FR2879875A1 - Graphical object e.g. avatar, personalizing method for e.g. portable telephone, involves generating personalization request towards server for displaying personalized object corresponding to combination of group of 3D objects in database - Google Patents

Abstract

The method involves generating a request towards a server for display of a personalized graphical object relative to a combination of 3D objects in a database. A combination identifier is stored in a request identifier database. A 3D object is generated in a 2D database and sent to the terminal, if the identifier does not exist in the identifier database. The object is sent to the terminal if the identifier exists in the database.

Description

METHOD AND SYSTEM FOR IMPLEMENTING AVATAR ON DEMAND

  The present invention relates to the display and transmission to a user of fixed or animated customizable graphic objects.

  The present invention relates more particularly to such a display using avatar type objects. An example of use of the present invention is the display on mobile terminals type mobile phone or PDA.

  The prior art knows many processes where avatars are prerecorded in a database and downloaded at the request of the user. Such solutions are not satisfactory since the size of the databases limits the diversity of available avatars.

  In the field of displaying avatars in a non-predetermined form, the prior art is particularly familiar with PCT application 2004/053799 (SONY Ericsson). In this request, an avatar library is used.

  It includes a set of layers corresponding to parts of the displayable final avatar (eg hair, head, mouth). These parts are preferably stored as an XML document and represent two-dimensional images.

  Such a two-dimensional database is quite limited in terms of modifications available to the user.

  One could therefore think of overcoming these disadvantages by using a database of parts of avatars stored as 3-dimensional objects being generated in real time at a remote server and transmitted to the user on request.

  However, in practice, this method requires computation and loading times that are too long to satisfy a user wishing to download his avatar.

  The present invention therefore intends to solve the problem of avatar loading by using on the one hand a database of 3-dimensional objects that are generated at the request of the user, and on the other hand, a database of two dimensions corresponding to the 3D avatars that have already been calculated.

  The obtained system therefore provides a great freedom of choice avatar customization possibilities through the database of 3D objects, while maintaining an acceptable loading speed in the case where the elements of the avatar had already been calculated at the server level.

  To do this, the present invention is of the type described above and is remarkable, in its broadest sense, as it relates to a method of customizing a graphic object for a user with a client station from a first remote server, characterized in that said first server comprises at least one database comprising a set of elementary 3D personalization objects, and in that it comprises the steps of: generating at said level; client station, a customization request to said server for displaying a personalized graphic object corresponding to the combination of a set of 3D elementary objects of said database.

  registering a unique identifier of said combination in a request identifier database; if said identifier does not already exist in the identifier database: generating at said remote server, in response to said request, at least one 3D object corresponding to said combination; storing in a second database of elements in two dimensions, in the form of a 2-dimensional image, said personalized graphic object corresponding to said combination; - send said graphic object at the client station.

  if said identifier already exists in the identifier database: - sending said graphic object from the database of elements in two dimensions at the client station.

  Preferably, said sending step at the client station is performed by a telecommunication network and is performed by a communication type GSM, GPRS or UMTS.

  Advantageously, said personalized graphic object corresponds to a virtual character (avatar).

  According to a variant, said personalized graphic object 5 corresponds to a part of a scene for the decoration.

  Preferably, said two-dimensional image of the method according to the invention is in a format comprising a depth component.

  The invention will be better understood by means of the following description, given purely for explanatory purposes, of one embodiment of the invention, with reference to the appended figures in which: FIG. 1 represents the general principle of the creation of an avatar on demand from 3D objects; FIG. 2 represents the exchange architecture between a client and a server for obtaining avatars; FIG. 3 represents a simple example of a database used for the invention; - Figure 4 shows an example of menu accessible to the customer; FIG. 5 represents a skeleton according to the invention; FIG. 6 represents the skinning process for the animation of the points of the models; FIG. 7 represents the operating diagram of the invention by the collaboration of a 3D base and a 2D base for the preexisting queries.

  An avatar is theoretically the representation of a user in a virtual environment. The present invention is particularly suitable for this type of representation, but it should be understood that it is equally applicable to any type of representation of objects or set of customizable objects. In the case of avatars, these objects are for example characters to which we add clothes or accessories. In the field of the representation of a living space, these objects are for example furniture to which is added a texture or a finish.

  We first describe the generation of avatars from the database of simple 3D elements. Illustrated in Figure 1, very generally, when the user issues the request meaning that he wishes to load an avatar consisting of elements elt1, elt2, elt3, elt4, elt5 and elt6, this request is sent to the database of 3D objects that calculates the avatar corresponding to this combination. Once the avatar has been calculated, it returns it to the user's terminal.

  We will now describe two embodiments of this stage of generation of an avatar from 3D objects corresponding on the one hand to a deformable human avatar object generated from skeletons and on the other hand to a non-deformable object . Those skilled in the art will however understand that these modes of implementation are not limiting.

  FIG. 2 represents a general client architecture of the mobile telephone type on which the system according to the invention is installed. Once again, it is understood that this embodiment is not limiting since the destination terminal can be a fixed personal computer or mobile accessing the Internet, and more generally, any terminal having a display and remote communication capability. . It can be of type Wap, I-Mode, or www, and also correspond to an interactive terminal on a local network.

  Server 1 includes software and databases necessary for implementation. It includes more particularly a first script (4) for interpreting user requests. This script allows the management of information needed to generate 3-dimensional objects for customization. For example, it manages the importation of objects into a blank scene, position and / or rotation data in 3D space, shading and / or mapping information, camera position and / or zoom information. The skilled person knows the information necessary for the generation of a 3D scene, and it is understood that the previous lite is not limiting. Finally, it should be noted that the script is adapted to the 3D processing software (5) for generating the personalization objects from the information of the script and the object databases (3).

  Figure 3 shows a simple example of a database used for the invention, heads, bodies, and sockets corresponding to different possible customization options.

  According to a first embodiment, the personalization 3D objects are objects that can change shape and position. They are generated from skeletons and are particularly suitable for the realization of human avatars.

  The posing fields correspond to the skeleton of the character on which the different accessories will fit.

  In the field of 3D, the skilled person knows the definition of objects by skeleton and skinning.

  We here simply describe the general principle of skinning for a better understanding of the invention. However, it is understood that this embodiment is not limiting and that the skilled person is able to use another mode of defining the objects. The person skilled in the art, for example, also knows a nearby technology called cloth> for simulating the movement of clothing. This technique is also applicable to the present invention.

  The first step of the creation of the object is to obtain a skeleton S, corresponding globally to a set of bones that will be brought to be animated. This skeleton is therefore a constrained system where we define the roots, the links, and the different constraints on the links. A representation of a skeleton is visible figure 5.

  In a general way, one can call skeleton any object on which points are assigned in a fixed way, the points moving with the displacement of the skeleton.

  From this skeleton, a skinning step is performed to define the 3D model.

  This model is defined by a set of points Mi, fixedly attached to one or more bones of the skeleton.

  The attachment to the different bones is represented by a weighting vector whose sum of weights is 1. The movement of the bones will then cause the movement of the points attached to it, as can be seen in Figure 6.

  The skeleton animations are defined by rotation and translation matrices, which are also applied to the fixed points in the skeletal space.

  We then define a pose of the skeleton as a combination of these matrices of rotation and translation on the different bones.

  Laying at rest is also defined as the skeleton not subject to movement (thus without rotation or translation).

  Finally, animated pose is defined as an image pose of resting pose by a set of geometric transformations.

  It is understood that since the points are fixed in the space of the bones, one can define in the same way a model at rest and an animated model corresponding to the geometrical transformations on the bones seen in the space of the model.

  According to the present invention, the poses are stored in a database of the remote server and different models are associated with them. These models correspond for example to different parts of the body for a human skeleton. The structure of the database is shown in FIG. 2, with for example a set of heads, bodies, and silhouette bottoms.

  The skinning properties make it possible to store only the models at rest, since the animated model will be automatically calculated by the transformations applied to the skeleton.

  Once the model is defined, it is possible to change its texture. Indeed, the 3D model can be meshed, and different textures are applied to this mesh. Once again, only a 2-dimensional texture can be stored, since it will automatically adapt to the model defined by the mapping coordinates. This mapping technique applied to the texture of objects is known to those skilled in the art.

  This change of texture can correspond to a simple change of color or to a change of mapped image or to a change of the physical values constituting the type of texture such as bump, reflection or transparency in a model. We then store the red color, and in case of request of the ur =, ilisateur, this color is applied on the part of the model selected. This possibility of texture change is particularly advantageous in the field of interior decoration, so it is possible for the user to view on request the changes on parts of a 3D scene.

  This notion of texture mapping also allows the user to define particular textures, for example in the form of a photograph. A face photograph can therefore be inserted in place of a default face on one of the 3D models according to the invention.

  To do this, a photograph is placed in a template to define areas of the face, and is uploaded to the server for processing and texturing of the avatar by photography. The personalized avatar with the face corresponding to the photograph is then sent back to the client station.

  The menu proposed to the user is then adapted to this type of definition of the data of the avatar, as in FIG.

  Using a browser or an application such as applet java on his mobile or fixed equipment, the user chooses the type of pose (skeleton). For more user-friendliness, the skeleton is for example presented according to a naked or default model, which can take different forms. The user also chooses specific models corresponding to different parts of the skeleton, as well as possibly a set of textures corresponding to parts of the models.

  Once the user has defined his request, it is sent to the server 1 by a communication network.

  The choice of a pose and models then determines an animated avatar (or avatar at rest If the user has chosen the pose at rest).

  The avatar corresponding to the association of the different models and textures is then displayed on the user's screen.

  If the user decides to change a model, for example if he decides that his avatar has a different haircut, he selects the corresponding feature in his menu and sends his request. The current pose being recorded at the server, the brown is then adapted to the current pose and is sent for display.

  If the user decides to change the silhouette while keeping the same models, they are reloaded to match the new pose. The user can also modify the general scale of his avatar for a better display on his remote equipment.

  It should be noted that the use of 3D models allows great freedom in the animation of custom avatars. For this, the user can optionally define different positions of a model and request the animation of his custom avatar. A program at the server level then calculates an animation over the specified period and sends this animation to the user station. The display of this animation can be done either via a media player (quicktime, flashplayer, windows media ...) or via real-time 3d systems.

  These 3D models can also be viewed from different viewing angles. Requests for camera changes are sent by the client, and the server generates the model under the requested shooting.

  Furthermore, it is understood that the different scenes viewed may also contain a 2D or 3D background to customize the evolution environment of the avatar.

  Finally, the user has access to accessories that can be associated with 3D models. These accessories are for example glasses associated with a face. In this case, the accessory is assigned to the model, as the model was assigned to the skeleton.

  According to a second embodiment, the objects to be personalized may be of the non-deformable type. This embodiment is particularly suitable for interior decoration applications online. In this case, reference objects are no longer defined as skeletons, but as a set of points with position, rotation, and pivot properties in three-dimensional space. The customization is this time performed by mapping methods known to those skilled in the art. As before, all the textures and / or the colors of the objects are modified at the server level on request of the user.

  Moreover, as already mentioned, all of these calculations from 3D objects often require significant processing time incompatible with the demand of users.

  The invention thus also consists, from these calculations on 3D objects, in keeping a database of 2-dimensional elements corresponding to the simple elements constituting the avatar already calculated from the 3D database when a first request.

  This type of process therefore corresponds globally to the combination with 3D processing and 2D learning.

  Illustrated figure 7, during the first request of a combination eltl, elt2, elt3, ..., elt6, the avatar is calculated at the level of the 3D database as it was seen previously. The avatar obtained in its 2D form is then stored in the database of 2D elements. Thus, during a request for a combination of elements, the combination is recognized.

  This combination is for example recorded as a unique identifier at an identifier base for future comparison with the new requests.

  If this combination exists in the 2D database, the corresponding 2D avatar will be sent directly to the user.

  If this combination does not exist, then it is a new combination that must be calculated from the 3D database. This new combination is then, at the end of processing stored in the 2D database for subsequent calls.

  In this way, we keep all the freedom to create new avatars by generating 3D objects, while having a fast computing time for known objects.

  Finally, the 2D images stored in the 2D database are in an IFF format that contains depth information (z depth) for each pixel in the image. It is therefore extremely easy to find the original constitution of the 3d image even if the entanglement of the layers is complex. This IFF format is for example used by the Maya software.

  Those skilled in the art will understand that the method according to the invention is particularly adapted to the virtual representation of an individual in a chat, on a blog, in a communication exchange via a mobile terminal or in an instant messenger. In general, it is applicable in any virtual space.

  The invention is described in the foregoing by way of example. It is understood that the skilled person is able to realize different variants of the invention without departing from the scope of the patent.

Claims (6)

  1. A method of customizing a graphic object for a user equipped with a client station from a first remote server, characterized in that said first server comprises at least one database comprising a set of 3D elementary objects. of personalization, and in that it comprises the steps of. generating, at the level of said client station, a customization request to said server for the display of a personalized graphic object corresponding to the combination of a set of 3D elementary objects of said database.   registering a unique identifier of said combination in a request identifier database; if said identifier does not already exist in the identifier database: generate at said remote server, in response to said request, at least one 3D object corresponding to said combination; storing in a second database of elements in two dimensions, in the form of a 2-dimensional image, said personalized graphic object corresponding to said combination; - send said graphic object at the client station.   if said identifier already exists in the identifier database: sending said graphic object from the database of two-dimensional elements at the client station.   2. A method of customizing a graphic object according to claim 1, characterized in that said sending step at the client station is performed by a telecommunications network and is performed by a communication type GSM, GPRS or UMTS .   3. A method of personalizing a graphic object according to claim 1, characterized in that said personalized graphic object corresponds to a virtual character (avatar).   4. A method of personalizing a graphic object according to claim 1, characterized in that said personalized graphic object corresponds to a part of a scene for decoration.   5. A method of personalizing a graphic object according to claim 1, characterized in that said two-dimensional image is in a format comprising a depth component.   6. A method of personalizing a graphic object according to claim 1, characterized in that it further comprises a mapping step.