KR100695156B1 - Method for converting graphics image of objects and apparatus therefor - Google Patents

Abstract

According to the present invention there is disclosed a method and apparatus for converting a graphical image of an object and a computer readable recording medium having recorded thereon a program for performing the method.

According to the present invention, a method for converting a graphic image of an object consisting of at least one triangle comprises: determining at least one triangle to be collapsed among the triangles constituting the object, three vertices of the triangle to be collapsed and the Determining the collapse peak to be produced by the collapse of the triangle by calculating the collapse peak based on the tangent plane of three vertices, based on the determined triangle to be collapsed and the information about the determined collapse vertex Converting the graphical image of the object. According to the configuration of the present invention as described above, rendering of a large three-dimensional object can be efficiently performed while saving resources.

Description

Method and converting graphics image of objects and apparatus therefor}

1 is a reference diagram for explaining an LOD according to the prior art;

2 is a reference diagram for explaining a concept of LOD according to the present invention;

3 is a schematic block diagram of an LOD implementing apparatus according to the present invention;

4 is a detailed configuration diagram of the preprocessing unit illustrated in FIG. 3;

5 is a flowchart illustrating a process of a LOD rendering method according to the present invention;

6 is a schematic block diagram of an apparatus implementing the LOD method in VRML according to the present invention;

FIG. 7 is a detailed configuration diagram of the preprocessing unit illustrated in FIG. 6;

8 is a preprocessing operation performed in the preprocessing unit shown in FIG.

FIG. 9 is a rendering operation process performed by the rendering unit illustrated in FIG. 7;

10A to 10D are reference diagrams for explaining a concept used for calculating an error metric according to the present invention;

11 is a reference diagram for explaining a concept of creating a collapsed vertex by collapse of one triangle according to the present invention,

12 is an operational flow diagram of collapse of one triangle to produce a collapsed vertex in accordance with the present invention;

13A to 13F illustrate an example of a process of creating a collapsed peak by collapsing one triangle according to the present invention;

14 is a reference diagram for explaining a method of obtaining texture coordinates of a collapsed vertex according to the present invention;

15 is a reference diagram for explaining a method for obtaining a normal vector of a collapsed vertex according to the present invention;

16 is a reference diagram for explaining a method for obtaining a color of a collapsed vertex according to the present invention;

17 is a reference diagram for explaining a seamless rendering according to the present invention;

18 is a reference diagram for explaining a distance relationship between a camera and a 3D mesh in seamless rendering;

19 is a view for explaining the gradual deformation between LOD level k-1 and LOD level k,

20 shows an embodiment of Geo-Morphing,

21 is a node syntax according to an LOD method according to the present invention;

22 is a structure of the lodSequence shown in FIG. 21,

FIG. 23 is a reference diagram for describing an updateDist illustrated in FIG. 21;

24 is a reference diagram for describing minScreenArea illustrated in FIG. 21;

25 is a reference diagram for explaining converting an original VRML model to a VRML model according to the present invention;

26A is an example of an object consisting of two or more triangles,

FIG. 26B is an original VRML model for the object shown in FIG. 26A; FIG.

FIG. 26C is a VRML model according to the present invention performed on the object shown in FIG. 26A. FIG.

The present invention relates to a method and apparatus for converting a graphical image of an object, and to a computer readable recording medium having recorded thereon a program for performing the method.

LOD stands for Level of Detail.

1 is a reference diagram for explaining an LOD according to the prior art.

Referring to Figure 1 (a), three cones are shown. The leftmost cone is not actually a cone, but rather a pyramid with four sides. The middle cone has eight sides, and the right cone has sixteen sides. In the cones shown in (a), the three cones can be distinguished from each other by the direction in which the light is reflected.

Referring to FIG. 1B, the cones show the cones seen at a position farther away from the user. Among the cones shown, the right and center cones can be identified from the left cone, but the difference between the right and center cones is difficult to clearly identify.

Referring to FIG. 1C, the cones show the cones seen at a further distance from the user. In (c), it is hard to tell the difference between the three cones.

This sequence of images shows that it is more difficult to identify the details of an object over distance. Thus, when representing an object that is very far from the user, it indicates that there is no need to draw the object complicated. As the shape of an object becomes more complex, more tasks are required to draw the object, which means that you don't have to draw the object in detail when you indicate that the object is far from the user. This represents the level of detail.

3D data is so large that it is difficult to render effectively on limited resources. Therefore, progressive rendering considering the user's view is required.

The present invention solves the above problems and provides a method and apparatus for converting a graphic image of an object capable of rendering a three-dimensional object effectively in a limited resource, and a computer-readable recording medium on which a program for performing the method is recorded. It aims to do it.

One feature of the present invention for solving the above problems is, in a method of converting a graphic image of an object consisting of at least one triangle, determining at least one triangle to collapse among the triangles constituting the object And determining a decay vertex to be generated by the collapse of the triangle, and converting a graphical image of the object based on the information about the determined triangle to be collapsed and the information about the determined decay vertex. It is.

The determining of the triangle to collapse involves determining an indicator that has less influence on the representation of the object when one triangle constituting the object is removed, and determines the triangle to collapse in the object according to the indicator. It is preferable to include the step of.

The determining of the collapse peak may also include calculating the collapse peak based on the three vertices of the triangle to be collapsed and the tangent plane of the three vertices. In addition, the decay vertex determination step, it is preferable to include the step of calculating the attribute information of the decay vertex based on the three vertices of the triangle to be collapsed.

Another aspect of the invention is a method for adaptively converting a graphical image of an object, comprising information on the vertices constituting the object and the triangles constituting the object to represent an object consisting of one or more triangles. Receiving a graphical model containing information about the triangle, determining a triangle to collapse among the triangles constituting the object and a collapse peak generated by the collapse of the triangle, the information about the triangle to collapse and the collapse peak Generating a graphic model containing information about, and performing a rendering while converting a graphic image of the object based on the generated graphic model.

The generating of the graphic model may include defining an error metric representing an index in which the triangle constituting the object affects the representation of the object, and an error value for each triangle constituting the object based on the error metric. It is preferable to include the step of obtaining a and determining the order of the triangle to collapse and the triangle to collapse based on the obtained error value.

Another aspect of the present invention is a method for adaptively converting a graphical image of an object consisting of at least one triangle, the error metric defining an indicator indicative of the triangles constituting the object affecting the representation of the object. And calculating an error value for each triangle constituting the object based on the error metric, and first collapsing the triangle having the smallest error value among the obtained error values.

Defining the error metric preferably includes defining the error metric as a sum of an error relating to a geometric characteristic of the triangle and an error relating to an attribute characteristic of the triangle.

Here, the error regarding the geometric characteristics of the triangle is determined in consideration of the radius of curvature of the triangle, the area occupied by the triangle in the whole mesh, and whether the triangle is a triangle at the boundary, and also the properties of the triangle The error regarding the characteristic is preferably determined in consideration of the amount of change in color, the amount of change in normality, and the amount of change in texture coordinate value between the triangle and triangles adjacent to the triangle.

In addition, the collapse of the triangle may include generating a collapsed vertex to replace the triangle, changing a connection relationship of triangles constituting the object based on the collapsed vertex, and based on the changed connection relationship. Preferably generating a graphical image of the object.

Another feature of the present invention is a method for decaying a triangle contained in an object represented by a graphic image composed of one or more triangles into vertices, each of which has a normal to the mesh surface of the object for three vertices of the triangle to be collapsed. Defining, for each of the three vertices of the triangle, respectively defining a tangent plane that is a plane that is perpendicular to the normal and includes a vertex, and the center points of the three vertices of the triangle on the three tangent planes, respectively. Projecting to obtain three projected points, and obtaining a collapsed vertex by calculating an average of the three projected points.

The collapse method preferably further comprises obtaining an attribute value of the collapsed vertex based on an attribute value of the three vertices of the triangle to be collapsed.

In the obtaining of the attribute value, the attribute value may be calculated using the average value of the three vertices or reflects the characteristics of the three vertices.

In the case of calculating an attribute value reflecting the characteristics of the three vertices, the texture coordinate value of the collapsed vertex is most affected by the texture coordinate value of the vertex located closest to the collapsed vertex among the three vertices of the triangle to be collapsed. Obtain a normal vector value of the collapsed vertex so as to be most affected by the normal vector value of the vertex located closest to the collapsed vertex among the three vertices of the triangle to be collapsed, or the color of the collapsed vertex You can preserve the color of the three vertices of the triangle.

Another feature of the present invention is a method for generating a graphical model for adaptively converting a graphical image of an object consisting of at least one triangle, the triangle being collapsed from the object, the order of triangles being collapsed and the collapse of the triangle. Determining a decay vertex to be generated by: generating a triangle identifier node in which the information about the triangles is arranged in the order of the triangle to be collapsed; and a level-of-detail sequence having the order information of the triangle to be collapsed; Creating a node, generating coordinate nodes in which coordinate information of vertices constituting the object and coordinate information of the collapsed vertices are arranged, and decaying vertices indicating positions of the collapsed vertices arranged in the coordinate nodes Creating a location node.

The determining may include obtaining an error value for each triangle constituting the object based on an error metric representing an index in which the triangle constituting the object affects an expression of the object, and the triangle with a small error value is obtained. It is preferred to include the step of determining the order of triangles to be collapsed so as to collapse first.

The method comprises the steps of creating an update node indicative of the extent to which the triangle collapses progressively between one level and the next level at which the triangle collapse is performed, and the level information at which to start the triangle collapse in the original object. Preferably, the method further includes generating the representing node.

In addition, the method generates a color node of the vertices constituting the object, a color node in which the color information of the decay vertex is arranged, and a collapsing vertex color node representing the position of the color information of the decay vertex arranged in the color node. Generating a normal node of the vertices constituting the object, a normal node in which normal information of the collapsed vertices is arranged, and a collapsed vertex normal node representing positions of normal information of the collapsed vertices arranged in the normal node; And decay vertex textures indicating the texture coordinate information of the vertices constituting the object, the texture coordinate node in which the texture coordinate information of the decay vertex is arranged, and the texture coordinate information of the decay vertex arranged in the texture coordinate node. Preferably, at least one of the steps of generating the coordinate node is included.

Another feature of the present invention is a VRML model generation method for adaptively transforming a graphic image of an object consisting of at least one triangle, the triangle to collapse from the object, the order of triangles to collapse and collapse of the triangle Determining a decay vertex to be generated by: generating a coordIndex node that arranges information about the triangles in the order of the decayed triangle; and creating a lodSequence node having the order information of the decayed triangle. Generating a coord node in which coordinate information of vertices constituting the object and coordinate information of the collapsed vertices are arranged, and generating a coordStart node indicating a location of the collapsed vertices arranged in the coord node. It will include.

The VRML model generation method includes: a normalIndex node in which information about the triangles are arranged in order of the collapsed triangle, normal information of vertices constituting the object, and normal node in which normal information of the collapsed vertex is arranged; generating a normalStart node indicating a position of normal information of the collapsed vertices arranged in a normal node, a colorIndex node in which information about the triangles is arranged in order of the collapsed triangle, and color information of vertices constituting the object And generating a color node in which color information of the decay vertex is arranged and a colorStart node indicating a position of color information of the decay vertex arranged in the color node, the information about the triangles in the order of the decayed triangle. A texcoordIndex node in which the coordinates are aligned, texture coordinate information of the vertices constituting the object, and the decay information. Is the texture coordinates that information further comprises an array texcoord node, at least one of generating a texcoordStart node that represents the location of the texture coordinate information of the collapse vertices arranged in the texcoord nodes is preferred.

Another feature of the invention is an apparatus for converting a graphical image of an object consisting of at least one triangle, comprising determining at least one triangle from among the triangles constituting the object and decaying the triangle A preprocessing unit determining a collapsed vertex to be generated, and a rendering unit converting a graphic image of the object based on the information about the collapsed triangle determined by the preprocessing unit and the information on the determined collapsed vertex. .

In still another aspect of the present invention, in an apparatus for adaptively converting a graphic image of an object, information about vertices constituting the object and triangles constituting the object to represent an object composed of one or more triangles Receive a graphical model containing information about and determine among the triangles constituting the object a triangle to collapse and a collapse peak generated by the collapse of the triangle, and the information about the triangle to collapse and the collapse peak. It includes a pre-processing unit for generating a graphical model containing the information, and a rendering unit for performing a rendering while converting the graphical image of the object based on the generated graphical model.

Another feature of the present invention is an apparatus for adaptively converting a graphical image of an object consisting of at least one triangle, the error metric defining an indicator in which the triangles constituting the object affect the representation of the object. And a preprocessing unit obtaining an error value for each triangle constituting the object based on the error metric, and rendering the object while first collapsing the triangle having the smallest error value among the obtained error values. It includes a rendering unit.

Another feature of the present invention is a graphic model generating apparatus for adaptively converting a graphic image of an object consisting of at least one triangle, wherein the triangle decays from the object in consideration of the degree of influence on the representation of the object. Information about the triangles in order of the triangle to be collapsed, an error value calculation unit to determine the order of the triangles to collapse, a collapsed vertex calculation unit to determine the collapsed vertices to be generated by the collapse of the triangle, and information about the triangles in the order of the triangles to collapse. An aligned triangular identifier node, a level-of-detail sequence node having sequence information of the collapsed triangle, coordinate information of vertices constituting the object, coordinate nodes on which the collapsed vertex coordinate information is arranged, and arranged in the coordinate node And a graphic model generator for generating a collapsed vertex position node representing the collapsed vertex position. Will.

According to another aspect of the present invention, in the VRML model generating apparatus for adaptively converting a graphic image of an object composed of at least one triangle, the triangle collapses from the object in consideration of the degree to which the triangle affects the representation of the object. Information about the triangles in order of the triangle to be collapsed, an error value calculation unit to determine the order of the triangles to collapse, a collapsed vertex calculation unit to determine the collapsed vertices generated by the collapse of the triangle, and information about the triangles in the order of the triangles to collapse. An ordered coordIndex node, a lodSequence node having the sequence information of the collapsed triangle, coordinate information of the vertices constituting the object, coordinate information of the collapsed vertex, and a coord node arranged with the collapsed vertex arranged in the coord node. It includes a graphical model generator that creates a coordStart node representing a location.

A further aspect of the invention is a computer readable recording medium having stored thereon a program for performing a method for converting a graphic image of an object consisting of at least one triangle, the method comprising collapsing among the triangles constituting the object. Determining at least one triangle to be formed, determining a collapsed vertex to be generated by the collapse of the triangle, and based on the information about the determined triangle to be collapsed and the information about the determined collapsed vertex Converting the image.

Another aspect of the invention is a computer readable recording medium having a program recorded thereon which performs a method for adaptively converting a graphic image of an object, said method comprising: said object for representing an object consisting of one or more triangles; Receiving a graphical model containing information about the vertices constituting the object and the triangles constituting the object, the triangle to be collapsed among the triangles constituting the object and generated by the collapse of the triangle Determining a collapsed vertex, generating a graphical model containing information about the collapsed triangle and information about the collapsed vertex, and performing a rendering while converting a graphical image of the object based on the generated graphical model It is to include a step.

Another aspect of the invention is a computer-readable recording medium having a program recorded thereon which performs a method for adaptively converting a graphical image of an object consisting of at least one triangle, the method comprising: a triangle constituting the object; Defining an error metric indicative of an indicator affecting the representation of the object, obtaining an error value for each of the triangles constituting the object based on the error metric, and having the smallest error value among the obtained error values The first step is to collapse the triangle with.

Another aspect of the present invention is a computer-readable recording medium having a program recorded thereon that performs a method of collapsing a triangle included in an object represented by a graphic image composed of one or more triangles to a vertex. Defining the normals to the mesh surface of the object for each of the three vertices of the triangle to be defined, and for each of the three vertices of the triangle, defining tangent planes, each of which is a plane perpendicular to the normal and including the vertices; Projecting center points of the three vertices of the triangle to the three tangent planes, respectively, to obtain three projected points, and obtaining a collapsed vertex by calculating an average of the three projected points. will be.

Another aspect of the invention is a computer-readable recording medium having a program recorded thereon that performs a method for generating a graphic model for adaptively converting a graphic image of an object consisting of at least one triangle, the method comprising: Determining triangles to collapse from, an order of triangles to collapse, and decay vertices to be generated by the collapse of the triangle; and generating a triangle identifier node in which information about the triangles is arranged in the order of the triangles to collapse. Generating a level-of-detail sequence node having the sequence information of the triangle to be collapsed, generating coordinate information of coordinates of vertices constituting the object and coordinate information of the collapsed vertex; A collapsed vertex position node representing the location of the collapsed vertex arranged at the coordinate node. It comprises the step of sex.

Another aspect of the present invention is a computer-readable recording medium having a program recorded thereon that performs a method for generating a VRML model for adaptively converting a graphic image of an object consisting of at least one triangle. Determining from the object a triangle to collapse, an order of triangles to collapse, and a decay vertex to be generated by the collapse of the triangle; and creating a coordIndex node that arranges information about the triangles in the order of the triangle to collapse. Generating a lodSequence node having sequence information of the triangle to be collapsed, generating a coord node in which coordinate information of vertices constituting the object and coordinate information of the collapsed vertex are arranged; And generating a coordStart node representing the location of the collapsed vertex arranged at.

The present invention will now be described in detail with reference to the accompanying drawings.

2 is a reference diagram for explaining the concept of the LOD according to the present invention.

On the left side of FIG. 2 is shown an original object consisting of ten triangles, and on the right is a deformed object consisting of six triangles by collapsing triangles from the original object.

In order to simplify and deform the graphic image of an object composed of triangles, the present invention is intended to collapse a triangle into vertices. Referring to FIG. 2, the original object 210 is composed of ten triangles T1 to T10, and when the triangle T1 is collapsed, the triangle T1 is composed of three vertices V1, V2, and V3. The triangles T2, T3, and T4, which share two vertices of V2 and V3 together, also collapse together. Therefore, when the original object 210 collapses the triangle T1, the triangles T2, T3, and T4 collapse together, resulting in the deformed object 220 composed of the triangles T5 to T10 because all four triangles T1 to T4 collapse. .

That is, the object deformation method according to the present invention is to collapse one triangle to generate one collapsing vertex. One such collapsed vertex becomes vertex V4 in the modified object 220 of FIG. 2.

3 is a schematic block diagram of an LOD implementing apparatus according to the present invention.

Referring to FIG. 3, the LOD implementation apparatus 300 includes a preprocessing unit 310 and a rendering unit 320.

The preprocessing unit 310 receives information about the object to be rendered, performs preprocessing for the LOD, and provides the information about the LOD to the rendering unit 320 together with the information about the object.

The object represented by the graphical image according to the invention consists of vertices and triangles. In other words, vertices exist as the most basic elements constituting an object, and a triangle or face is represented as a connection relationship between three vertices. In general, the surface constituting the object may be represented by various polygons, but the present invention considers triangles. The preprocessing unit 310 receives the information about the vertices and triangles constituting the object to be rendered, information about the triangle, and attribute information as optional information, and determines the order of triangles to collapse in the object. The information about the collapsed vertex and the property of the collapsed vertex generated by the collapse of is calculated and used as information for the LOD. That is, the preprocessing unit 310 provides the rendering unit 320 with information about the vertices constituting the object and the collapsed vertex information, the triangle information, the order information of the triangle to collapse in the object and the attribute information of the collapsed vertex.

The rendering unit 320 receives the information about the object to be rendered and the information for the LOD from the preprocessing unit 310 and outputs the graphic image to be displayed by rendering the object while performing the deformation of the object based on the information for the LOD. do.

4 is a detailed configuration diagram of the preprocessing unit illustrated in FIG. 3.

Referring to FIG. 4, the preprocessing unit 310 includes an error value calculator 311, a collapsed vertex calculator 312, and a collapsed vertex attribute information calculator 313.

The error value calculator 311 calculates an error value for all triangles constituting the object based on the error metric. The error metric refers to an index indicating how much a triangle affects the representation of the object in order to find a triangle that less affects the representation of the object when one of the triangles constituting the object to be rendered is removed. do. Since it is desirable to collapse a triangle that does not affect the representation of the original object even if it is removed from the triangles that make up the object, it is best to determine the error metric to find a triangle that does not significantly affect the representation of the object. It is important. How to determine the error metric will be described in detail below.

When the error value calculator 311 calculates an error value for all triangles constituting the object, the triangle having the smallest error value will have the highest priority of the triangle to collapse. In other words, the order of triangles to be collapsed is determined by the order in which the error values are small. In this order, the rendering part collapses the triangle when performing LOD.

The collapsed vertex calculator 312 receives information about the triangle to be collapsed from the error value calculator 311 and calculates the collapsed vertex using vertex information and triangle information constituting the object. That is, when the collapsed vertex calculation unit 312 is determined to collapse, using the vertex information and the triangle information of the triangle to collapse, calculates the collapsed vertex generated by the collapse of the triangle and outputs it to the rendering unit 320 do. The method of calculating the collapse peak is also described in detail below.

The collapsed vertex attribute information calculation unit 313 receives information about the triangle to be collapsed from the error value calculator 311 and calculates the attribute information of the collapsed vertex using the attribute information of the vertices constituting the object. The property information of the collapsed vertex is preferably calculated for each characteristic. For simplicity of calculation, the attribute information of the collapsed vertex may be calculated as the average of each vertex of the triangle to be collapsed.

 The rendering unit 320 may simplify the graphic image of the object by collapsing the triangle and generating the collapsing vertex when performing LOD by using the information about the collapsing vertex and the attribute information of the collapsing vertex.

5 is a flowchart illustrating a process of a LOD rendering method according to the present invention.

Referring to FIG. 5, the preprocessing unit receives information about vertices and triangles constituting an object to be LOD rendered (510).

The preprocessing unit calculates an error value for each triangle by using the error metric (520). Next, the preprocessing unit determines the order of triangles to collapse in the object based on the error values calculated for each triangle as described above (530).

Next, the preprocessing unit calculates a collapse peak generated by the collapse of the triangle (540).

Then, the rendering unit renders the object while performing LOD based on the information about the vertices constituting the object, the information about the triangle, the order information of the triangle to be collapsed, and the collapsed vertex information (550).

6 is a schematic block diagram of an apparatus implementing the LOD method in VRML according to the present invention.

Referring to FIG. 6, the LOD implementing apparatus 600 includes a preprocessing unit 610 and a rendering unit 620.

The preprocessing unit 610 receives an input VRML model related to the object to be rendered, performs preprocessing for the LOD, generates an LOD VRML model according to the present invention, and provides it to the rendering unit 620.

The input VRML model for the object to be rendered includes a coord with coordinate information of the vertices constituting the object and a coordIndex with information about a triangle, that is, a plane made by the connection of vertices. In addition, the input VRML model may further include attribute information, that is, normal, normalIndex, color, colorIndex, texcoord, and texcoordIndex. normal represents a normal value for a vertex, normalIndex represents a face composed of vertices representing normal values, color represents a color value for vertices, colorIndex represents a surface composed of vertices representing color values, texcoord represents a texture coordinate value for a vertex, and texcoordIndex represents a face constituted by a vertex representing a texture coordinate value.

The preprocessing unit 610 receives these coords and coordIndex to determine the order of the triangles to collapse in this object to create a lodSequence node, and also calculates the collapsed vertices generated by the collapse of the triangles, and thus the positional information of the collapsed vertices. Create a coordStart node with In addition, when the attribute information is present, the preprocessing unit 610 calculates normal values of decay vertices based on normal and normalIndex, calculates color values of decay vertices based on color and colorIndex, and includes texcoord and texcoordIndex. Compute the texture coordinates of the collapsed vertex based on Then, a normalStart node, a colorStart node, and a texcoordStart node with attribute information of the collapsed vertex are created. However, in the case of color, instead of calculating the color value of the collapsed vertex, it is possible to preserve the color information of the three vertices of the existing collapsed triangle as it is. In such a case, the "colorStart node" will not be created. .

In addition, the preprocessing unit 610 determines a level at which to start the collapse of the triangle in the original object to create a minScreenArea node, and one level and the next level in order to reduce annoyance to the user due to the collapse of the triangle. Create an updateDist node that will allow you to gradually collapse the triangle in between. The preprocessing unit 610 provides the rendering unit 620 with the LOD VRML model including the nodes generated in this way.

The rendering unit 620 receives the LOD VRML model from the preprocessing unit 610 and transforms the object while rendering the object to output a graphic image to be displayed.

FIG. 7 is a detailed configuration diagram of the preprocessing unit illustrated in FIG. 6.

Referring to FIG. 7, the preprocessing unit 610 may include an error value calculator 611, a collapsed peak calculator 612, a collapsed peak attribute information calculator 613, and a LOD VRML model generator 614. It includes.

The error value calculator 611 calculates an error value for all triangles constituting the object based on the error metric.

When the error value calculator 611 calculates an error value for all triangles constituting the object, the triangle having the smallest error value will have the highest priority of the triangle to collapse. In other words, the order of triangles to be collapsed is determined by the order in which the error values are small. The error calculator 611 provides the information about the order of the triangles to be collapsed thus determined to the collapsed vertex calculator 612 and the LOD VRML model generator 613.

The collapsed vertex calculator 612 receives information about the triangle to collapse from the error value calculator 611 and calculates a collapsing vertex by using the vertex information constituting the object, that is, coord and triangle information, that is, coordIndex. do. The collapsed vertex calculator 612 provides the LOD VRML model generator 613 with information about the collapsed vertex calculated as described above.

The collapsed vertex attribute information calculating unit 613 receives information about the triangle to be collapsed from the error value calculating unit 611 and also includes attribute information of vertices constituting the object, that is, normal, normalIndex, color, colorIndex, texcoord, texcoordIndex, and the like. Compute the attribute information of the collapsing vertex using. The collapsed vertex attribute information calculator 613 provides the LOD VRML model generator 614 with information about the attribute of the collapsed vertex calculated as described above.

The LOD VRML model generator 614 receives the coord, the coordIndex, the collapsed vertex information, the attribute information of the collapsed vertex, and the order information of the collapsed triangle, inserts the collapsed vertex information into the coord, and generates a deformed coord node. Deformed normal node is created by inserting normal information of decay vertex into it, deformed color node is created by inserting color information of decay vertex into color, and deformed texcoord node by inserting texcoord information of decay vertex into texcoord. Create However, in the case of a color node, a modified color node in which the color information of the collapsed vertex is inserted may be generated as described above, or the color information of the three vertices of the triangle to be collapsed is generated without generating new color information of the collapsed vertex. If used, a color node will be created in which only the color information of the triangle to be collapsed is removed.

In addition, the LOD VRML model generator 614 includes a coordStart node indicating the position of the collapsed vertex inserted in the coord, a normalStart node indicating the position of the collapsed vertex attribute inserted in the normal, and a colorStart node indicating the position of the collapsed vertex attribute inserted in the color. , create a texcoordStart node that represents the location of the collapsed vertex attribute inserted in texcoord. Of course, attribute information such as normal, color or text code is optional. Also, in the case of color, the "colorStart node" will not be created unless it is creating a new color for the decay vertex.

In addition, the LOD VRML model generation unit 614 generates a deformed coordIndex by arranging the triangles in the coordIndex in the order of the decayed triangles using the order information of the decayed triangles, and generates the deformed coordIndex in the order of the decayed triangles at the normalIndex. Arrange according to create the modified normalIndex, arrange the triangle in the colorIndex according to the order of the triangle to collapse, create the modified colorIndex, arrange the triangle in the order of the triangle to collapse in the texcoordIndex and create the modified texcoordIndex. In addition, the LOD VRML model generator 613 generates a lodSequence node indicating the order information of the triangle to be collapsed. In addition, the LOD VRML model generator 613 generates an updateDist node and a minScreenArea node according to a user input or predetermined setting information in the system. The LOD VRML model generator 613 outputs the LOD VRML model including the coord node, the coordIndex node, the lodSequence node, the coordStart node, the updateDist node, and the minScreenArea node generated in this way to the rendering unit 620. Optionally, when there is further attribute information, a normal node, a normalIndex node, a normalStart node, a color node, a colorIndex node, a colorStart node, a texcoord node, a texcoordIndex node, and a texcoordStart node may be further included.

8 is a flowchart of a preprocessing operation performed by the preprocessing unit shown in FIG. 7.

First, the preprocessing unit receives coord and coordIndex, which are information about an object to be rendered, according to the VRML model, and calculates an error value of each triangle included in the object using an error metric (810).

Next, the preprocessing unit calculates collapsing vertices for each triangle (820).

Next, the preprocessing unit sorts the triangles included in the coordIndex in order of decreasing error values (830).

Next, the preprocessing unit generates LodSequence and CooordStart based on the sorted coordIndex (840).

Then, the preprocessing unit sets minScreenArea and UpdateDist values according to a user input or a predetermined setting value (850).

Next, the preprocessing unit stores the coord, coordIndex, lodSequence, coordStart, updateDist, and minScreenArea as a LOD VRML model (860).

FIG. 9 is a flowchart of a rendering operation performed in the rendering unit illustrated in FIG. 7.

Referring to FIG. 9, the rendering unit receives an LOD VRML model generated from the preprocessing unit (910).

Next, the rendering unit determines the LOD starting point according to minScreenArea (920).

The rendering unit then performs LOD based on LodSequence and CooordStart (930).

Next, the rendering unit performs geomorphing according to the updateDist (940).

Now, the LOD method according to the present invention will be described in more detail.

Error metrics

First, an error metric as an index for determining a triangle to collapse in accordance with the present invention will be described.

The error metric, as described above, indicates the degree of the triangle's effect on the object. The error metric for each triangle F constituting the object is the error value E (F) for each triangle. The large E (F) of a triangle indicates that this triangle affects the object a lot, so it is best to collapse this triangle as much as possible when deforming the object. Also, the small E (F) of a triangle indicates that it does not affect the object so much that it can be determined to be the first triangle to collapse when deforming the object.

The error metric E (F) may be defined by Equation 1 as follows.

Equation 1

That is, the total error E (F) may be expressed as the sum of E _geometry (F), which is an error regarding geometrical characteristics, and E _attribute (F), which is an error regarding attribute properties.

Error regarding geometrical characteristics With regard to E _geometry (F), the smaller the coverage, the smaller the area of the triangle, and the earlier the inner triangle is removed, the better the visual quality. Therefore, E _geometry (F), which is an error about geometrical characteristics, is defined by Equation 2 as follows.

Equation 2

"curvature (F)" represents the coverage (curvature radius) of this triangle, more specifically, it can be found to be 1 / (the radius of the sphere closest to F and neighboring triangles). Here, F denotes a target triangle for which an error metric is to be obtained, and Nei _F denotes a triangle neighboring the target triangle. That is, the radius of curvature herein refers to the radius of curvature of a sphere that can be formed closest to this plane by the plane formed by the object triangle and neighboring triangles.

"area (F) / area (the largest triangle in mesh)" represents the influence of the area of the triangle and represents the ratio of the area of the target triangle to the area of the largest triangle in the object.

"| Nei _F <3 |" indicates whether the triangle is at the boundary, and Nei _F indicates the number of triangles adjacent to the triangle F, and thus determines whether the number of neighboring triangles is smaller than three. For example, the number of neighboring triangles of triangle F shown in FIG. 10A is three, but the number of neighboring triangles of triangle F shown in FIG. 10B is two. Since collapsing triangle F with three neighboring triangles, as shown in FIG. 10A, will lessen the representation of the object than collapsing triangle F with two neighboring triangles, as shown in FIG. 10B | Nei _F <3 | is an index for distinguishing these triangles.

Error regarding attribute characteristics For E _attribute (F), triangles with small change in color, small change in normal, and small change in texture coordinate between adjacent triangles are first. Eliminating can improve visual quality. Therefore, E _attribute (F), which is an error regarding an attribute characteristic, is defined by the following equation.

Expression 3

here,

IsFaceSharp (F) is for handling the case where the texture coordinate values given at the vertices are different when two or more triangular faces share a given vertex.

는 면 I의 neighborhood의 개수를 나타내고,

Denotes the number of neighborhoods on face I,

는 컬러의 변화량을 나타내고,

Represents the amount of change in color,

은 노말의 변화량을 나타낸다. 도 10c를 참조하면, F는 에러 메트릭을 구하려는 대상 삼각형을 나타내고, G는 이 대상 삼각형에 이웃하는 삼각형들중의 하나를 나타내고, V1과 V2는 삼각형 F와 삼각형 G가 공유하는 정점(vertex)을 나타낸다.

Represents the amount of change of normal. Referring to FIG. 10C, F denotes a target triangle for which to obtain an error metric, G denotes one of the triangles neighboring the target triangle, and V1 and V2 denote a vertex shared by the triangle F and the triangle G. Indicates.

IsFaceSharp (F) is for handling the case where the texture coordinate values given at the vertices are different when two or more triangular faces share a given vertex. That is, when different texture coordinate values are given to one vertex, the vertex is defined as "sharp", and IsFaceSharp (F) has a value of "1". Otherwise, that is, given a texture coordinate value at a given vertex, it has a value of "0". 10D, four triangles F1. If the texture coordinates that share the vertex F0 among F2, F3, and F4 are not the same, it is defined as "sharp", otherwise it is not defined as "sharp".

The amount of color change is the difference between the color value of vertex V1 in triangle F and the color value of vertex V1 in triangle G, and the difference between the color value of vertex V2 in triangle F and the color value of vertex V2 in triangle G. It is defined as the plus value.

The change amount of the normal is obtained by adding the difference between the normal value of the vertex V1 in the triangle F and the normal value of the vertex V1 in the triangle G, and the difference between the normal value of the vertex V2 in the triangle F and the normal value of the vertex V2 in the triangle G. Defined by value.

In addition, the coefficient of each item shown in Equation 2 and Equation 3 is described as follows.

The default value of each coefficient is selected such that each corresponding item has a value between 0 and 1.

a is defined as the radius of curvature of the sphere with the smallest radius of curvature among all triangles. In this way, a * curvature (F) has a value between 0 and 1.

b is defined as "1". This selects an item corresponding to b to have a value between 0 and 1.

As c is a special case, the item corresponding to c is defined as some predetermined large default value for c, since its purpose is to prevent the triangles at the boundary from collapsing.

Since an item corresponding to d is intended to prevent the triangles to which different textures are mapped from being collapsed, d is defined as a predetermined default value. However, it is based on a smaller value than c because it has less visual change when collapsed than c. For example, when c is set to "100", d may be set to "10".

e =

f =

  By calculating an error metric, that is, an error value, for all triangles constituting the object to be rendered by the error metric calculation method as described above, it is possible to determine which triangle should be collapsed first when deforming the object. That is, among the obtained triangles, the triangle with the smallest error value is determined as the triangle that must collapse first, and the second small triangle is determined as the triangle that must collapse second. In the preprocessing step, the triangles constituting the object to be rendered are arranged in the order of the smallest error values using the error value calculated as described above, and if the information about the order is provided to the rendering step, the rendering step determines the LOD according to the order information. At run time, the object collapses the triangle.

Calculation of collapsing vertices

If the calculation of the error metric determines the triangles to collapse in the object to be rendered or the order of the triangles to collapse, then the collapsed vertices that must be created when this triangle collapses must now be calculated. When a triangle collapses in an object, it is desirable to represent the newly created decay vertex close to its original shape.

FIG. 11 is a reference diagram for explaining a concept of generating a collapse peak by collapse of one triangle according to the present invention.

Referring to FIG. 11, the left side schematically shows a cross section 1100 of an object before the triangle collapses, and the right side schematically shows a cross section 1110 of the object after the triangle collapses.

That is, if the cross section of the object consists of vertices M1 to M4 before collapse, and attempts to collapse the triangle by vertex M2 and vertex M3 (the other vertex of the triangle is not shown because the section is displayed), It is desirable to collapse the triangle close to the shape (G) of the object to create a new decay vertex. However, since the shape G of the original object is not known in practice, it is important to select the collapse peak M5 that approximates the shape of this original object.

12 is an operational flow of collapse of one triangle to produce a collapse peak in accordance with the present invention.

First, the vertices of the triangle to be collapsed are defined as V1, V2, and V3, and normals to the surface of the mesh at V1, V2, and V3 are defined as N1, N2, and N3, respectively (1210).

Then, the planes containing vertices Vi and perpendicular to Ni are defined as tangent planes P1, P2, and P3 (1220).

Next, the center point of the triangle to collapse, that is, C = (V1 + V2 + V3) / 3 is obtained (1230).

Then, the center point C of the triangle is projected onto the tangent planes P1, P2, and P3, respectively, to obtain three projected points C1, C2, and C3 (1240).

Finally, the collapsed vertex V4, the point generated by the collapse of the triangle, is computed by (C1 + C2 + C3) / 3, which is the average of these three projected points C1, C2, C3 (1250).

The properties of the collapse point are determined by calculating the average of the properties of P1, P2, P3.

An example of a process of creating a collapsed peak by decaying a triangle in accordance with the present invention will now be described with reference to FIGS. 13A-13F.

FIG. 13A shows an object 1300 consisting of four triangles T1, T3, T7, and T8. The triangle T1 is composed of vertices V1, V2, and V3, and it is assumed that the triangle T1 is determined to be a triangle to collapse. If the original object 210 shown in FIG. 2 is viewed from the perspective of the triangle T1 from above, FIG. 13A shows the user of the triangle T1 of the original object 210 in order to facilitate the explanation of the object in a three-dimensional manner. It is shown from the point of view.

Referring to FIG. 13B, the normals to the surface of the mesh 1310 at three vertices V1, V2, and V3 of the triangle to be collapsed are defined as N1, N2, and N3, respectively.

Referring to FIG. 13C, a plane including vertex Vi and perpendicular to normal Ni is defined as tangent plane Pi. That is, the plane including vertex V1 and perpendicular to normal N1 is defined as tangent plane P1. The plane containing vertex V2 and perpendicular to normal N2 is defined as tangent plane P2. The plane containing vertex V3 and perpendicular to normal N3 is defined as tangent plane P3.

Next, referring to FIG. 13D, the center point of the triangle T1 to collapse is defined as C = (V1 + V2 + V3) / 3, and the center point C of the triangle T1 to be collapsed onto the tangent planes P1, P2, and P3. Project each to obtain three projected points C1, C2 and C3.

Next, referring to FIG. 13E, the collapsed vertex V4, the point generated by the collapse of the triangle, is calculated by (C1 + C2 + C3) / 3 which is the average of these three projected points C1, C2, C3.

Thus, an object formed by the collapsed vertex newly created by the collapse of triangle T1 may be drawn as shown in FIG. 13F.

The properties of the collapsed vertex can also be defined as the average of the property values defined for the three projected points.

In addition, in order to more precisely determine the properties of the collapsed vertex, it is desirable to obtain the properties of the collapsed vertex according to the characteristics of the three vertices of the collapsed triangle. Hereinafter, a method of obtaining the properties of the collapsed vertex according to the characteristics of the three vertices of the triangle to be collapsed will be described. Properties of decay vertices include texture coordinates, color, and normal vectors.

First, the texture coordinates of the collapsed vertex are obtained as follows. The texture coordinate value of the collapsed vertex is preferably obtained by obtaining the texture coordinate value of the collapsed vertex such that the texture coordinate value of the collapsed vertex is most affected by the texture coordinate value of the vertex located nearest to the collapsed vertex among the three vertices of the collapsed triangle.

Referring to FIG. 14, assume that (P1, P2, P3) is three coordinates of a triangle, texture coordinates of (T1, T2, T3), coordinates of collapsed vertices are P, and texture coordinates are T. At this time, if P is closer to P3 than P1 and P2, the T value should be affected by T3.

At this time, if P is mathematically expressed using P1, P2, and P3, it can be expressed as the following equation.

P = a1 * P1 + a2 * P2 + a3 * P3

Here, T values can be expressed as T = a1 * T1 + a2 * T2 + a3 * T3 using the obtained a1, a2, and a3. However, since the P value does not exist on the plane consisting of P1, P2, and P3, the value of Q can be obtained by projecting P onto a triangle, and this Q can be expressed by the following formula.

Q = a1 * P1 + a2 * P2 + a3 * P3

At this time, the coefficient values a1, a2, and a3 are called barycentric coordinates of the Q values on the triangles (P1, P2, P3), and the coefficient values are defined by the following equation.

, a1+a2+a3=1

, a1 + a2 + a3 = 1

Here, S _abc represents the area of the triangle ABC.

According to the above formula, a1, a2, and a3 can be defined as follows.

Therefore, the texture coordinate values T = a1 * T1 + a2 * T2 + a3 * T3 are obtained using a1, a2 and a3 obtained here.

Next, the normal vector of the decay peak is obtained as The normal vector of the collapsed vertex is to obtain the normal vector value of the collapsed vertex so that the normal vector value of the collapsed vertex is most affected by the normal vector value of the vertex closest to the collapsed vertex among the three vertices of the collapsed triangle. desirable.

Normal vectors are located on the surface of a sphere with a radius of 1 in the sphere coordinate system. Using this property, a normal vector assigned to three vertices of a given triangle is represented by a sphere coordinate system as shown in FIG. 15. n1, n2, and n3 represent the normal vectors of the three vertices of the triangle to be collapsed, respectively, and n represents the normal vectors of the collapsed vertices, respectively.

n1 = (1, theta1, phi1)

n2 = (1, theta2, phi2)

n3 = (1, theta3, phi3)

n = (1, theta, phi)

Therefore, the normal vector of the decaying vertex is obtained by interpolating (theta, phi) on a sphere coordinate system. Where the new normal vector n is

n = a * n1 + b * n2 + c * n3

At this time, the coefficient a + b + c = 1, and each coefficient is represented by the barycentric coordinates used when obtaining the texture coordinate value.

By replacing the next sphere coordinate system (1, theta, phi) with (x, y, z) again, we get the normal vector in the x, y, z coordinate system.

Next, the color of the collapsed vertex is obtained as follows.

Referring to FIG. 16, when a given triangle (2, 4, 5) collapses to a new vertex 9, if a new color is generated by blending the color of the vertex assigned to the vertex 9 with the surrounding color, a completely different color This can come out. Therefore, for the color value, the original given color is preserved as it is without generating a new color. That is, the newly created vertex 9 has the color assigned to each vertex of the triangles 2, 4 and 5 collapsing in the left figure of FIG. That is, it only removes the color for the missing triangles (2, 4, 5). For example, if vertex 2 is red, vertex 4 is green, and vertex 5 is blue, red, green and blue may be assigned to collapsed vertex (9).

Therefore, even if the original model triangle (1,4,7) collapses to (1,9,7), the triangle preserves the color of the original triangle (1,9,7).

For example, suppose you have the following model:

original coordIndex [

          2, 1, 7, -1,

          0, 1, 5, -1,

          0, 4, 2, -1,

          2, 3, 1, -1,

          0, 5, 4, -1,

          1, 3, 5, -1,

          2, 4, 3, -1,

          4, 5, 3, -1,

          0, 7, 1, -1,

          6, 0, 2, -1,

          6, 2, 7, -1,

          6, 7, 0, -1]

original colorIndex [

          c1, c2, c3, -1,

          c4, c5, c6, -1,

          c7, c8, c9, -1,

          c10, c11, c12, -1,

          c13, c14, c15, -1,

          c16, c17, c18, -1,

          c19, c20, c21, -1,

          c22, c23, c24, -1,

          c25, c26, c27, -1,

          c28, c29, c30, -1,

          c31, c32, c33, -1,

          c34, c35, c36, -1]

In the above model, if triangle (6,7,0) collapses to a new vertex 8, we assign ColorIndex as

new coordIndex [

          2, 1, 8, -1,

          8, 1, 5, -1,

          8, 4, 2, -1,

          2, 3, 1, -1,

          8, 5, 4, -1,

          1, 3, 5, -1,

          2, 4, 3, -1,

          4, 5, 3, -1]

The new coordIndex, from the original coordIndex, shares two points with the triangle (6,0,7) and triangle (6,0,7) (6,2,7), (6,0,2), (0 It can be seen that, 7,1 is removed, and in the triangle sharing one point with triangle (6,0,7), one point sharing with triangle (6,0,7) is changed to eight. In other words, triangle (2,1,7) turns into (2,1,8), triangle (0,1,5) turns into (8,1,5), and triangle (0,4,2) turns into ( 8,4,2), and the triangle (0,5,4) is changed to (8,5,4).

new colorIndex [

          c1, c2, c3, -1,

          c4, c5, c6, -1,

          c7, c8, c9, -1,

          c10, c11, c12, -1,

          c13, c14, c15, -1,

          c16, c17, c18, -1,

          c19, c20, c21, -1,

          c22, c23, c24, -1]

The new colorIndex removes the four color values from the bottom of the original colorIndex, and the color is preserved even if the coordinates of the vertices of the triangle change. For example, (2,1,7) in the original coordIndex is changed to (2,1,8) in the new coordIndex, but original colorIndex (c1, c2, c3) replaces (c1, c2, c3) in the new colorIndex as well. Keep it.

LOD Levels and Geomorphing

17 is a reference diagram for explaining seamless rendering according to the present invention.

Referring to FIG. 17, LOD level = 0 represents rendering 1710 in its original state without any object deformation.

LOD level = 1 (1720) is the next stage of LOD level = 0 (1710) is a state where the collapse of a series of triangles has been performed, the object is a one-step deformation, LOD level = 2 (1730) is LOD level = 1 Next step and so on.

Referring to FIG. 18, a distance relationship between a camera, a screen, and a 3D object formed of a mesh is illustrated.

If q becomes smaller than a predetermined threshold as shown in the following equation, then a simplification process is performed.

That is, q represents the ratio of the size of the area projected onto the screen by the 3D object to the size of the area of the entire screen. Thus, if the ratio of the projected area of the 3D object to the area of the entire screen becomes smaller than the predetermined threshold, then simplification or deformation of the object begins.

The LOD level is proportional to the distance "z" from the camera to the mesh. That is, as shown below, the LOD level is determined on a logarithmic scale according to distance.

Where px and py represent camera projection coefficients, and r represents the radius of the 3D object.

19 is a view for explaining Geo-Morphing according to the present invention.

When the LOD level changes, for example, a sudden drop in the number of triangles or a sudden decrease in the number of vertices in the original object causes an unobtrusive "popping" of the viewer.

To prevent this, as shown in FIG. 19, when proceeding from the level k-1 shown in (a) to the level k shown in (c), the process does not immediately proceed from the level k-1 to the level k (b). You can prevent this popping phenomenon by adding an intermediate step as shown in. In the preprocessing step according to the invention, since the collapsed vertices are calculated when the triangle collapses in the object, it has both information about the collapsed triangle and information about the collapsed vertices generated by the collapse of the triangle. Only by adjusting it is possible to easily create an intermediate step as shown in (b) during the rendering process.

FIG. 20 illustrates an embodiment of Geo-Morphing according to the present invention, and proceeds through step (b) in the middle of steps (a) and (c) rather than immediately proceeding from step (a) to (c). It can be seen that the transformation can proceed more naturally and gradually.

VRML node syntax

As far as working with the VRML browser, the input mesh for the simplification algorithm is the IndexedFaceSet. The output is an improved IndexedFaceSetLOD and simplified sequence. Improvements to the original IndexedFaceSet introduce new vertices. That is, the collapse of each triangle creates one new vertex. The optimization of the original IndexedFaceSet requires reordering its coordIndex, normalIndex, colorIndex, and texCoordIndex arrays. The basic idea of reordering is to put information about the collapsing surface first in the array.

21 shows node syntax newly defined according to the LOD method according to the present invention.

lodSequence is the node whose triangular information is to be collapsed.

coordStart is a node that contains position information that contains coordinate information of the triangle to collapse.

texCoordStart is a node with position information that contains texture coordinate information of the triangle to collapse.

normalStart is a node that contains location information containing the normal information of the triangle to collapse.

CoordStart must exist because the coordinates of the triangle's properties must be present, but texCoordStart and normalStart are both optional nodes because texture coordinates, normals, and colors are optional. And, if more than one triangle is to be collapsed, the coordinate information of the first collapsed vertex generated by the collapse of the first triangle is obtained from coordStart, and the coordinate information of the second collapsed vertex is obtained from coordStart + 1. The same applies to the remaining texCoordStart, normalStart or colorStart.

updateDist is a node that defines the detailed steps at that time when one or more steps are placed between one level and another for seamless rendering between LOD levels as described above.

minScreenArea is a node that represents the starting point for starting the LOD.

Meanwhile, among the nodes included in the VRML model, the coord node, the texCoord node, the normal node, the color node, the coordIndex node, the texCoordIndex node, the normalIndex node, and the colorIndex node are rearranged according to the present invention.

The coord node has coordinate information about the vertices constituting the object to be rendered according to the VRML model definition. According to the present invention, the coord node is located next to the coordinate information where the information about the coordinates of the collapsed vertex that is newly created due to the collapse of the triangle is included. Are arranged in. The texCoord node has texture coordinate information about the vertices constituting the object to be rendered according to the VRML model definition. According to the present invention, the texture coordinate where the information about the texture coordinates of the collapsing vertex newly generated by collapse of the triangle is included. The information is arranged at the following location. The normal node has normal information about the vertices constituting the object to be rendered according to the VRML model definition. According to the present invention, the normal node has the normal information about the normal of the collapsed vertex that is newly created due to the collapse of the triangle. Are arranged in. The color node has color information about the vertices constituting the object to be rendered according to the VRML model definition. According to the present invention, in the case of color, when the color of the collapsed vertex is obtained using the average, the triangle is collapsed and newly created. If the information about the color of the collapsing vertex that is generated is arranged next to the color information that originally contained it, and is used as the color of the collapsing vertex using the color information of the triangle's vertices to collapse without using an average, the collapsing is only at the color node. Only the color information for the triangles that have been removed is removed.

The coordIndex node, the texCoordIndex node, the normalIndex node, and the colorIndex node rearrange the elements arranged therein in the order of the triangle to collapse.

The simplification sequence is encoded as MFInt32 lodSequence of the type as shown in FIG.

A mesh with LOD = 0 represents the original object state before it was deformed, that is, before it was simplified.

Referring to FIG. 22, when generating LOD = 1 mesh, collapse of n ₁ triangles is performed. "-1" between LOD = 1 mesh and LOD = 2 mesh is a separator.

The number of triangles collapsing when generating LOD = 2 mesh is n ₂ . And the number of triangles collapsed when generating LOD = N mesh is n _N.

The texCoordIndex node, the normalIndex node, and the colorIndex node are not described separately, but face removal is performed in the same manner as the coordIndex node.

FIG. 23 is a diagram referred to for describing updateDist illustrated in FIG. 21.

updateDist is a node for further setting the level of detail to gradually transform an object between one LOD level and the next.

FIG. 24 is a reference diagram for describing minScreenArea illustrated in FIG. 21.

The minScreenArea is determined by the ratio of the rendered mesh in the entire screen as described above, and is a node that determines when the first LOD starts.

25 illustrates an example of transforming an original VRML model 2510 into a LOD VRML model 2520 according to the present invention, where one triangle 2530 collapses and one collapsed vertex 2540 is generated. It is a reference figure for demonstrating, for example.

Referring to FIG. 25, the original VRML model 2510 includes a coord node 2511 that is information about a vertex of an object 2530 and a coordIndex node 2512 that is information about a face of an object 2530.

The node coord 2511 has coordinate information 0.0 1.1 2.0, 1.0 0.1 2.0, 2.0 1.1 2.0 about the vertices 0, 1, and 2 of the object 2530. In addition, the node coordIndex 2512 has 0, 1, and 2 which are information about the face of the object 2530.

The preprocessing unit according to the present invention receives the original VRML model 2510 and determines the triangle to collapse and calculates the collapsed vertex. In this example, since there is only one triangle in the object, it is determined to be the triangle to collapse and the collapsed vertex Let is the coordinate of (1.0 0.76 2.0).

Then, in the LOD VRML model 2520 according to the present invention, the coord 2521 has (1.0 0.76 2.0) which is the coordinate of the collapsed vertex at the last position.

coordIndex 2522 will be rearranged in the order of triangles to collapse, but in this example there is only one triangle to collapse, which is the same as coordIndex 2512 in the original VRML model 2510.

lodSequence 2523 also has only one triangle to collapse, so 1 is arranged.

coordStart 2524 contains positional information " 3 " in which collapsing vertices are arranged in coord 2521.

updateDist 2525 is set to 0 and minScreenArea 2526 is set to 0.08.

A case where there is only one triangle to collapse will be described with reference to FIG. 26, but now a case where two triangles to collapse is described.

26A shows the object 2600 to render. Referring to FIG. 26A, an object 2600 to be rendered is composed of eight vertices of vertices 0 to 7.

FIG. 26B shows the original VRML model 2610 for the object 2600 to render shown in FIG. 26A. There may be other attribute information about the object 2600 such as texture and color, but only coordinate information is illustrated in FIG. 26B. Referring to FIG. 26B, coordinate information of vertices 0 to 7 included in the object 2600 to be rendered is arranged in the coord 2611. The coordinate information of vertex 0 is represented by x0, x0, z0, and the coordinate of vertex 1 is represented by x1, y1, z1. The same is true for the remaining vertices.

The coordIndex 2612 is arranged with information about the plane of the object 2600 to be rendered, that is, the triangle.

The preprocessing unit according to the present invention having received the original VRML model 2610 uses the information about the vertices in the coord 2611 and the information on the planes in the coordIndex 2612 to generate errors for each triangle. Get the value and arrange the triangles based on this error value to determine the order of triangles to collapse. FIG. 26C illustrates an LOD VRML model 2620 generated by the preprocessing unit according to the present invention.

First, assume that the triangle to be collapsed in the object 2600 is determined as a triangle composed of vertices (6,7,0) and a triangle composed of vertices (3,4,5). The information about the face of coordIndex 2622 in LOD VRML model 2620 is then sorted in the order of this collapsed triangle. That is, referring to FIG. 26C, at coordIndex 2622, information m1 regarding the triangle to be collapsed first is located at the end of the array. Referring next to FIG. 26A, by the collapse of a triangle consisting of vertices (6, 7, 0), a triangle sharing two vertices with this triangle collapses accordingly. The triangles thus collapsed include triangles formed by vertices (6, 2, 7), triangles formed by vertices (6, 0, 2), and triangles formed by vertices (0, 7, 1). In this way, information about the triangle to collapse along with the first triangle to collapse is arranged in front of m1 at coordIndex 2622 as m2, m3, m4.

Since the next triangle to collapse is a triangle formed by vertices (3, 4, 5), information (m5) about this triangle is arranged before m4. Referring also to FIG. 26A, by the collapse of a triangle consisting of vertices (3, 4, 5), a triangle sharing two vertices with this triangle collapses accordingly. The triangles thus collapsed include triangles composed of vertices (2, 4, 3), triangles composed of vertices (1, 3, 5), and triangles composed of vertices (0, 5, 4). The information about the triangle to be collapsed accompanying the second triangle to collapse is thus arranged in front of m5 at coordIndex 2622 as m6, m7, m8. The description of the remaining vertices is omitted.

In addition, lodSequence 2623 has the number of triangles that collapse according to the collapse order of these triangles. lodSequence (2623) collapses the four triangles associated with (6,7,0) from LOD level 0 to LOD level 1 as described above, and 4 associated with (3,4,5) from LOD level 1 to LOD level 2 This means that the dog's triangle will collapse. "-1" between levels serves to distinguish between LOD levels and levels.

Meanwhile, in the preprocessing unit according to the present invention, when the triangle to collapse in the object to be rendered is determined, the collapsed vertex generated by the collapse of the triangle is calculated. The information of the collapsed vertices thus calculated is located at the end of the array of coord nodes 2621. In this example, we have determined two collapsed triangles, so we also create two collapsed vertices, the coordinates of which are (x8, y8, z8) and (x9, y9, z9).

Information about the location of this collapsed vertex is stored in the coordStart node 2624. Referring to FIG. 23C, since coordStart 2624 contains "9", the rendering unit may obtain coordinate information of the first collapsed vertex at the ninth position in the coord 2621. To get the coordinates of the second decay vertex, the renderer simply finds the position in the coord based on the increment of the value of coordStart.

The LOD rendering method as described above can also be embodied as computer readable code on a computer readable recording medium. The computer-readable recording medium includes all kinds of recording devices in which data that can be read by a computer system is stored. Examples of computer-readable recording media include ROM, RAM, CD-ROM, magnetic tape, floppy disk, optical data storage, and the like, and may also be implemented in the form of a carrier wave (for example, transmission over the Internet). Include. The computer readable recording medium can also be distributed over network coupled computer systems so that the computer readable code is stored and executed in a distributed fashion. In addition, functional programs, codes, and code segments for implementing the LOD rendering method may be easily inferred by programmers in the art to which the present invention belongs.

So far I looked at the center of the preferred embodiment for the present invention. Those skilled in the art will appreciate that the present invention can be implemented in a modified form without departing from the essential features of the present invention. Therefore, the disclosed embodiments should be considered in descriptive sense only and not for purposes of limitation. The scope of the present invention is shown in the claims rather than the foregoing description, and all differences within the scope will be construed as being included in the present invention.

According to the present invention as described above, it is possible to effectively render the three-dimensional object in a resource-limited system. The present invention is an embedded system that renders and loads 3D data effectively in terminal devices such as 3D telematics or low performance PDA, phone, etc., which perform effective laddering for massive 3D terrain data. ) Can be applied.

Claims (51)

In the method for converting a graphic image of an object consisting of at least one triangle, Determining at least one triangle to be collapsed among the triangles forming the object, Determining the collapse peak to be created by the collapse of the triangle by calculating the collapse peak based on the three vertices of the triangle to be collapsed and the tangent plane of the three vertices; And converting the graphical image of the object based on the determined information about the collapsed triangle and the information about the determined collapsed vertex. The method of claim 1, Determining the triangle to be collapsed, Determining an indicator that has less influence on the representation of the object when one triangle constituting the object is removed; Determining a triangle to collapse in the object according to the indicator. delete The method of claim 1, The decay vertex determination step, And calculating attribute information of the collapsed vertex based on three vertices of the triangle to be collapsed. The method of claim 4, wherein The attribute information calculation step of the collapsed vertex, And calculating attribute information of the collapsed vertex by using an average of three vertices of the triangle to be collapsed or by reflecting the characteristics of the three vertices. delete delete In the method for adaptively converting a graphical image of an object consisting of at least one triangle, Defining an error metric representing an index in which a triangle constituting the object affects the representation of the object as a sum of an error relating to a geometric characteristic of the triangle and an error relating to an attribute characteristic of the triangle; Obtaining an error value for each triangle constituting the object based on the error metric; And first collapsing the triangle having the smallest error value among the obtained error values. delete The method of claim 8, The error about the geometrical characteristics of the triangle is And the radius of curvature of the triangle, the area occupied by the triangle in the entire mesh, and whether the triangle is a triangle at a boundary. The method of claim 8, The error about the attribute characteristic of the triangle is And the amount of change in color, the amount of change in normality, and the amount of change in texture coordinate value between the triangle and the triangles adjacent to the triangle. The method of claim 8, Collapse the triangle, Creating a collapse vertex to replace the triangle; Changing a connection relationship of triangles constituting the object based on the decay vertex, And generating a graphic image of the object based on the changed connection relationship. In a method for decomposing a triangle included in an object composed of one or more triangles represented by a graphic image into vertices, Defining the normals to the mesh surface of the object for each of the three vertices of the triangle to collapse; Defining, for each of the three vertices of the triangle, a tangent plane that is a plane perpendicular to the normal and including vertices; Projecting the center points of the three vertices of the triangle onto the three tangent planes, respectively, to obtain three projected points; Obtaining a decay vertex by calculating an average of the three projected points. The method of claim 13, Obtaining an attribute value of the collapsed vertex based on the attribute values of the three vertices of the triangle to be collapsed. The method of claim 14, The attribute value obtaining step is characterized in that using the average value of the three vertices. The method according to claim 14, The property value obtaining step includes calculating a property value by reflecting the characteristics of the three vertices. The method of claim 16, The attribute value obtaining step, Obtaining a texture coordinate value of the collapsed vertex so as to be most affected by the texture coordinate value of the vertex located nearest to the collapsed vertex among the three vertices of the triangle to be collapsed. The method of claim 16, The attribute value obtaining step, Obtaining a normal vector value of the collapsed vertex so as to be most affected by the normal vector value of the vertex located closest to the collapsed vertex among the three vertices of the triangle to be collapsed. The method of claim 16, The attribute value obtaining step, And the color of the collapsed vertex preserves the color of the three vertices of the triangle to be collapsed. A graphic model generation method for adaptively converting a graphic image of an object consisting of at least one triangle, Determining a triangle to collapse from, an order of triangles to collapse, and a collapse peak to be generated by the collapse of the triangle; Creating a triangle identifier node in which information about the triangles is arranged in the order of the triangle to be collapsed; Generating a level of detail sequence node having the sequence information of the triangle to be collapsed; Generating a coordinate node in which coordinate information of vertices constituting the object and coordinate information of the collapsed vertex are arranged; Generating a collapsed vertex position node indicative of the location of the collapsed vertex arranged at the coordinate node. The method of claim 20, The determining step, Obtaining an error value for each triangle constituting the object based on an error metric representing an index in which the triangle constituting the object affects the representation of the object; And determining the order of the triangles to be collapsed so that the triangle with the smaller error value first collapses. The method of claim 21, Creating an update node indicative of the degree to which the triangle collapses gradually between one level and the next level at which the triangle collapse occurs; Generating a node indicative of level information at which to start the collapse of the triangle in the original object. The method of claim 21, Generating a color node of the vertices constituting the object, a color node in which the color information of the collapsed vertex is arranged, and a collapsed vertex color node indicative of a position of the color information of the collapsed vertex arranged in the color node; Generating collapsed vertex normal nodes representing positions of normal information of vertices constituting the object, normal nodes of the collapsed vertex normal information, and positions of normal information of the collapsed vertices arranged in the normal node; A texture coordinate node representing texture coordinate information of vertices constituting the object, texture coordinate node in which texture coordinate information of the collapsed vertex is arranged, and texture coordinate information node of the collapsed vertex texture coordinate node arranged in the texture coordinate node; Graphical model generation method comprising at least one of the steps of generating. The method of claim 21, Generating a color node in which color information of vertices constituting the object except color information regarding collapsed triangles is arranged; Generating collapsed vertex normal nodes representing positions of normal information of vertices constituting the object, normal nodes of the collapsed vertex normal information, and positions of normal information of the collapsed vertices arranged in the normal node; A texture coordinate node representing texture coordinate information of vertices constituting the object, texture coordinate node in which texture coordinate information of the collapsed vertex is arranged, and texture coordinate information node of the collapsed vertex texture coordinate node arranged in the texture coordinate node; Graphical model generation method comprising at least one of the steps of generating. In the VRML model generation method for adaptively transforming the graphic image of the object consisting of at least one triangle, Determining a triangle to collapse from, an order of triangles to collapse, and a collapse peak to be generated by the collapse of the triangle; Creating a coordIndex node that arranges information about the triangles in order of the triangle to collapse; Creating a lodSequence node having the sequence information of the triangle to be collapsed; Generating a coord node in which coordinate information of vertices constituting the object and coordinate information of the collapsed vertex are arranged; Generating a coordStart node indicative of the location of the collapsed vertex arranged in the coord node. The method of claim 25, A normalIndex node in which information about the triangles is arranged in the order of the collapsed triangle, normal information of vertices constituting the object, normal node in which normal information of the collapsed vertex is arranged, the collapsed array arranged in the normal node Creating a normalStart node representing the position of the vertex normal information, A colorIndex node in which information about the triangles is arranged in order of the collapsed triangle, a color node of vertices constituting the object, a color node in which color information of the decay vertex is arranged, and the decay arranged in the color node Creating a colorStart node indicating the position of the vertex color information, A texcoordIndex node in which information about the triangles is arranged in the order of the collapsed triangle, texture coordinate information of vertices constituting the object, texture coordinate information of the collapsed vertex, and arranged in the texcoord node. And generating at least one of a texcoordStart node indicating a position of texture coordinate information of the collapsed vertex. The method of claim 25, A normalIndex node in which information about the triangles is arranged in the order of the collapsed triangle, normal information of vertices constituting the object, normal node in which normal information of the collapsed vertex is arranged, the collapsed array arranged in the normal node Creating a normalStart node representing the position of the vertex normal information, Creating a colorIndex node in which information about the triangles is arranged in order of the collapsed triangle, a color node in which color information of vertices constituting the object except color information regarding collapsed triangles is arranged; A texcoordIndex node in which information about the triangles is arranged in the order of the collapsed triangle, texture coordinate information of vertices constituting the object, texture coordinate information of the collapsed vertex, and arranged in the texcoord node. And generating at least one of a texcoordStart node indicating a position of texture coordinate information of the collapsed vertex. An apparatus for converting a graphical image of an object consisting of at least one triangle, Determining at least one triangle among the triangles constituting the object, and determining the collapsed vertex to be generated by the collapse of the triangle based on the three vertices of the triangle to be collapsed and the tangent plane of the three vertices With preprocessing part to say, And a rendering unit that renders while converting a graphic image of the object based on the information about the triangle to be collapsed and the information about the collapsed vertex determined by the preprocessing unit. The method of claim 28, The preprocessing unit, And determine an indicator that has less influence on the representation of the object when one triangle constituting the object is removed, and determines a triangle to collapse in the object according to the indicator. delete delete delete An apparatus for adaptively converting a graphical image of an object consisting of at least one triangle, An error metric representing an index in which a triangle constituting the object affects the representation of the object is defined as a sum of an error relating to a geometric characteristic of the triangle and an error relating to an attribute characteristic of the triangle, and based on the error metric. A preprocessing unit for calculating an error value for each triangle constituting the object; And a rendering unit which renders the object while first collapsing the triangle having the smallest error value among the obtained error values. delete The method of claim 33, wherein The error about the geometrical characteristics of the triangle is And determine a radius of curvature of the triangle, an area occupied by the triangle in the entire mesh, and whether the triangle is a triangle at a boundary. The method of claim 33, wherein The error about the attribute characteristic of the triangle is And the amount of change in color, the amount of change in normal, and the amount of change in texture between the triangle and the triangles adjacent to the triangle. The method of claim 33, wherein The rendering unit may generate a collapsed vertex to replace the triangle, change a connection relationship of triangles constituting the object based on the collapsed vertex, and generate a graphic image of the object based on the changed connection relationship. Characterized in that the device. An apparatus for generating a graphic model for adaptively converting a graphic image of an object consisting of at least one triangle, An error value calculation unit for determining a collapsing triangle and an order of collapsing triangles from the object in consideration of the degree to which the triangle affects the representation of the object; A collapsed vertex calculation unit for determining a collapsed vertex to be generated by the collapse of the triangle; A triangle identifier node in which the information about the triangles is arranged in order of the collapsed triangle, a level-of-detail sequence node having the sequence information of the collapsed triangle, coordinate information of vertices constituting the object, and And a graphic model generator for generating a coordinate node in which coordinate information is arranged and a collapsed vertex position node indicating a position of the collapsed vertex arranged in the coordinate node. The method of claim 38, The error value calculation unit, An error value is obtained for each triangle constituting the object based on an error metric representing an index in which the triangle constituting the object affects the representation of the object, and the triangle of the triangle to be collapsed so that the triangle with the smaller error value first collapses. Device for determining the order. The method of claim 38, The graphic model generation unit, Between the one level at which the collapse of the triangle is performed and the next level to create an update node for indicating the degree to which the triangle is to be gradually decomposed, and a node for indicating level information at which to start the collapse of the triangle in the original object. Characterized in that the device. The method of claim 38, The graphic model generation unit may include a color node of vertices constituting the object, a color node in which color information of the decay vertex is arranged, and a decay vertex color node representing a position of color information of the decay vertex arranged in the color node; A collapsed vertex normal node representing positions of normal information of vertices constituting the object, normal nodes of the collapsed vertex normal information and normal information of the collapsed vertices arranged in the normal node, vertices constituting the object Further generate at least one of texture coordinate information of the decayed vertex texture coordinate node and a texture coordinate node in which the texture coordinate information of the decay vertex is arranged and the texture coordinate information of the decay vertex arranged in the texture coordinate node. Device characterized in that. The method of claim 38, The graphic model generation unit may include: a color node in which color information of vertices constituting the object excluding color information regarding collapsed triangles is arranged; normal information of vertices constituting the object and normal information of the collapsed vertex are arranged; The normal node and the collapsed vertex normal node indicating the position of the normal information of the collapsed vertices arranged in the normal node, texture coordinate information of the texture coordinates of the vertices constituting the object, and texture coordinate information of the collapsed vertex And at least one of decay vertex texture coordinate nodes indicative of the location of texture coordinate information of the decay vertex arranged in the texture coordinate node. An apparatus for generating a VRML model for adaptively converting a graphic image of an object composed of at least one triangle, An error value calculation unit for determining a collapsing triangle and an order of collapsing triangles from the object in consideration of the degree to which the triangle affects the representation of the object; A collapsed vertex calculation unit for determining a collapsed vertex to be generated by the collapse of the triangle; A coordIndex node in which the information about the triangles is arranged in order of the collapsed triangle, a lodSequence node having the sequence information of the collapsed triangle, coordinate information of vertices constituting the object, and coordinate information of the collapsed vertex And a graphic model generator for generating a coord node, the coordStart node representing the position of the collapsed vertices arranged in the coord node. The method of claim 43, The graphic model generation unit, A normalIndex node in which information about the triangles is arranged in the order of the collapsed triangle, normal information of vertices constituting the object, normal node in which normal information of the collapsed vertex is arranged, the collapsed array arranged in the normal node NormalStart node indicating the position of the vertex normal information, A colorIndex node in which information about the triangles is arranged in order of the collapsed triangle, a color node of vertices constituting the object, a color node in which color information of the decay vertex is arranged, and the decay arranged in the color node A colorStart node indicating the position of the vertex color information, A texcoordIndex node in which information about the triangles is arranged in the order of the collapsed triangle, texture coordinate information of vertices constituting the object, texture coordinate information of the collapsed vertex, and arranged in the texcoord node. And at least one of a texcoordStart node representing the location of texture coordinate information of the collapsed vertex. The method of claim 43, The graphic model generation unit, A normalIndex node in which information about the triangles is arranged in the order of the collapsed triangle, normal information of vertices constituting the object, normal node in which normal information of the collapsed vertex is arranged, the collapsed array arranged in the normal node NormalStart node indicating the position of the vertex normal information, A colorIndex node in which information about the triangles is arranged in order of the collapsed triangle, a color node in which color information of vertices constituting the object except color information about collapsed triangles is arranged; A texcoordIndex node in which information about the triangles is arranged in the order of the collapsed triangle, texture coordinate information of vertices constituting the object, texture coordinate information of the collapsed vertex, and arranged in the texcoord node. And at least one of a texcoordStart node representing the location of texture coordinate information of the collapsed vertex. A computer-readable recording medium having stored thereon a program for performing a method for converting a graphic image of an object consisting of at least one triangle, the method comprising: Determining at least one triangle to be collapsed among the triangles forming the object, Determining the collapse peak to be created by the collapse of the triangle by calculating the collapse peak based on the three vertices of the triangle to be collapsed and the tangent plane of the three vertices; And converting the graphical image of the object based on the determined information about the collapsed triangle and the information about the determined collapsed vertex. delete A computer-readable recording medium having a program recorded thereon that performs a method for adaptively converting a graphical image of an object consisting of at least one triangle, the method comprising: Defining an error metric representing an index in which a triangle constituting the object affects the representation of the object as a sum of an error relating to a geometric characteristic of the triangle and an error relating to an attribute characteristic of the triangle; Obtaining an error value for each triangle constituting the object based on the error metric; And first collapsing the triangle having the smallest error value among the obtained error values. A computer-readable recording medium having recorded thereon a program for performing a method for collapse of a triangle contained in an object represented by a graphic image by being composed of one or more triangles, the method comprising: Defining the normals to the mesh surface of the object for each of the three vertices of the triangle to collapse, Defining, for each of the three vertices of the triangle, a tangent plane that is a plane perpendicular to the normal and including vertices; Projecting the center points of the three vertices of the triangle onto the three tangent planes, respectively, to obtain three projected points; And obtaining a collapsed vertex by calculating an average of the three projected points. A computer-readable recording medium having recorded thereon a program for performing a method for generating a graphic model for adaptively converting a graphic image of an object consisting of at least one triangle, the method comprising: Determining a triangle to collapse from, an order of triangles to collapse, and a collapse peak to be generated by the collapse of the triangle; Creating a triangle identifier node in which information about the triangles is arranged in the order of the triangle to be collapsed; Generating a level of detail sequence node having the sequence information of the triangle to be collapsed; Generating a coordinate node in which coordinate information of vertices constituting the object and coordinate information of the collapsed vertex are arranged; Generating a collapsed vertex position node indicative of the position of the collapsed vertex arranged at the coordinate node. A computer-readable recording medium having a program recorded thereon that performs a method for generating a VRML model for adaptively converting a graphic image of an object consisting of at least one triangle, the method comprising: Determining a triangle to collapse from, an order of triangles to collapse, and a collapse peak to be generated by the collapse of the triangle; Creating a coordIndex node that arranges information about the triangles in order of the triangle to collapse; Creating a lodSequence node having the sequence information of the triangle to be collapsed; Generating a coord node in which coordinate information of vertices constituting the object and coordinate information of the collapsed vertex are arranged; And generating a coordStart node indicative of the location of the collapsed vertices arranged in the coord node.

Images