KR20020013224A - VRML Node Expansion and Processing Method for Progressive Data Transmission in Internet Virtual Reality Service - Google Patents

Abstract

PURPOSE: A method for extending and processing a VRML(virtual reality modeling language) is provided to transmit a virtual space as the unit of a scene in detail. CONSTITUTION: A virtual space is divided as the unit of a scene, which is a detail division unit(S100). List information capable of discriminating a plurality of scenes based on local information positioned by a user is necessary for selecting the corresponding scene positioned by the user out of a plurality of scenes in the virtual space divided as the unit of a scene. Thus, a scene list including scene list information capable of discriminating a plurality of scenes and coordinate information positioned by the scene in the virtual space is created(S200). The scene list is interpreted and scene information of the virtual space positioned by a user is read, and the virtual space is embodied(S300).

Description

VRML Node Expansion and Processing Method for Progressive Data Transmission in Internet Virtual Reality Service

Field of invention

The present invention relates to a method of transmitting data in units of scenes by subdividing a three-dimensional cyberspace on the Internet into units of "scenes." The present invention also relates to a method for extending a scene node, which is a new node that supports the concept of a scene, and a map node that supports the scene node, in order to implement the method. More specifically, the present invention divides a virtual space into a scene which is a smaller unit, and generates a list of scenes representing list information about a scene dividing the virtual space and coordinates in which the scene is located in the virtual space. The server transmits the scene list and the scene information to a client, and the received client interprets the information of the scene list to represent the virtual space where the scene is located. The present invention relates to a method of extending VRML and a method of processing an extended VRML to represent a virtual space.

Background of the Invention

The problem of the client-server system for serving a multi-avatar virtual space in the conventional Internet is that VRML files constituting virtual reality, or virtual world, when a user connects to the server. The whole thing has to be down-loaded to the client. In addition, the user must wait for the process while parsing and rendering the entire file after it is downloaded. This waiting time has been a factor in which the multi-participation virtual reality service is not widely activated by the user.

In order to solve the problem, the present inventors propose a method of subdividing a virtual space into scenes and transmitting the scenes in units of scenes, and a new node supporting a scene concept to realize the method. We want to add a scene node, which is a node, and a map node that supports it, to the VRML. The VRML stands for ISO / IEC 14772-1: 1997 Virtual Reality Modeling Language (VRML97), which was declared as an international standard in 1997, and defines 54 nodes, which are expression units of objects.

It is an object of the present invention to provide a method of subdividing a virtual space into "scene" units and transmitting them in a scene unit in order to solve the problems described above.

It is another object of the present invention to provide a method for extending a VRML by adding a scene node, which is a new node that supports a scene concept, and a map node that supports it, to realize the above method. do.

It is another object of the present invention to provide a method for preventing a time delay when a user waits for all information of a VRML file to be down upon initial connection to a server.

It is still another object of the present invention to provide a method of minimizing memory requirements by parsing information of a scene visible from a user's current location and resident in a memory without parsing the entire virtual space file and resident in the memory.

It is still another object of the present invention to provide a method for providing fast user navigation, since a user does not need to perform collision detection during navigation for an object of an invisible scene.

The above and other objects of the present invention can be achieved by the present invention described in detail below.

1 is a flowchart illustrating a method of designing a VRML file made in a scene unit according to the present invention.

2 is a conceptual diagram illustrating a scene concept of implementing a virtual space.

3 shows a flow for designing a scene node, which is a new node representing a virtual space as a scene that is a subdivision unit.

4 shows a flow for designing a map node, which is a new node defining scene list information.

Figure 5 illustrates the order and method of sending virtual space information to a client accessing a server in a virtual space implemented according to the present invention.

6 illustrates a client access process of a conventional multi-participation virtual reality service.

7 illustrates a client access process of the multi-participation virtual reality service proposed in the present invention.

* Brief description of the major symbols in the drawings *

10: Scene 20: Scene List

30: virtual space

The present invention divides a virtual space into a scene, which is a subdivision unit, and generates a list of scenes representing list information about a scene dividing the virtual space and coordinates of the scene located in the virtual space. The server transmits the scene list and the scene information to a client, and the received client interprets the information of the scene list to represent the virtual space where the scene is located.

The present invention relates to a method of transmitting data in units of scenes by subdividing a virtual space into units of "scenes." In order to implement the above method, a method of designing a scene node, which is a new node supporting a concept of a scene, and a map node, which supports the scene node, is provided. Therefore, since the node concept and the field concept need to be understood, the node and field will be described in more detail below.

First, in general terms, nodes, which are objects handled by the object-oriented language VRML, can have almost any type of information and have children. The node may be reused through instancing. VRML, like typical object-oriented languages, constructs each independent element nonlinearly or randomly. On the other hand, a VRML source file existing inside a scene graph is a hierarchical file describing the order in which the nodes are parsed.

Hereinafter, the node and the field will be described in detail by way of example. Example 1 is a copy of the example from the VRML1.0 specification.

(Example 1) node example

VRML V1.0 ascii

Separator {

DirectionalLight {

direction 0 0 -1

}

PerspectiveCamera {

position -8.6 2.1 5.6

orientation -0.1352 -0.9831 -0.1233 1.1417

focalDistance 10.84

}

}

In (Example 1) above, when a command comes out, it must be followed by '{', followed by other commands, and '}' is wrapped around other commands. All commands take this form. As in the example, a sentence in the form of 'command {...}' is defined as a node in VRML. The node is the basic unit constituting the VRML. Nodes are represented as objects in three-dimensional space, directly visible objects such as cameras, lights, properties of objects such as color, position, and direction, and links of the World Wide Web. In the case of Example 1, it is easy to see that the nodes are Separator, DirectionalLight, and PerspectiveCamera. Nodes defined in VRML are classified into three types, type node, attribute node, and group node, as shown in (Table 1) below. The shape node is a node that designates the shape of an actual three-dimensional object. The attribute node specifies the nature of the object that the shape node will draw. The group node serves to distinguish several nodes so that they can be treated as a single entity.

Classification by node type Node type Subgroup Related Nodes Shape nodes AsciiText, Cone, Cube, Cylinder, IndexedFaceSet, PointSet, Sphere Attribute node Form and shape group Coordinate3, FontStyle, Info, LOD, Material, MaterialBinding, Normal, NormalBinding, Texture2, Texture2Transform, TextureCoordinate2, ShapeHints Transformation matrix group MatrixTransform, Rotation, Scale, Transform, Translation Camera / Light Group OrthographicCamera, PerspectiveCamera, DirectionalLight, PointLight, SpotLight Group node Group, Separator, Switch, TransformSeperator, WWWAnchor Other nodes WWWInline

(Example 2) Field example

Cube {#Node Name

width 0.1 # width field

height 0.1 # height field

depth 4 # depth field

}

The above (Example 2) is for explaining the field, and if you look inside the node, it is different from (Example 1). As such, a command without '{}' cannot exist solely on a node, and is intended to specify a number that determines the nature of the node. This is called a field. If you interpret Example 2, the field consists of the field name and field value. Most nodes will specify one or more fields to represent their attributes. However, some nodes do not have a field, and even if a field exists, the field is sometimes omitted. In this case, the field has a default value. VRML has several predefined numerical systems to represent the numerical systems required for three-dimensional graphics. The numerical system is called a field type, and all fields belong to one of these field types. There are two types of field types, one with one value and the other with two or more values (Table 2).

Type of field type Single object field Multi-Object Field SFBitMask, SFBool, SFmatrixSFColor, SFEnum, SFRotationSFFloat, SFImage, SFLongSFVec2f, SFVec3f, SFString MFColor, MFLong, MFVec2f, MFVec3f,

In the former example, two characters, SF (Single Field) are preceded by the field type in the former case, and two characters, MF (Multiple Field) in the latter case.

(Example 3) Types of Fields

SFFloat 1.0 # Single Object Field with One Value

SFVec3f 1.0 1.2 -0.3 # single value field with multiple values

MFVec3f [ # Multi-object field with three three-dimensional vector elements

1.0 1.2 -0,3

0.2 1.5 -8.7

-2.1 -2.1 0

]

The nodes and fields were described in detail above. Once again, the nodes of the VRML are defined as follows.

[DEF node-name] node-type {[fields ..] [children-node ..]}

Hereinafter, a method of subdividing a virtual space to be implemented in the present invention into "scene" units and transmitting the above-described scene unit will be described.

1 is a flowchart illustrating a method of designing a VRML file made in a scene unit according to the present invention.

First, the virtual space is subdivided into scenes which are subdivided units (S100). In order to divide the virtual space into scene units which are subdividing units constituting the virtual space as described above, the node cannot be represented by a conventional node, thereby newly defining the node. The method of designing the node will be described in detail below.

As described above, in order to select a corresponding scene in which the user is located in a plurality of scenes in the virtual space divided into scene units, list information for distinguishing a plurality of scenes based on region information where the user is located is required. Therefore, in this step, a scene list including list information of a scene capable of distinguishing a plurality of scenes and coordinate information of the scene located in a virtual space is generated (S200). In this step as well, since the conventional node cannot express the scene list, the node is newly defined. The method of designing the node will also be described in detail below.

As described above, the virtual space is divided into a plurality of scenes, a scene list including list information and coordinate space information of the scene is generated, and the scene list is interpreted to represent the virtual space where the user is located. The scene information of the virtual space where the user is located is loaded to implement the virtual space (S300).

Hereinafter, a method of designing a scene node, which is a new node supporting the scene, and a map node, which is a new node supporting the scene list, will be described.

2 is a conceptual diagram illustrating a scene concept of implementing a virtual space.

As shown, in the present invention, the virtual space is divided into the scene 10 which is a subdivision unit. Subsequently, in representing the virtual space, the scene where each scene is located in the virtual space 30 is referred to the scene list 20 to represent the virtual space where the user is located. In the related art, in order to express the virtual space where the user is located, the entire VRML file representing the virtual space is received, parsed, and rendered. On the other hand, according to the above-described scene concept, the user can select and express only the minimum region located in the virtual space. Therefore, it is possible to minimize the size of the VRML data transmitted from the server to the client to represent the virtual space, thereby minimizing the time for parsing and rendering.

As described above, the VRML language is composed of a node representing an object and a field representing a property of the object. 54 nodes are currently defined and used, but in order to support the above-described scene concept, a new node expresses list information and coordinate information of a scene node representing a scene, which is a division unit of a virtual space, and a scene constituting a virtual world. You must design a 'map node'.

3 shows a flow for designing a scene node, which is a new node representing a scene that is a smaller division unit of the virtual space.

A scene is a smaller unit for dividing a virtual space, representing local information located in a portion of the virtual space. Therefore, multiple scenes exist in the virtual space. Therefore, the scene node first assigns a unique ID (id) of the scene to distinguish each scene (S110). Using the nodes and fields described above, if expressed in VRML language, it becomes "field SFInt32 id".

Next, we need to represent the scene that needs to be seen in the current scene as the scene that is adjacent to the scene where the current user is located. Therefore, it indicates the unique identifier of the neighboring scene that needs to be seen in the current scene (S120). When expressed in VRML language using the nodes and fields described above, it becomes "exposedField MFInt32visibleRegions".

Next, the virtual space is divided into a plurality of scenes by indicating an area boundary on a coordinate where the scene is located (S130). Using the nodes and fields described above, expressed in VRML language, it becomes "exposedField MFNode walls".

Next, when the virtual space is divided into a plurality of scenes as described above, an object included in each scene is represented (S140). When expressed in VRML language using the nodes and fields described above, it becomes "exposedField MFNode contents".

Finally, the objects in the scene are initially closed (S150). Using the nodes and fields described above in the VRML language, it becomes "exposedField SFBool closed".

When the scene node made as above is organized in VRML language,

Scene {

field SFInt32 id # unique id of the scene

exposedField MFInt32 visibleRegions # A unique identifier of the scene that needs to be visible in the current scene as a scene (neighbor scene) neighboring the current scene.

exposedField MFNode walls # Area boundary in scene coordinates

exposedField MFNode contents # object in scene

It indicates that the object in the exposedField SFBool closed # scene have been closed in the early

}

Becomes

4 shows a flow for designing a map node, which is a new node defining scene list information.

The map node defines list information of all the scenes that partitioned the virtual space. The list information has a function of a list and a description of all the scenes defined in the virtual space. Therefore, the map node should contain information about all the scenes dividing the virtual space, and represent the area where the scene is located in the virtual space.

Therefore, first, the ID of the scene defined by the scene node is expressed as a list (S210). Using the nodes and fields described above, if expressed in VRML language, it becomes "exposedField MFInt32 ids".

Next, in order to express coordinate information on the virtual space of each scene, information about a normal meaning a direction vector of a plane surrounding the scene is expressed (S220). Using the nodes and fields described above, the expression in VRML language is "field MFVec3f normals".

As described above, the direction of each scene is set, and finally, a distance list indicating a position from the midpoint of the plane surrounding the area to the origin is defined to indicate the area where the scene is located (S230). . In the VRML language, using the nodes and fields described above, it becomes "field MFFloat distances".

If the map node made as above is once again arranged in VRML language,

Map {

exposedField MFInt32 ids # list of ids for scene

field MFVec3f normals # normals list. normal is the direction vector of the plane surrounding the scene

field MFFloat distances #distance list. distance is the position from the midpoint to the origin of the plane surrounding the region

}

Becomes

So far, the new nodes, the map node and the scene node, have been described in detail. Hereinafter, a process of implementing a virtual space with the node added as described above will be described.

FIG. 5 illustrates a sequence and method for sending virtual world information to a client accessing a server in a virtual world implemented according to the present invention.

When the client connects to the server and requests virtual world information (S1), the server lists scenes representing information about the scene partitioning the virtual space and coordinates where the scene is located in the virtual space. list) and the scene corresponding to the entrance of the virtual world are first transmitted to the client (S2). When the client's browser parses this information and renders the information to the user (S3), the user moves the scene space (S4). In this case, additional scenes to be fetched from the server must be determined. For this purpose, the scene list initially transmitted from the server is referred to (S5).

6 illustrates a client access process of a conventional multi-participation virtual reality service. 7 illustrates a client access process of the multi-participation virtual reality service proposed in the present invention. Hereinafter, the present invention will be described with reference to FIGS. 6 and 7.

6 and 7, the client accesses the server to check user authentication (S10 and S'10), and receives a virtual world list from the server to select a virtual world of interest. Same (S20, S'20). Next, in the conventional multi-participation virtual reality service described in FIG. 6, when the client selects a specific virtual world, the server transmits a VRML file for the entire virtual world, and the client parses and renders the file to perform virtual The world will be implemented (S30, S40).

On the other hand, in the multi-participation virtual reality service proposed by the present invention described with reference to FIG. 7, instead of transmitting the entire VRML file of the virtual world selected by the user as described above, the scene list and the virtual constituting the virtual world selected by the user are described. The scene corresponding to the entrance of the world is transmitted first (S'30). Thereafter, the client receives the scene and parses and renders the virtual world to implement and navigate (S'40). As the user navigates, the client requests the server for a new scene by referring to the scene list according to the navigation direction (S'50).

Therefore, in the conventional multi-participation virtual reality service, the entire VRML file constituting the virtual space is transmitted to the client, which consumes a lot of time as the file is parsed and rendered. However, in the multi-participation type virtual reality service proposed by the present invention, it is possible to save time by using only the minimum VRML data to express the virtual world selected by the user.

The present invention does not need to wait for all information of the VRML file to be downloaded when the user initially connects to the server, and does not have to wait for the user to render all the invisible scenes. Rendering performance can be improved by rendering only the scene visible from the user's current location. It also minimizes memory requirements by parsing and storing only the information of the scene visible from the user's current location, without having to parse the entire virtual world and residing in memory, and when the user navigates to invisible scene objects. This eliminates the need for collision detection, enabling fast user navigation.

Claims (4)

Dividing the virtual space into a scene which is a subdivision unit; Generating a scene list indicating list information of a scene dividing the virtual space and spatial coordinate information at which the scene is located in the virtual space; A server transmits the scene list and the scene information to a client; And Receiving this information, the client interprets the information of the scene list to represent the virtual space in which the scene is located; How to extend VRML and process extended VRML to represent multi-participation virtual space. The method of claim 1, wherein dividing the virtual space into a scene that is a subdividing unit comprises: Assigns a unique ID for each scene to distinguish a plurality of scenes from the virtual space; Indicates a unique identifier of a neighboring scene seen in the current scene; Represent an area boundary on a coordinate that divides the virtual space into a plurality of scenes; Expressing an object included in each scene by dividing the virtual space into a plurality of scenes as described above; And Expressing that objects in the scene are initially closed; A method of extending VRML and processing an extended VRML to represent a multi-participant virtual space, characterized in that the step consisting of. The method of claim 1, wherein the generating of a scene list including list information of a scene dividing a virtual space, and representing a region where the scene is located in the virtual space, comprises: Expresses an ID of a scene defined in a scene node as a list; Expressing information about a normal meaning of a direction vector of a plane surrounding the scene; And Expressing a distance list indicating the position from the midpoint of the plane surrounding the area to the origin so that the scene can be located; A method of extending VRML and processing an extended VRML to represent a multi-participant virtual space, characterized in that the step consisting of. The server confirms the authentication when the client connects to the server; After selecting a virtual space of interest to the user, the client requests information of the virtual space from the server; The server preferentially transmits to the client a scene list representing the information about the scene dividing the virtual space and coordinates located in the virtual space of the scene and a scene corresponding to the entrance of the virtual space; The client parses and renders the scene corresponding to the entrance of the virtual space and the scene list received from the server to display to the user; Requesting the scene from the server after determining a scene to be additionally obtained from the server according to a moving direction of the user by referring to a scene list initially transmitted from the server when the user moves a scene space; And The server sends the determined scene to a client; Sending the information of the virtual space to the client connected to the server, characterized in that consisting of steps.