The
impact of this feature is to allow VI images generated by one application to
be shared by
other MVM compliant applications. fn Figure 2, the MVM Face Mapping
application 14d
generates 2D or 3D images of the user's face. Once stored in VIIS, other
applications such
as games can make use of this user's asset.
The MVM Virtual Dressing Room
The MVM Virtual Dressing Room (VDR) is an example of a technology based on the
MVM
VI engine. Online apparel vendors create custom versions of VDR to provide
their
customers the service of trying on virtual garments online. In this domain, a
user's visual
identity or Virtual Model is designed to represent the user's body. The Visual
Identity Profile
for the virtual model contains parameters such as weight, height, shoulder
size, waist size,
and hip size, which establish the user's morphology. Once a Virtual Model is
created by the
VDR, users benefit by visualizing how specific garments and outfits fit their
morphology.
The MVM Network
Network centric VI engines allow a multitude of independent applications to
share the same
VI user account. As a result, users need to remember a single sign-in name and
password
and manage and disclose their information once across many applications. The
MVM
Network is composed of all the Web applications that use the MVM VI engine.
Rather than
having to re-create their Virtual Model from application to application, users
create a Model
the first time they use an MVM-enabled application. The Virtual Identity
Profile associated
the model is stored in a network centric repository, which is then accessed by
MVM Network
application providers on behalf of the user-and only with the user's
permission. At any time,
users may choose to share the information needed for a particular product or
service with
their preferred MVM-enabled Web site. MVM securely protects the users' privacy
online. In
Figure 3, this mobility of the Virtual Model is demonstrated for The Virtual
Dressing Room
15539-1 US Page 10
CA 02373459 2001-11-07
application. Within each apparel site; a user will have the same visual
identity derived from
their VIP account with the MVM Network.
Site Development Kit (SDK)
Finally, a part of the MVM VI implementation, a Site Development Kit (SDK) is
provided
which allows and eases the integn~tion of MVM visual identity business logic
(BL) as well as
domain dependent business logic onto a web site application. It is based on
the "tag library"
concept recently introduced. It allows a greater flexibility on the user
interface (U1) and
minimizes coupling between UI and business logic and permits web application
developers
to use the MVM core functionalities without instantiating a single programming
class. This
is accomplished using the MVM Java Server Page Tag Library Application
Programming
Interface (TL-API).
The basic concept behind this TL-API is the use of a series of calls to the
application
business logic to assemble the required information for a given page. Each
application BL
method call returns, when required, a fragment of information formatted into
an XML
document. Each of these fragments is then appended to the page global XML
document
(often refer-ed as the master document). XSL is then used to create and format
the end
result into the client language (HTML for example). Alternative client
language can also be
outputted for more specialized clients such as cell phones, web TV, etc.
The client makes a request to a JSP page that contains an SDK master tag. In
between the
"master" tag, other SDK tags are called in order to access application BL
services. These
services return XML fragments that are maintained in the master XML document
(managed
by the SDK). Once all business sent back to logic requests are done, the
resulting XML is
processed with XSL to generate the HTML that is sent to the client.
The current implementation combines the fD and Likeness Dataset Storage
device, the
Likeness Data Retrieval + Filter device and the Likeness Data Export device
into a
central server called VMIS. VMIS uses a central Oracle database accessed
through a
JAVA interface local to the Likeness Editor device. The local VMIS interface,
called VMIS
15539-1 US Page 11
CA 02373459 2001-11-07
Client communicates across the Internet using the HTTP protocol. WebObjects by
Apple is
used as the application server, and Apache is used as the web server.
VMIS Client uses a class called MobitityServer to establish communication with
the remote
Dataset Storage system. The constructor of the class as documented in the VMIS
Client
JavaDoc is:
MobiIityServer Class Constructor
public MobilityServer(java.lang.String newServerURL,
java.lang.String newSiteName,
java.lang.String newSitePassword)
throws Mobilit)rError,
InvalidLogin,
VersionMismatch,
InvalidDatabaseVersion,
CommunicationError
Constructor that takes the URL of the mobility server and the application site
address.
Parameters:
newServerURL - server URL
newSiteName - site name
newSitePassword - site password
Once the class is successfully instantiated and a communication channel to
VMIS is open,
the method getSignln() is called to obtain an Object that represents a user's
set of
parameters. VMIS uses its knowledge of the local site to filter the user's
dataset and return
only the information pertinent to the site. The Signin object returned by the
method
getSignInQ contains. a filtered version of the user's profile.
GetSignin method of the MobilityServer Class
public Si nln getSignln(java.lang.String loginl~ame,
java.utiLHashtable passwords)
15539-1 US Page 12
CA 02373459 2001-11-07
throws MobilityError,
InvalidLogin,
InvalidSession,
InvalidDatabaseVersion,
CommunicatyonError
This method implements a login of a user account and returns its descriptor:
The password
parameter is a dictionary whose key is the way to access the password and the
contents are
the password value. The group of passwords provided should have an index
greater than or
equal to 100% for the login to be successful.
Parameters:
foginName - usemame
passwords - password dictionary
Returns:
The object Signln related to the user
Throws:
MobitityError, - InvalidLogin, InvalidSession
For VM1S to be able to filter the user's dataset, the owner of the site (owner
of the
environment of the Likeness Editor) must input the information into VMIS.
The conceptual model of VMIS is as follows:
The Site owner configures the information on the left, while the User fills in
the information
on the right. The filtering occurs when a user signs-in to a specific site at
which point the
relationships between right and left are enforced and a subset of the
information in selected
from the central repository and sent to the site. For example, only the
Answers and
AnswerSets in the user's Sign-in that match the sites Questions and
Qu~stionSets are
selected. All of the other Answers attached to the User's Sign-in are filtered
out.
An example of the information available through the methods of the Signln
Object filtered for
a specific site will be described. !n this case the site has 2 domains: UP
(User) and VM
15539-1 US Page 13
CA 02373459 2001-11-07
(Model). The VM domain has 3 question sets each with one question. The A
(Appearance)
Question set also shares the BodyShape question with the F (Fashion Advice)
Question
set.
VM1S also handles the Likeness Data Export device.
Likeness Data Retrieval & Filter
Data is filtered and sent to the Likeness Editor. The filter is a function of
the likeness data
and the environment of the Imaging system. In the current implementation, this
device is
combined with the Storage system.
Likeness Data Export
Data is exported for use with a remote system. In the current implementation,
the data is
serialized into a binary stream using Java classes.
Likeness Editor (3D Model Editor)
Mappers ,
questionnaire
Current Implementation
The current implementation of the Likeness Editor is called a Fashion Server.
It combines
the Likeness Editor and the Rendering Engine. A user signs-in to the
application using
the Identity plug-in. The Identity Plug-in uses VMIS Client to retieve and
filter the user's
likeness dataset. The application then forwards the likeness data to the 2D
Visualisation
plug-in. The 2D visualization plug-in uses a Mapper to map the likeness data
to a series of
parameters that it can then use to reconstruct the likeness of the user with
the available local
resources and the Rendering Engine.
If the user is not satisfied with their Likeness, they can modify it. The user
can modify their
likeness dataset using the Identity plug-in. Each time data is changed by the
user it is re-
forwarded to the 2D Visualisation plug-in where it is re-mapped and re-
displayed. Once the
user is satisfied, the dataset is saved back on VMIS using the Indentity Plug-
in
The API of the fdenity Plug-in is as follows:
15539-1 US Page 14
CA 02373459 2001-11-07
Image Generator PIu4-In API
The following methods available in the Image Generation Plug-in (25) are used
by the Site
Application (14) to render a Likeness Image (18) as illustrated in Figure 1.
1. showVM (IMAGE SIZE): render the image of the current VM to be displayed and
return the image name
2. turnVM (DIRECTION) : change view ID of the current VM
3. changeBackground (IMAGEFILE_NAME) : change the background image
Identfir Plua-In API
The following methods available in the Identity Plug-In (22) are used by the
Site Application
(14) to manage the user's likeness dataset and virtual identity.
Application related:
1. getSitelnfo () : get information about the retailer site
2. getDomains(): return the fist of the domains available for the connected
retailer site
3. getQuestionnaire(domain, questionSet): return the XML Fragment of the
chosen
question set name.
User related:
4. signln (signlnName, password): get the user object from Vmis
5. modifySignln (QUESTION SET) update the modifications of the answers of user
6. createVM() : create a "guest" signln with all VM profiles (Apparence,
sizeSuggestion, Fashion) provide by Vmis for the chosen domain
7. getVM () : get the cun-ent VM in the session after a signln or a create
8. createAnswerSet (VMName, QUESTION SET): create an answer set from a
(Appearance, sizeSuggestion, Fashion)
9. getAnswerSetsStatus () : allow to get infom~ation about the AnswerSets
(filled or not
filled) the signin or guest user has filled.
15539-1 US Page 15
CA 02373459 2001-11-07
10. getAnswers (VMName, QUESTION SET ): to populate an Answer Set
11. modifyAnswerSet (VMName, QUESTION SET): update the modifications of the
answers of VM Answer Set
12. updateAnswerSet (VMName, QUESTION_SET ): update the modifications of the
answers of VM Answer:Set after missingQuestion event
13. saveAnswerSet (QUESTION SET): save one answerSet of the current VM
14. getVMlnfo() : get informations about the currentVM
15. saveProfile () : save all the answerSets of the current VM
16. signOut() : Reset all the user info except the datacollection Info from
the current user
' session.
The Fashion Server application uses XMUXSL technology to build a compelling
user
experience. To enable developers of the user experience to have a simple and
easy to use
entry point to the plug-ins we developed t>1e Fashion Tag Libraries. With this
approach,
developers can build many different user experiences and share a common set of
services.
In the preferred embodiment, the system would function as follows:
The user would create their likeness using the Likeness editor
The user would go to a site
The user would identity themselves
The user would retrieve their likeness data
The system would show the rendered likeness to the user
The user would accept of refuse the likeness
The likeness dataset would be updated
Mobility Scenarios
Case 1: A model going to a site with a more recent render engine
Case 2: A model going to a site with an older render engine
Case 3: A model with parameters that are missing
15539-1 US Page 16
CA 02373459 2001-11-07
Case 4: A model with parameters out of range
Case 5: A model with unsupported parameters
3D Model Imaging
Current Implementation is Compositing - Layering (+alpha). Any off the-shelf
compositing
software could do the job. Our compositor is in C++ and has minimum
functionality to
maximize speed.
Other possible implementations are:
3D rendering
Compositing - Z-buffer (+alpha)
image selection (2D sketches)
Low resolution bitmaps (cell phones)
Vectors (NAPLPS)
ASCII art
Hologram
Model Display
Web site
Cell phone
Kiosk
15539-1 US Page 17
