FR2870374A1 - Client-server architecture for displaying three dimensional digital model, has server and clients whose network synchronizing modules permit to update scene state relative to model, within communication network, between clients and server - Google Patents

Abstract

The architecture has a server (2) with a network synchronizing module (24). Clients (3, 3a) have objects display modules (30) for representing a three dimensional model and network synchronizing modules (31). The server and the clients are included in a communication network (4) e.g. Internet. The modules (24, 31) permit to update a scene state relative to the model, within the network, between the clients and the server. An independent claim is also included for a method of displaying a three dimensional digital model in a client-server architecture.

Description

CLIENT-SERVER ARCHITECTURE

  VISUALIZATION OF A DIGITAL MODEL

THREE DIMENSIONAL

  The invention relates to a client-server architecture for viewing a three-dimensional digital model (3D) and a method for viewing such a model.

  Many companies design products using computer-aided design (CAD) tools. CAD tools essentially allow to design but also to visualize and interact on 3D digital mockups representing future products.

  These models are used during the various stages of design and validation of products to verify their aesthetic appearance, to simulate their behavior, for example, their impact resistance or their size, to resolve manufacturing constraints, and even to promote products from potential customers or to train staff on these products.

  However, these stages of design, validation, promotion and training, often require the simultaneous intervention of a plurality of users on the models, these users being located in different places, sometimes even in different countries.

  This is why we have developed client-server architectures allowing a plurality of users, physically distant, to view the same model.

  Such an architecture is disclosed in WO 01/84501. This architecture implements a server, a communication network and at least one client. The server comprises, on the one hand, a rendering module configured to render a two-dimensional image (2D) generated from 3D scene data and, on the other hand, an interface configured to transmit these images. to customers, through the network. The weight of the 2D images, that is to say the amount of information they contain in bits, bytes or multiples of these units, is sufficiently small to allow the fast transmission of the images through a weak network. flow and their display at the client level even if the latter does not benefit from a powerful machine. A command / response system allows the client to interact on the image and, in conjunction with the server, on the scene data of that server.

  Also, in the architecture disclosed in WO 01/84501, the client does not have 3D data representative of the scene data so that it is not possible for the user to interact on said data. of scene.

  Other architectures are disclosed in US-6,377,257 and US-6,384,821. In these architectures, the 3D data of a scene, in particular geometric data, is transmitted between a server and a set of clients in a network. The client usually has a rendering module for viewing 3D data.

  Nevertheless, in these architectures, the connection of the clients to the communication network must imperatively be a broad bandwidth connection so as to allow a large flow of geometric data from the server to said clients. In addition, the client machines themselves must have powerful graphics cards with a high quality rendering module to be able to manage these geometric data and give a sufficient rendering to the graphic representations visualized.

  Considering the above-mentioned state of the art, a problem to be solved by the invention is to provide a client-server architecture for the visualization of a three-dimensional digital model, as well as a corresponding method, which allow a set of users to view, interact and ultimately collaborate on shared 3D data within a network.

  The solution of the invention to the aforementioned problem is firstly a client-server architecture for viewing a three-dimensional digital model comprising: a three-dimensional database relating to said model; a server comprising a network synchronization module; and one or more clients, each client comprising a display module for the representation of said model and a network synchronization module, said server and said client (s) being included in a communication network, said synchronization modules being able to put update a scene state relating to the model, within the network, between the client or clients and the server.

  Its second object is a method for viewing a three-dimensional model in a client-server architecture, characterized in that it comprises the following steps according to which: a three-dimensional database relating to said model is provided, a server comprising a network synchronization module, one or more clients, each client comprising a display module for the representation of said model and a network synchronization module, and a communication network; the server generates, from the three-dimensional database, a three-dimensional representation of the model; and - scene states of this three-dimensional representation are updated, within the network, between the client or clients and the server, by means of the synchronization modules.

  Thus, the modifications of the scene states of the three-dimensional model made in a client are updated within the network, via the server, so that the set of users can not only view the model synchronously and coherently, but also interact collaboratively with it and with a high visual quality, regardless of the complexity of the model, regardless of the bandwidth of the network and regardless of the graphics power available to customers.

  The invention will be better understood on reading the nonlimiting description which follows, written with reference to the drawings, in which: FIG. 1 schematizes the client-server architecture according to the invention; FIG. 2 illustrates the division of the calculation of an image into independent layers representing different types of visual effects for the implementation of the invention; FIG. 3 illustrates the different quality levels associated with the Shadow layer shown in FIG. 2; FIG. 4 illustrates the management of the progressive rendering performed for the creation of an image according to the invention; FIG. 5 is an example of level metalanguage used for the definition of layers according to the invention; FIG. 6 illustrates the actions of composition of an image performed by the client display module in an architecture according to the invention, when the parameters of a shadow are modified; FIG. 7 schematizes various steps of a network synchronization according to the invention; FIG. 8 illustrates the decoupling of the complexity according to the invention; FIG. 9 illustrates the remote progressive rendering operated by the server as well as the progressive enrichment locally; FIG. 10 illustrates the collaborative visualization principle according to the invention; FIG. 11 details the various software or hardware modules of a server part of an architecture according to the invention; and FIG. 12 details the various software or hardware modules of a client that is part of an architecture according to the invention.

  As shown in FIG. 1, the client-server architecture according to the invention comprises a three-dimensional database 1, a server 2 and one or more clients 3, 3a distant from the server 2. The server 2 and the clients 3 are connected to each other within a communication network 4.

  The base 1 comprises native three-dimensional data concerning a 3D digital model. This data is stored in a non-volatile memory for example of a hard disk. This hard disk is part of the server 2 or in connection therewith, as in the case of Figure 1.

  The server 2 is of a known type comprising a motherboard provided with at least one microprocessor for data processing, computer cards as well as volatile and non-volatile memory media. This server 2 further comprises, according to the invention, a module 20 for importing and converting native data, a module 21 for three-dimensional generation and the management of generating lightweight 3D representations, managing the attributes of a scene the combination of materials or them and a module 3D rendering 3d progressive to obtain photo-realistic representations of 3D models. The server 2 finally includes a synchronization module 24 server / client virtual spaces.

  Each client 3, 3a is formed by a computer, for example of the personal computer type, comprising a low-resolution lightweight object display module 30 and a network synchronization module 31. A user interface 32 allows a user to interact with the different client modules.

  The network 4 is a network for example of the Internet or private type. The server 2 and the clients 3, 3a have access to this network 4 according to a bandwidth of their own. The data circulating within the network for the implementation of the invention may be encrypted, for example according to the SSL (Secure Sockets Layer) protocol.

  The conversion module 20 of the server 2 imports the native data of the 3D models, which come from existing CAD tools such as CatiaTM, Pro-EngineerrM, VDA-FS ', IGESTM, AutodeskTM, AliasTM or SoftimageTM, translates them, and generates a description of these data in a specific format that can be managed by the detail representation module adapted to a module 22 of 3D such as quality parameters, very high quality and client visualization 30. This specific format is adapted to the needs of the invention and allows in particular a subsequent management levels of detail, a continuous reading, a decoupling of complexity and a combination of materials.

  The details management module 21 of the server 2 generates levels of topological details from the 3D data imported and converted by the module 20. These representations are more or less precise and more or less heavy in terms of memory weight. In particular, the details can be generated from the surface or polygonal information according to a method minimizing the deformations. The 3D representation obtained by the detail management module 21 is thus lightened so as to facilitate, on the one hand, their transfer within the network 5 and, on the other hand, their display by the client machines that do not have capacities. high graphics.

  The scene management module 22 manages the modifications made to the 3D scenes, which relate, for example, to the position of the objects, their rendering attributes such as visibility, color, transparency, texture or reflections, and manages the properties global 3D rendering such as quality, edge effects, fog or optical lens effects.

  The module or rendering engine 23 calculates high quality 2D images from the descriptions of 3D scenes obtained from the module 22 and the topological representations of the 3D models obtained from the module 21.

  For this purpose, it divides the calculation of the image into independent layers representing different types of visual effects, such as shadows, reflection, transparency, volume effects such as smoke and such as lens effects. such as the depth of field. This division or cutting into independent layers is illustrated in Figure 2 with regard to layers of Matter and Illumination, Shadow, Transparency and Reflection. It allows the execution of the calculations of each layer independently and on a plurality of machines, the customers performing the simplest work, the server taking over for the more complex work. It also makes it possible to improve the quality of the image in a progressive manner, this progressive improvement being visible to the user. It finally allows the realization of calculations on each layer according to different methods, more or less fast and / or more or less good quality.

  Moreover, the progressive rendering module 23 defines and manages quality levels in hardware or software for each of the layers and hence manages the calculation time of each of the image layers. Figure 3 illustrates the different quality levels associated with the shadow layer of Figure 2. At quality level 0, the computation time is 40 ms and the projected shadow is of low quality. At Quality Level 1, the computation time is 1 s and the projected shadow is of good quality. At Quality Level 2, the computation time is 10 s and a Shadow Ray roll takes into account the transparency and color of the object. At quality level 3, the computation time is 1 min and the aforementioned spoke run then takes into account the calculation of the penumbra.

  Finally, as shown in Figure 4, the progressive rendering is described by a graph of operations in which each node corresponds to a type of effects, a node that can be calculated with different engines, each engine providing a higher or lower quality depending on its performance.

  Of course, a metalanguage of levels is defined so as to allow the encapsulation of the concepts managed by the classic rendering engines. This metalanguage supports the concept of layered cutting and ensures optimal performance of visual effects. The different variables of this language refer to the layers themselves, to the quality levels available for a given layer and to the effects that can be achieved for that layer. An example of organization of levels according to this metalanguage is shown in Figure 5, for the layers of Matter and Illumination (Illumination Layer), Transparency (Transparency Layer) and Shadow (Shadow Layer). As shown in this figure, parameters for each layer are defined, such as Color, Texture, Shininess. In addition, for each layer, available quality levels such as Quality 0, 1, 2 or 3 (Quality 0, 1, 2 or 3) are defined, as well as pointers such as phong-OGL or phong-software, designating the rendering engines for performing the calculations.

  The visualization module 30 provides local visual support to each client 3, 3a. It allows you to interact with 3D models to change your point of view or edit the rendering properties. This module 30 receives light versions of 3D models in which the quality level is automatically defined according to the power of the host machine. It inherits technologies implemented in the progressive rendering module 23 of the server 2 since it is able to calculate certain levels of low quality depending on the power of the client host machine. In addition, the display module 30 incorporates a 2D image composition engine which makes it possible to assemble in real time the different layers calculated by the progressive rendering module 23. In the example of FIG. shadow settings of a digital mockup. The server 2 recalculates this shadow and returns the result of this calculation to the client 3, 3a. The client 3, 3a then recomposes a new image with this recalculated shadow and the other layers that have not been modified.

  The network synchronization modules 24, 31 update the scene states, not only between a given client and the server, but also between the server and other clients, which allows collaboration between users. FIG. 7 represents the different actions generated within the architecture according to the invention when a user U1 modifies the color of an object. In this case, the modification is transmitted (A) to the synchronization module 31 of the client 3 corresponding to this user U1. This synchronization module 31 informs (B) the synchronization module 24 of the server 2 of the modification performed. The scene manager 22 receives (C) this modification which is subsequently transmitted (D) to the rendering module, which restarts the calculation of the image by taking changes. The synchronization module 24 informs (E) then the other clients 3a of the modification. In particular, the user U2 receives (F) the color change made by the user U1.

  Finally, according to the invention, a specific format of the data has been created which can be managed by the client visualization module. There is therefore a decoupling between the native format of the three-dimensional data and the specific format of the data of the invention. This decoupling allows a user unfamiliar with the CAD languages to interact on three-dimensional models regardless of the format of the original data of this model. Since this format supports detail levels, degraded versions of native data can be generated. This facilitates the transfer of the representative data of the models through the network and the visualization of these models by the users having machines of weak graphical capacities. The decoupling of complexity is illustrated in Figure 8. It appears that the degradation of native data decreases the visualization quality of 3D digital mockups (low resolution image). However, thanks to the high quality of the 3D rendering server, the local and degraded image of the model is enriched, in time, by sending photorealistic 2D pixels calculated on the powerful server. The representation of the model is enriched, on the screen of the user, gradually. The local work at the client level as well as the progressive enrichment of the representation of the model visualized by the client by sending data from the server are illustrated in Figure 9. Finally, thanks to its high quality, the module of server rendering comes in customer support by generating high quality 2D images to gradually enhance this representation. The user sees the image that he visualizes to improve by progressive enrichment.

  Finally, FIG. 10 illustrates the collaborative visualization principle according to the invention. As shown in this figure, users located in different locations can, at the same time, view the same representative image of a digital model such as an eyeglass model or a bodywork model and interact on this model. , collaboratively.

  The following figures, referenced 11 and 12, detail the various software and hardware modules present on the server 2 and the client 3, in a particular embodiment of the invention.

  As shown in FIG. 11, the server 2 comprises a data management module 200, a progressive rendering module 214 and a synchronization module 222.

  The data management module 200 performs the functions described above of the conversion modules 20, the details management 21 and the scene manager 22. For this purpose, it comprises a command management module 201, a data conversion module. 202, a geometric detail level generation module 203, a data model manager 204, a scene manager 205, a session manager 206, a material library manager 207, a user manager 208, a database manager, geometric resolutions associated with the scenes 209, an event manager 210 and a query dictionary 211.

  The scene manager 205 maintains a database of 3D scenes including native files as well as data associated with said files such as resolution levels. The data model manager 204 allows to associate tools specific to each type of data. For example, this manager includes tools that make it possible to convert the data of the Catiarm type to the specific format according to the invention and which make it possible to generate the different levels of geometrical details. The data model manager 204 manages such models. The data conversion module 202 converts the native data to the specific format of the invention. However, this module merely executes the commands managed by the command management module 201 and does not contain data conversion technology sensu stricto. The command management module 201 comprises a script and interpreter module 212 and an external tool manager 213. The script and interpreter module 212 makes it possible to write and execute scripts. The external tool manager 213 manages and references external software that can be called from the scripting language to form complex commands. The geometric detail level generation module 203 generates the detail levels by executing the commands available in the command manager 201. The session manager 206 manages the accesses as well as the associated visualization and interaction processes on a 3D scene. by a user. The user manager 208 manages the users who can participate in the sessions. The library of material libraries 207 manages the libraries associated with the different elements of a scene. The geometric resolutions manager associated with the scenes makes it possible to choose a level of resolution adapted to each client as part of a session. The interaction on a scene generates different events that must be synchronized between the clients 3, 3a and the server 2. The event manager 210 performs this function. These are for example the following events: modification of the point of view, movement of objects, association of materials, edition of the parameters of the materials. The synchronization of the different instances of the same scene between the clients 3, 3a and the server 2 is achieved by sending and receiving requests. A request makes it possible to define an action and contains information describing the type of action, the target, the user having performed the action, etc. The query dictionary 211 makes it possible to define the possible actions. These queries are then translated into events.

  The progressive rendering module 214 comprises a layer manager 215, a quality level manager 216, a task manager 217, a level metalanguage manager 218 and a rendering engine manager 219.

  The layer manager 215 includes a dependency manager between the layers 220 and a module for optimizing the calculations between the layers 221. It manages the execution of the calculation of each effect layer constituting an image. An effect layer, for example, relates to color, shadow, reflections or reflection. Some layers must be calculated before others, for example, the reflection of an object. The dependency manager between the layers 220 ensures the proper scheduling of the calculation of the different layers to obtain a coherent final image. The optimization module 221 analyzes the parameters of a scene to eliminate unnecessary calculations (hidden objects and / or small sizes). The quality level manager 216 makes it possible to define the level of quality for the calculation of the different layers. It makes it possible to manage the sending of the data corresponding to the high quality layers to the clients 3, 3bis connected to a session. The task manager 217 makes it possible to perform the computation of certain layers on remote computers to improve the speed of calculation by involving other computing hardware resources. The level metalanguage manager 218 realizes the syntax and interpretation of this metalanguage. It allows to describe the visual effect of a material taking into account the different layers and processes to calculate the different levels of quality. The rendering engine manager 219 lists all available software and hardware rendering engines and manages these rendering engines for calculating layers and quality levels.

  The synchronization module 222 comprises a flow manager 223, a request manager 224 and a connection manager 225.

  The flow manager 223 manages the transfer of information flows in the network (file transfer or streaming). The request manager 224 makes it possible to receive and send requests between the clients 3, 3bis and the server 2. The connection manager 225 manages the particularly low-level connections within the network between the clients and the server and manages in particular the firewalls and routers.

  As shown in FIG. 12, each client 3, 3a comprises a programming interface (API) 300, a 3D scene graph manager 301, an event manager 302, a request dictionary 303, a composer of layers 304, a layer manager 305, the level metalanguage manager 306, a hardware rendering engine 307, a scene topology manager 308, and a network timing module 309.

  The API 300 is used to build end-user applications and provides access to scene information and control actions. The 3D scene graph manager 301 manages scene graph parameters such as kinship information, camera position, translation parameters, object orientation and size, or object names. The event handler 302 and the client request dictionary 303 have identical functions to the event handler 210 and the server request dictionary 211, respectively. The layer composer 304 performs the composition of the various layers, such as the color layer, the shadow layer or the reflective layer, to obtain the final image. The layer manager 305 and the level metalanguage manager 306 have functions identical to the layer manager 215 and the server level metalanguage manager, respectively. The hardware rendering engine 307 is a rendering engine allowing a quick but approximate calculation of the different layers. The scene topology manager 308 manages the geometric information of the scenes at a resolution level compatible with the graphical capabilities of the client and the bandwidth of the network between the latter and the server. Finally, the network synchronization module 309 of the client is similar to the network synchronization module 222 of the server and therefore comprises, like this module 222, a flow manager 310, a request manager 311 and a connection manager 312 .

Claims (12)

  A client-server architecture for viewing a three-dimensional digital mockup comprising: a three-dimensional database (1) relating to said mock-up; a server (2) having a network timing module (24); and one or more clients (3, 3a), each client having a display module (30) for the representation of said model and a network synchronization module (31), said server (2) and said one or more clients (3, 3a) being included in a communication network (4), said synchronization modules (24, 31) being able to update a scene state relating to this model, within the network (4), between the client or clients (3, 3a) and the server (2).   2. Architecture according to claim 1, characterized in that the server (2) further comprises a data conversion module (20) able to convert the three-dimensional data of the database (1) three-dimensional, a plurality of formats, in a specific format that can be managed by the client visualization module (30).   3. Architecture according to one of claims 1 or 2, characterized in that the module (21) for generating three-dimensional representations of the server (2) is able to generate light representations of said three-dimensional model for facilitated communication within of the communication network (4).   4. Architecture according to one of claims 1, 2 or 3, characterized in that the server (2) further comprises a module (22) for managing scene attributes.   5. Architecture according to one of the preceding claims, characterized in that the server (2) further comprises a rendering module (23).   6. Architecture according to claim 5, characterized in that the rendering module (23) is progressive.   7. Architecture according to claim 5, characterized in that the rendering module (23) is capable of dividing a layered image, a quality coefficient being assigned to each layer.   8. Architecture according to one of the preceding claims, characterized in that a level metalanguage is defined, the metalanguage being able to manage the layers and the levels of qualities associated with these layers.   9. Method for viewing a three-dimensional digital model in a client-server architecture, characterized in that it comprises the following steps according to which: a three-dimensional database (1) relating to said model, a server (2) is provided; ) comprising a module (21) for generating three-dimensional representations of said model and a network synchronization module (24), one or more clients (3, 3a), each client (3, 3a) comprising a visualization module (30); ) for the representation of said model and a network synchronization module (31), and a communication network (4); stage states of this three-dimensional representation are updated, within the network (4), between the client or clients (3, 3a) and the server (2), by means of the synchronization modules (24, 31).   10. Method according to claim 9, characterized in that the three-dimensional data of the database (1) are converted, in a specific format, by a data conversion module (20).   11. Method according to one of claims 9 or 10, characterized in that a rendering module (23) of the server (2) divides the calculation of an image into independent layers each representing a visual effect and manages levels of quality.   12. Method according to one of claims 9, 10 or 11, characterized in that the rendering module (23) generates high quality 2D images which will progressively enrich the representation of the model visualized by the customer.