JP3372034B2 - Rendering method and apparatus, game apparatus, and computer-readable recording medium storing program for rendering stereo model - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and apparatus for rendering a stereo model in a virtual space, and a computer-readable recording medium storing a rendering program.

[0002]

2. Description of the Related Art In recent years, technology relating to computer graphics (CG) has made rapid progress. One area of CG technology being studied is how to do more realistic rendering. With these techniques, more realistic images are being expressed.

On the other hand, other fields of CG technology being studied include non-photorealistic rendering.
endering). This non-photorealistic rendering technique attempts to represent a handwritten image by CG. As one of them, the viewpoint position in the virtual three-dimensional space, the line-of-sight direction or the placement position of the stereo model,
Various researches have been made on an image generation technique for automatically drawing the contour line of the stereo model even when the state such as the direction and the shape is changed.

For example, Japanese Patent Laid-Open No. 7-85310 discloses that
A technique is disclosed in which, when a stereo model is rendered, a side of a polygon forming the stereo model is used as a unit to detect whether or not each side is a contour portion and draw a contour line. Further, Japanese Patent Application Laid-Open No. 7-160905 discloses a technique of drawing a contour line by detecting whether or not each pixel is a contour portion in units of pixels in a display image on which the stereo model is rendered. .

[0005]

As described above, according to the conventional technique, in order to draw a contour line, a process of decomposing a stereo model into edge units of a polygon or pixel units to be rendered and detecting the contour portion is performed. is necessary. Therefore, the process of drawing the contour line is very complicated.

At present, as a non-photorealistic rendering technique, a technique for generating an image such as a cel animation (hereinafter referred to as cel animation) by CG is also desired. This is because it takes a lot of time to create images of various states of a character by human hands in cel animation, but if these images can be generated by CG, the number of times can be greatly reduced. is there.

As a method of rendering a cel animation image by rendering a three-dimensional model, it is conceivable to apply coloring similar to that performed by human hands. In addition, if the contour line of the solid model is drawn, the generated image becomes more cel-animation-like.

Therefore, an object of the present invention is to perform cel-animation-like coloring on a three-dimensional model arranged in a virtual space, draw contour lines of the three-dimensional model by a simple process, and draw the outline of the three-dimensional model in the cel-animation type. A rendering method and apparatus for rendering an image, and a computer-readable recording medium storing a rendering program.

[0009]

According to a first aspect of the present invention, a rendering method for rendering a stereo model placed in a virtual space includes a contour drawing model generated based on the stereo model. A first step of arranging the model at a position including the model, a second step of drawing the inside of the contour drawing model with a predetermined color scheme, and a typical example in which the brightness is divided into levels within a certain range and different for each level. A second lightness distribution in which the lightness is assigned and the first lightness distribution, which is the lightness distribution of the area where the three-dimensional model is to be drawn, is replaced with a representative lightness assigned to each lightness level, and a three-dimensional model. Third, a color distribution is generated based on the color set in advance and the stereo model is drawn with the color distribution.
And steps.

The stereo model is usually rendered outside the table that is visible from the table, whereas the contour drawing model is rendered inside the screen that is not visible from the table.
By drawing the contour drawing model as described above, the contour line is drawn, and by drawing the stereo model as described above, cel animation tone coloring is possible.

In the first aspect of the present invention, the second step described above is a step of drawing the inside of the contour drawing model by mapping a texture having a pattern including a change in brightness or transparency. Such a configuration may be adopted. This makes it possible to make the contour look like a blur line.

A rendering method according to a second aspect of the present invention for rendering a stereo model composed of a plurality of polygons arranged in a virtual space and representing the outside of an object to be represented corresponds to the stereo model. And a first step of obtaining a contour drawing model in which the front and back of polygons corresponding to each polygon of the stereo model are reversed, a second step of arranging the contour drawing model at a position including the stereo model, and a contour drawing The third step of drawing only polygons facing the table to a predetermined viewpoint position in the model with a predetermined color scheme, and the brightness is divided into levels within a certain range, and a typical brightness is assigned to each level. Has been
The first is the lightness distribution of the area where the stereo model is to be drawn.
A color distribution is generated based on a second lightness distribution in which the lightness distribution of is replaced with a representative lightness assigned to each level of lightness and a color preset in the stereo model, and the stereo model is generated using the color distribution. And a fourth step of drawing.

A rendering method according to a third aspect of the present invention for rendering a stereo model composed of a plurality of polygons arranged in a virtual space and representing the outside of an object to be represented corresponds to the stereo model. A first step of obtaining the contour drawing model, a second step of arranging the contour drawing model at a position including the stereo model, and a polygon whose back is directed to a predetermined viewpoint position of the contour drawing model. The third step is to draw only a predetermined color scheme, and the lightness is divided into levels within a certain range, and a typical lightness is assigned to each level. Color based on a second lightness distribution obtained by replacing a certain first lightness distribution with a representative lightness assigned to each lightness level and a color preset in the stereo model. It generates a cloth, and a fourth step of drawing a three-dimensional model in the color distribution.

In the second and third aspects of the present invention, the above-mentioned third step may be configured so as to include a step of mapping a texture having a pattern including a change in brightness or transparency. is there.

A program for rendering a stereo model placed in a virtual space according to a fourth aspect of the present invention includes, in a computer, a contour drawing model generated based on the stereo model and the stereo model. 1st step of arranging in a position, 2nd step of drawing the inside of the contour drawing model with a predetermined color scheme, and the brightness is divided into levels within a certain range, and different typical brightness is assigned to each level. The second lightness distribution obtained by replacing the first lightness distribution, which is the lightness distribution of the area in which the stereo model is to be drawn, with a representative lightness assigned to each lightness level, and the preset lightness distribution set in the stereo model. And a third step of drawing a three-dimensional model with the color distribution.

In the fourth aspect of the present invention, the above-mentioned second step is a step of drawing the inside of the contour drawing model by mapping a texture having a pattern including a change in brightness or transparency. It is also possible.

A program according to a fifth aspect of the present invention for rendering a three-dimensional model composed of a plurality of polygons arranged in a virtual space and representing the outside of an object to be expressed is stored in a computer. A first step of obtaining a contour drawing model in which front and back of polygons corresponding to each polygon of the stereo model are reversed; a second step of arranging the contour drawing model at a position including the stereo model; The third step of drawing only the polygons of the drawing model facing the predetermined viewpoint position with a predetermined color arrangement, and the brightness divided into levels within a certain range, and the typical brightness at each level. Is assigned, and the first lightness distribution, which is the lightness distribution of the area in which the stereo model is to be drawn, is set to the representative lightness assigned to each lightness level. A second lightness distribution instead come, on the basis of a preset color to solid model generates a color distribution,
It is a program for executing a fourth step of drawing a stereo model with a color distribution.

A program according to a sixth aspect of the present invention for rendering a three-dimensional model composed of a plurality of polygons, which is arranged in a virtual space and represents the outside of an object to be represented, is stored in a computer as a three-dimensional model. The first step of obtaining the corresponding contour drawing model, the second step of arranging the contour drawing model at a position including the stereo model, and turning the back to the predetermined viewpoint position of the contour drawing model. The third step of drawing only the existing polygons in a predetermined color scheme, and the brightness is divided into levels within a certain range, and typical brightness is assigned to each level.
The first is the lightness distribution of the area where the stereo model is to be drawn.
A color distribution is generated based on a second lightness distribution in which the lightness distribution of is replaced with a representative lightness assigned to each level of lightness and a color preset in the stereo model, and the stereo model is generated using the color distribution. It is a program for executing the fourth step of drawing.

In the fifth and sixth aspects of the present invention, the third step described above may be configured so as to include a step of mapping a texture having a pattern including a change in brightness or transparency. is there.

The programs according to the fourth to sixth aspects of the present invention are stored in a recording medium or a storage device such as a CD-ROM, a floppy (registered trademark) disk, a memory cartridge, a memory or a hard disk. By thus causing the computer to read the program stored in the recording medium or the storage device, the rendering device and the game device described below can be realized. In addition, the recording medium can be easily distributed and sold as a software product independently of the device. Furthermore, by executing this program using hardware such as a computer, the graphics technology of the present invention can be easily implemented with these hardware.

By causing a computer to execute each step in the rendering method according to the first to third aspects of the present invention, it is possible to obtain the same effect as the rendering method described above. Therefore, by performing the described processing steps by using hardware such as a computer, the rendering technique of the present invention can be easily implemented by these hardware.

A rendering device for rendering a stereo model placed in a virtual space according to a seventh aspect of the present invention places a contour drawing model generated based on the stereo model at a position including the stereo model. Means to place,
A contour drawing model drawing means for drawing the inside of the contour drawing model in a predetermined color scheme, and the brightness is divided into levels within a certain range, and different representative brightness is assigned to each level. A second lightness distribution obtained by replacing the first lightness distribution, which is the lightness distribution of the area to be drawn, with a representative lightness assigned to each lightness level;
And a means for generating a color distribution based on the preset color of the stereo model and drawing the stereo model with the color distribution.

A rendering device according to an eighth aspect of the present invention, which renders a stereo model composed of a plurality of polygons arranged in a virtual space and representing the outside of an object to be represented, corresponds to the stereo model. Further, a means for obtaining a contour drawing model in which the front and back of polygons corresponding to the respective polygons of the stereo model are reversed, a means for arranging the contour drawing model at a position including the stereo model, and a predetermined one of the contour drawing models A means for drawing only polygons facing the table with respect to the viewpoint position in a predetermined color scheme, and the brightness is divided into levels within a certain range, and typical brightness is assigned to each level. , The second brightness distribution in which the first brightness distribution, which is the brightness distribution of the area to be drawn, is replaced with a representative brightness assigned to each brightness level, and To generate a color distribution based on a preset color Le, and means for drawing a three-dimensional model in the color distribution.

A rendering device according to a ninth aspect of the present invention, which renders a stereo model composed of a plurality of polygons arranged in a virtual space and representing the outside of an object to be represented, corresponds to the stereo model. A means for obtaining the contour drawing model, a means for arranging the contour drawing model at a position including the stereo model, and a predetermined polygon of the contour drawing model which faces the predetermined viewpoint position are predetermined. The contour drawing model drawing means for drawing in the given color scheme and the lightness is divided into levels within a certain range, and a typical lightness is assigned to each level, which is the lightness distribution of the area in which the stereo model is to be drawn. A color distribution based on a second lightness distribution obtained by replacing the first lightness distribution with a representative lightness assigned to each lightness level and a color preset in the stereo model. Generated, and means for drawing a three-dimensional model in the color distribution.

In the seventh to ninth aspects of the present invention. The contour drawing model drawing means described above may be configured to include means for mapping a texture having a pattern including a change in brightness or transparency.

A game apparatus for rendering a three-dimensional model arranged in a virtual space according to a tenth aspect of the present invention has a computer and a computer-readable recording medium storing a program to be executed by the computer. . Then, the program has a function of arranging a contour drawing model generated based on the stereo model on a computer at a position including the stereo model, and a contour for drawing the inside of the contour drawing model with a predetermined color scheme. The drawing model drawing function and the first brightness distribution, which is the brightness distribution of the area in which the stereo model is to be drawn, are divided into levels within a certain range and different representative brightness is assigned to each level. And a function of generating a color distribution based on a second lightness distribution replaced with a representative lightness assigned to each lightness level and a color preset in the three-dimensional model, and drawing the three-dimensional model with the color distribution. To implement.

According to an eleventh aspect of the present invention, a game device for rendering a three-dimensional model composed of a plurality of polygons arranged in a virtual space and representing the outside of an object to be represented is a computer and a computer. And a computer-readable recording medium storing a program to be executed. Then, the program has a function of obtaining, in the computer, a contour drawing model corresponding to the stereo model and in which the front and back of polygons corresponding to the polygons of the stereo model are reversed, and a contour drawing model at a position including the stereo model. Of the contour drawing model, the contour drawing model drawing function that draws only the polygons of the contour drawing model facing the predetermined viewpoint position with a predetermined color scheme, and the level within a certain range of brightness. Each level is assigned a representative lightness, and the first lightness distribution, which is the lightness distribution of the area in which the stereo model is to be drawn, is replaced with the representative lightness assigned to each lightness level. A function of generating a color distribution based on the second lightness distribution and a color set in advance in the stereo model and rendering the stereo model with the color distribution is performed.

A game device according to a twelfth aspect of the present invention for rendering a three-dimensional model composed of a plurality of polygons arranged in a virtual space and representing the outside of an object to be represented includes a computer and a computer. And a computer-readable recording medium storing a program to be executed. Then, the program includes, on the computer, a function of acquiring a contour drawing model corresponding to the stereo model, a function of arranging the contour drawing model at a position including the stereo model, and a predetermined viewpoint position of the contour drawing model. A contour drawing model drawing function that draws only the polygons that face the back with a predetermined color scheme,
The lightness is divided into levels within a certain range, and a typical lightness is assigned to each level. The first lightness distribution, which is the lightness distribution of the area in which the stereo model is to be drawn, is assigned for each lightness level. The second which was replaced with typical brightness
The color distribution is generated based on the lightness distribution and the color preset in the stereo model, and the function of drawing the stereo model with the color distribution is executed.

In the tenth to twelfth aspects of the present invention, the contour drawing model drawing function described above includes a function of mapping a texture having a pattern including a change in brightness or transparency. Good.

[0030]

BEST MODE FOR CARRYING OUT THE INVENTION First, the present invention is applied to a computer.
FIG. 1 shows an example of a computer 1000 that executes the computer program when the program is implemented. The computer 1000 includes a computer main body 101, and this computer main body 101
Is the arithmetic processing unit 10 connected to the internal bus 119.
3, memory 105, hard disk drive HDD1
07, sound processing unit 109, graphics processing unit 1
11, a CD-R drive 113, a communication interface 115, and an interface unit 117.

The sound processing section 109 of the computer main body 101 is a sound output device 125 which is a speaker.
In addition, the graphics processing unit 111 is connected to a display device 121 having a display screen 122. Also, CD-R
A CD-R 131 can be mounted on the drive 113. The communication interface 115 is connected to the network 151 via the communication medium 141. Interface section 11
An input device 161 is connected to 7.

The arithmetic processing unit 103 includes a CPU, a ROM and the like, executes a program stored on the HDD 107 or the CD-R 131, and controls the computer 1000. The memory 105 is a work area of the arithmetic processing unit 103. The HDD 107 is a storage area for storing programs and data. When the program executed by the arithmetic processing unit 103 gives an instruction to output sound, the sound processing unit 109 interprets the instruction and outputs a sound signal to the sound output device 125.

The graphics processing unit 111 follows the display device 1 according to the drawing command output from the arithmetic processing unit 103.
A signal for displaying is output on the display screen 122 of 21. The CD-R drive 113 reads / writes programs and data from / to the CD-R 131. The communication interface 115 is connected to the network 151 via the communication medium 141, and communicates with other computers and the like. The interface unit 117 includes an input device 161.
The input from is output to the memory 105, and the arithmetic processing unit 103
Interprets it and performs arithmetic processing.

The program and data according to the present invention are initially stored in, for example, the CD-R 131. Then, this program and data are stored in the CD-R drive 11 at the time of execution.
3 and is loaded into the memory 105.
The arithmetic processing unit 103 processes the program and data according to the present invention loaded in the memory 105, and outputs a drawing command to the graphics processing unit 111. The intermediate data is stored in the memory 105. The graphics processing unit 111 performs processing in accordance with the drawing command from the arithmetic processing unit 103, and outputs a signal for displaying on the display screen 122 of the display device 121.

Next, an example of the graphics processing unit 111 shown in FIG. 1 will be described in detail with reference to FIG. The graphics processing unit 111 communicates with the internal bus 119, a bus control unit 201, a geometric calculation unit 207 that exchanges data with the bus control unit 201, and a triangle drawing processing unit 2.
05, a pixel color processing unit 209 that receives data from the triangle drawing processing unit 205 and performs processing, a Z buffer 211 that stores the Z value of each pixel and is used by the pixel color processing unit 209, and a pixel color processing unit 209. And a frame buffer 213 for storing display screen data. The frame buffer 21
The display signal from 3 is output to the display device 121.

The bus control unit 201 of the graphics processing unit 111 receives the drawing command output from the calculation processing unit 103 via the internal bus 119, and the geometric calculation unit 207 or the triangle drawing process in the graphics processing unit 111. Part 20
Output to 5. Depending on the case, a process for outputting the output of the geometric calculation unit 207 or the triangle drawing processing unit 205 to the memory 105 via the internal bus 119 is also performed. The geometrical operation unit 207 executes geometrical operations such as coordinate conversion, light source calculation, rotation, and reduction / enlargement. The geometric calculation unit 207 outputs the result of the geometric calculation to the triangle drawing processing unit 205.

The triangle drawing processing unit 205 interpolates the data of each vertex of the triangular polygon to generate the data at each point inside the triangular polygon. The pixel color processing unit 209 writes the display image in the frame buffer 213 using the data at each point inside the triangular polygon generated by the triangle drawing processing unit 205. At this time, the pixel color processing unit 209 uses the Z buffer 211 to perform hidden surface removal.

For example, the arithmetic processing unit 103 outputs to the graphics processing unit 111 a drawing command for performing perspective transformation and light source calculation, using information regarding the position and color of each vertex of a triangular polygon in the world coordinate system and the light source as data. In that case, the following processing is executed in the graphics processing unit 111. Upon receiving the drawing command, the bus control unit 201 outputs the command to the geometric calculation unit 207. The geometric calculation unit 207 performs perspective transformation and light source calculation, and calculates the coordinate value (Z
(Including values) and color. The geometric calculation unit 207 outputs the calculation result to the triangle drawing processing unit 205.

The triangle drawing processing unit 205 calculates the coordinate value (including the Z value) and the color at each pixel inside the triangular polygon by using the coordinate value (including the Z value) and the color at each vertex of the triangular polygon. . Further, the triangle drawing processing unit 205 outputs the coordinate value (including the Z value) and the color of each pixel to the pixel color processing unit 209. The pixel color processing unit 209 reads the current Z value of the pixel from the Z buffer 211, and the triangle drawing processing unit 205
It is compared with the Z value output from. If the output Z value is smaller than the current Z value, the pixel color processing unit 2
09 stores the output Z value in the storage position in the Z buffer 211 corresponding to the pixel, and stores the color of the pixel in the storage position in the frame buffer 213 corresponding to the coordinate value of the pixel.

The transparency (α value) may be set for the color of the pixel. In that case, the pixel color processing unit 209 synthesizes the color stored in the storage position in the frame buffer 213 corresponding to the coordinate value of the pixel and the color of the pixel based on the α value. As a result, a composite color is generated. Pixel color processing unit 209
Stores the generated composite color in the same storage location as before.
Particularly in the present invention, the pixel color processing unit 209 also executes a process of storing the color of the pixel in a predetermined position of the frame buffer 213 only when the α value is within a certain range.

Each of the following embodiments is carried out by the computer shown in FIG.

1. Embodiment 1 Next, an outline of Embodiment 1 of the present invention will be described using the functional block diagram of FIG. The rendering apparatus illustrated as the first embodiment includes a contour drawing model drawing unit 350 and a stereo model drawing unit 390. The contour drawing model drawing unit 350 includes a contour drawing model acquisition unit 300.
And a contour setting model placement matrix setting unit 305
A contour drawing model processing unit 310, a blur expression texture mapping unit 315, and a stereo model drawing unit 390.
And a pixel processing unit 330 shared by Each of these functions transfers data in the order described above.

The three-dimensional model drawing unit 390 includes a vertex conversion and light source calculation unit 360, a brightness calculation unit 365, a brightness range table 375, a drawing color calculation unit 370, a brightness range setting unit 380, and a contour. Model drawing unit 350 for drawing
And a pixel processing unit 330 shared by The output of the vertex conversion and light source calculation unit 360 is input to the brightness calculation unit 365. The output of the brightness calculation unit 365 is input to the pixel processing unit 330. The brightness range table 375 is referenced by both the drawing color calculation unit 370 and the brightness range setting unit 380. Drawing color calculation unit 370 and brightness range setting unit 3
The output of 80 is input to the pixel processing unit 330. The pixel processing unit 330 shared by the contour drawing model drawing unit 350 and the stereo model drawing unit 390 includes a brightness comparison unit 333 used in the stereo model drawing process, and a contour drawing model drawing process and a stereo model drawing process. The hidden surface removal processing unit 337 used is included.

The contour drawing model acquisition unit 300 generates a contour drawing model corresponding to a stereo model composed of, for example, triangular polygons. If the contour drawing model is generated in advance, the contour drawing model acquisition unit 3
00 reads out the preliminarily generated contour drawing model composed of triangular polygons. It should be noted that each polygon of the obtained contour drawing model has the opposite side to the corresponding polygon of the stereo model. The contour drawing model is larger than the stereo model and is defined by a predetermined color scheme for the contour line.

Although the contour drawing model must finally be relatively larger than the corresponding stereo model, the size of the contour drawing model at this stage may be the same as that of the stereo model. In this case, the contour drawing model is processed to be drawn relatively larger than the stereo model before the contour drawing model and the stereo model are drawn. Further, the color of the contour drawing model may inherit the color of the material of the corresponding stereo model as it is. In this case, the drawing color is specified separately.

The reference position of the contour drawing model is usually defined to be the same as or near the reference position of the corresponding stereo model. For example, FIG. 4 shows a case where the size of the contour drawing model 510 is defined to be slightly larger than the size of the stereo model 500. In FIG. 4, the arrow direction on each surface indicates the front surface. In the three-dimensional model 500, the outside of each hexagonal surface is the front surface, and in the contour drawing model 510, the inside of each hexagonal surface is the front surface.

The stereo model reference position 520 which is the reference position of the stereo model 500 and the contour drawing model reference position 530 which is the reference position of the contour drawing model 510 are both defined at the center of each model. Also, the contour drawing model 510
Is defined to be slightly larger than the stereo model 500 with the contour drawing model reference position 530 as the center.

Then, the contour drawing model placement matrix setting unit 305 sets a placement matrix for placing the contour drawing model reference position 530 in the virtual space at the same position as the stereo model reference position 520. That is, by setting the arrangement matrix of the contour drawing model 510 so as to include a conversion for translating the contour drawing model reference position 530 to the coordinates of the stereo model reference position 520, a position including the stereo model 500 is set. A contour drawing model 510 is arranged.

The contour drawing model processing unit 310 performs vertex conversion (enlargement / reduction / rotation / translation / perspective conversion) for each vertex of the contour drawing model, and the front and back of each surface of the contour drawing model. Make a decision. The placement matrix described above is also used for this vertex transformation. Note that the light source calculation is not performed here. For example, it is expanded according to the specified state in the virtual space, which is a virtual three-dimensional space.
In addition to the reduction, rotation, translation, and perspective transformation, when the contour drawing model acquisition unit 300 acquires a contour drawing model of the same size as the stereo model, the contour drawing model processing unit 310 Perform vertex conversion to increase the size of the line drawing model. Even when enlarged here, the relationship between the stereo model and the contour drawing model is as shown in FIG.

The front / back surface determination is performed in order to exclude a surface whose direction is the same as the direction of the line of sight 540 from the camera 550, from the drawing target. In the example of FIG. 4, the surface 5 near the camera 550 of the contour drawing model 510
11 and 512 are excluded from the drawing target. In this way, the surface outside the stereo model 500 and close to the camera 550 is removed from the drawing target, so the stereo model 500 is drawn according to the processing of the stereo model drawing unit 390. On the other hand, in the contour drawing model 510, only the faces 513, 514, 515 and 516 behind the stereo model 500 are drawn. However, since the hidden surface removal processing unit 335 of the pixel processing unit 330 performs hidden surface removal, not all of the surface is drawn even if it is a drawing target.

Blurred expression texture mapping unit 315
Performs a process for mapping the blur expression texture on the contour drawing model so that the resulting contour line becomes faint. This faint expression texture is a texture having a pattern including a change in brightness or transparency, an example of which will be shown later. Since the contour line does not necessarily have to be blurred, the processing of the blurred expression texture mapping unit 315 is selectively performed.

The vertex conversion and light source calculation unit 360 of the three-dimensional model drawing unit 390 performs the vertex conversion (enlargement / reduction / rotation / translation / perspective) for each vertex of the triangular polygons forming the three-dimensional model arranged in the virtual three-dimensional space. (Conversion) is performed (the area in which the triangular polygon is drawn is calculated), and the light source is calculated for each vertex of the converted triangular polygon. In addition, the vertex conversion and light source calculation unit 360 of the stereo model drawing unit 390 also determines the front and back of each triangular polygon of the stereo model. The light source calculation is to calculate a shadow (luminance) generated by a virtual ray emitted from the light source.

Also in the vertex conversion and light source calculation section 360 of the stereo model drawing section 390, enlargement / reduction / rotation / translation / movement are performed in accordance with the specified state in the virtual three-dimensional space.
In addition to perspective conversion, the contour drawing model processing unit 31
When the contour drawing model after being processed with 0 is the same size as the stereo model, the size of the stereo model is reduced so that the stereo model becomes relatively smaller than the contour drawing model. Vertex transformation is performed.

The relationship between the stereo model 500 and the contour drawing model 510 is as shown in FIG. 4 even when the vertex conversion and light source calculation section 360 performs the stereo model reduction processing. In addition, the surface front / back determination is the same as that of the contour drawing model processing unit 310, and excludes, from the drawing targets, the surface of the three-dimensional model whose direction, which is the same as the line-of-sight direction of the camera, is the front surface. In the example of FIG. 4, the rear surfaces 503, 50 viewed from the camera are shown.
4, 505 and 506 are excluded from the drawing target.

The brightness calculator 365 calculates the brightness from the color at each vertex of the triangular polygon calculated by the vertex conversion / light source calculator 360. Normal vertex conversion and light source calculation unit 3
Since 60 calculates the color in the RGB system, the lightness calculation unit 365 obtains the lightness Y by YIQ converting this RGB.
The brightness at each vertex of this triangular polygon is output to the pixel processing unit 330.

The brightness range table 375 is a table as shown in FIG. 5, for example. That is, in the table in which the threshold value and the reference lightness are paired, the reference lightness is 0.75 for the threshold value 0.75, the reference lightness is 0.50 for the threshold value 0.50, the threshold The standard brightness is set to 0.25 with respect to the value of 0.00, and three levels are set. In addition,
Here, the brightness takes a real value from 0 to 1. It is also possible to specify the range by the upper limit and the lower limit instead of the threshold value (see, for example, FIG. 18). With reference to the brightness range table 375, the drawing color calculation unit 370 calculates the drawing color corresponding to each threshold value. The drawing color corresponding to each threshold value is calculated using the reference lightness corresponding to the threshold value and the color information preset for each triangular polygon of the stereo model. The drawing color calculation unit 370 outputs the calculated drawing color to the pixel processing unit 330.

The brightness range setting section 380 selects one threshold value of the brightness range table 375 and sets it in the pixel processing section 330. When using the brightness range table 375 as shown in FIG. 5 as it is, the brightness range setting unit 380 sets the brightness range table 375 one by one from the top. If upper and lower bounds are specified instead of thresholds,
It can be randomly selected and set.

The pixel processing unit 330 shared by the contour drawing model drawing unit 350 and the stereo model drawing unit 390 is
The color or brightness of each pixel in the triangular polygon is obtained by interpolating the color or brightness of each vertex of the triangular polygon. The interpolation method may be the Gouraud shading algorithm or the von shading algorithm.

When processing the triangular polygons which are the drawing targets of the contour drawing model, the pixel processing unit 330 uses the hidden surface removal processing unit 337 to perform the hidden surface removal processing, and at the same time, the contour drawing model. The color of each pixel in the triangular polygon that is the drawing target of is determined.

For example, in the case of FIG. 4, the three-dimensional model 500.
Two surfaces 501 and 502 that are closest to the camera 550 of the contour drawing model are drawn farther from the camera 550 of the contour drawing model.
Two faces 513, 514, 515 and 516 are drawn. Since these four surfaces of the contour drawing model 510 are projected to the left and right of the stereo model 500 when viewed from the camera 550, only the protruding portion is drawn without hidden surface removal. This protruding portion becomes the contour line. The pixel processing unit 330 determines the color in consideration of the color of the material of the contour drawing model. However, the color of the contour line (black or dark contour line color) may be set as the color of the contour drawing model by completely ignoring the color of the material.

On the other hand, when processing the triangular polygon which is the drawing target of the three-dimensional model, the pixel processing unit 330 first interpolates the brightness at each vertex of the triangular polygon output from the brightness calculation unit 365 to obtain the polygon. The brightness (brightness distribution in the polygon) of each pixel inside is calculated.

Then, the brightness comparing section 333 compares the brightness of each pixel with the threshold value set by the brightness range setting section 380. If the brightness of the pixel is greater than or equal to the threshold value, the pixel processing unit 330 draws the pixel with the drawing color based on the reference brightness corresponding to the threshold value.
At the time of this drawing processing, hidden surface removal processing is also performed using the hidden surface removal processing unit 337. If the brightness at that pixel is less than the threshold, then this pixel is not drawn at this stage. The brightness range setting unit 380 sets the brightness range table 37.
All the threshold values of 5 are set in the pixel processing unit 330,
Correspondingly, if the pixel processing unit 330 performs drawing processing on all the pixels in the triangular polygon, the inside of the triangular polygon is divided into three stages in the example of FIG. This process is executed for all triangular polygons of the three-dimensional model.

In the process described above, the drawing color is calculated in advance by the drawing color calculation unit 370, so it is difficult to understand. However, in the above process, the lightness of each pixel (lightness distribution in the polygon) The drawing color generated by replacing the reference lightness corresponding to the lightness range to which the lightness belongs (a second lightness distribution in the polygon is generated) and the preset color of the polygon and its reference lightness (color distribution in the polygon ) Is substantially the same as the drawing process.

In the present invention, hidden surface removal processing by the Z buffer may be necessary for hidden surface removal processing. For example, when the stereo model has a human shape and the arm is located in front of the torso, the Z-buffer method is not used from the positional relationship between the contour drawing object and the plane of the stereo model. This is because it may be difficult to draw.

In addition, in the stereo model drawing unit 390, as shown in FIG.
When the lightness range table 375 as described above is used as it is, the hidden surface removal by the Z buffer is used. For example, according to FIG. 5, the lightness of 0.75 or more is 0.5 or more and 0.0 or more, so it is necessary to set the upper limit value of the lightness range so that the drawing color is not overlaid. If the brightness of a pixel is 0.75 or more, this pixel is drawn with a drawing color corresponding to the threshold value 0.75,
The Z value of that pixel is stored in the Z buffer.

When the threshold value reaches 0.5, the Z value of the pixel is read from the Z buffer and compared with the Z value of the same pixel to be written, but naturally they are the same. Therefore, the drawing color corresponding to the threshold value 0.5 is not written to the frame buffer for that pixel.
The same applies to the threshold value 0.0.

The lightness at the vertices of the polygon and the pixels inside the polygon are treated as attribute values of a color (RGB) usually used as transparency. Usually α
Since the value is defined in the range of 0 to 255, the attribute value α actually used is 255 times the lightness. Therefore,
Threshold of brightness range table 375 (upper limit and lower limit)
May be a value in the range 0-255.

Next, the processing flow of the first embodiment will be described. The processing described below is performed by the arithmetic processing unit 103.
(FIG. 1) is a process executed by controlling other elements in the computer main body 101.

[CD-R Recording Processing] FIG. 6 shows the contour drawing model generation processing which is performed in advance. When the processing is started, the data of the stereo model stored in the HDD 107 in advance is read (step S303) and is acquired as the conversion target model.

Next, the size of the model to be converted is enlarged so as to be slightly larger (step S30).
5). For example, in the normal direction of each vertex of the conversion target model,
The vertex is moved by a length of 2% of the total length of the conversion target model, and is expanded by about 2% as a whole. That is, for example, if the conversion target model is a human type and the height thereof is equivalent to 1.8 m, each vertex is 0.0
It is moved by a length corresponding to 36 m. When the enlargement ratio is larger, the contour line is drawn thicker, and when the enlargement ratio is smaller, and the conversion target model is only slightly enlarged, the contour line is drawn thinner. Further, if it is not uniform and a part is enlarged, only the outline of the enlarged portion is drawn thick. Since this size adjustment is usually performed by the creator of the three-dimensional model, it is possible to draw a contour line that reflects the creator's intention.

When the normal line of each vertex of the three-dimensional model is not defined, the normal line of the vertex which is obtained by interpolating the normal line of each surface sharing the vertex is used, Can also be moved in the direction of the normal to the vertex. Also, the surface can be moved in the normal direction of each surface of the three-dimensional model. However, when the surfaces are simply moved, a gap is created between the surfaces, and a separate process for filling the gap is required. Further, since the reference position is usually defined in the stereo model, each vertex of the conversion target model can be moved around the reference position of the corresponding conversion target model.

Next, the color of the material of each polygon of the model to be converted is set to a color having the same saturation but lower brightness (step S307). Note that all polygons may be set to a single color such as black. Further, the setting for mapping the texture for expressing blur may be performed. Since the color of the material is adjusted by the manufacturer, the contour line can be drawn in the color intended by the manufacturer.

Next, the front and back of each polygon of the conversion target model are reversed (step S309). Specifically, the order in which the vertices of each triangle forming the conversion target model are defined is changed at one place. The details of the front / back determination method will be described later.

The data of the conversion target model converted so far is used as the contour drawing model data in the HDD 107.
(Step S311), and the contour drawing model generation process ends (step S313).

Next, various data stored in the HDD 107, including the stereo model data and the contour drawing model data, are transferred to the CD-R 13 by the CD-R drive 113.
Written to 1. FIG. 7 schematically shows an example of data written in the CD-R 131.

The program area 132 stores a program for causing the computer 1000 to carry out the present invention. However, the program for implementing the present invention is a CD-
Processing until writing to R131 and FIG.
Can be divided into the processing shown in. Therefore, the contour drawing model described above is generated and the contour drawing model
A program for writing various data including data to the CD-R 131 may not be included here.
By doing so, the processing shown in FIG.
It can also be implemented by a computer other than the computer 1000, which has a CD-ROM drive instead of the D-R drive 113.

The system data area 133 stores various data processed by the program stored in the program area 132 described above. The image data area 134 stores data including the stereo model data 137 and the contour drawing model data 135. However, when the contour drawing model is generated in the contour drawing model acquisition process described later, the contour drawing model data 135 need not be stored. It should be noted that data such as texture that expresses fading is also stored in the image data area 134.

In the sound data area 136, the sound output unit 1 by the sound processing unit 109 shown in FIG.
Data for outputting sound from 25 is stored. Since sound processing is not directly related to the present invention, it is not necessary to store data in the sound data area 136.

The size of the contour drawing model stored in the CD-R 131 may be defined as the same size as the size of the corresponding stereo model. In this case,
After the contour drawing model is acquired by the contour drawing model acquisition processing described below, the contour drawing model is expanded until the contour drawing model arrangement matrix is set by the contour drawing model arrangement processing described below. To be done. Alternatively,
When the placement matrix of the contour drawing model is set in the contour drawing model placement process, the placement matrix may be determined so that the placement matrix includes the expansion conversion. Conversely, when arranging the three-dimensional model, the three-dimensional model arrangement matrix may be determined such that the three-dimensional model arrangement matrix includes reduction conversion.

The color of the material on each surface of the contour drawing model stored in the CD-R 131 may be the same as the color of the material on each surface of the corresponding stereo model. In this case, the contour drawing model is drawn in a separately defined color such as black in the drawing process of the contour drawing model described later.

[Overall Processing Flow] FIG. 8 shows the first embodiment.
The whole processing flow of is shown. When the process starts, initial setting is performed (step S2). This initial setting processing includes data acquisition processing for the contour drawing model (FIG. 9), which will be described in detail later, and data acquisition processing for the stereo model to be drawn. Then, the state in the virtual space is set (step S3). This process is, for example, a process of changing the state in the virtual space in response to a change in the position of the viewpoint, a change in the position of the light source, a movement of the model, or a deformation of the model. By performing this processing, the position coordinates of the stereo model and the contour drawing model
The determination processing of the direction, the enlargement ratio, the reduction ratio, etc. is performed. More specifically, the process of determining the placement matrix (used in FIG. 11) for the stereo model and the contour drawing model is performed. Further, according to a key operation of the input device 161 (FIG. 1) or the like, the setting as to whether or not contour line drawing is performed in step S4 is performed in step S3.

Next, a judgment process as to whether or not to draw the contour line is performed (step S4). This is determined based on the setting by the key operation of the input device 161, or the setting by another program as described above. When it is determined that the contour line is drawn, the drawing process of the contour line drawing model is executed (step S5). This will be described later with reference to FIG. Then, the drawing process of the stereo model is executed regardless of whether the contour line is drawn or not (step S6). This processing will be described later with reference to FIG. The steps S3 to S6 are repeatedly executed until the processing is completed (step S7). Whether or not the processing is ended is determined by whether or not an operation input indicating that the processing should be ended is performed.

[Contour drawing model acquisition process] FIG.
The acquisition processing of the contour drawing model is shown. Here, first, it is determined whether or not a contour drawing model is generated (step S203). This is because there are cases where the contour drawing model is prepared in advance and cases where the contour drawing model is generated at this stage. Here, this judgment is made, for example, when the contour drawing model corresponding to the stereo model is CD-R1.
It is carried out by determining whether or not it is stored in 31. If it is determined that the contour drawing model is stored, it is determined that the contour drawing model is not generated, and if it is determined that the contour drawing model is not stored, it is determined that the contour drawing model is generated.

If it is determined that the contour drawing model is not generated, the data of the contour drawing model stored in the CD-R 131 is read (step S20).
7). Each surface of the contour drawing model has its front and back inverted from the corresponding surface of the stereo model, as described above with reference to FIGS. 4 and 6. Further, the size of the contour drawing model read out is defined to be slightly larger than that of the corresponding stereo model. Further, the color of the contour drawing model is defined as a color darker than the corresponding stereo model.

If it is determined that the contour drawing model is generated, the processing for generating the contour drawing model is performed (step S205). Similar to step S207, even when the contour drawing model is generated at this stage, each polygon of the contour drawing model is
As described above with reference to FIG. 4, the polygon corresponding to the stereo model is turned upside down.

The size of the contour drawing model is slightly larger than that of the corresponding stereo model. Step S30
As in the case of FIG. 5 (FIG. 6), for example, a contour drawing model is generated by moving the vertices of the stereo model in the normal direction of the vertices and enlarging the contour drawing model. When the contour drawing model is larger than the stereo model, the contour line is drawn thicker, and when the contour drawing model is slightly larger than the stereo model, the contour line is drawn thinner.

Further, as described in the description of step S305 (FIG. 6), even if the surface is moved in the normal direction of each surface of the stereo model, the enlarged contour drawing model is generated. Good. Furthermore, the contour drawing model that is enlarged by moving each vertex of the stereo model around the reference position normally defined in the stereo model may be generated.

At this point, the size of the contour drawing model may be the same as the size of the corresponding stereo model. In this case, after the contour drawing model is acquired by the contour drawing model acquisition processing, the contour drawing model is arranged by the contour drawing model arrangement processing described later until the arrangement matrix of the contour drawing model is set. Is expanded. Alternatively, when the placement matrix of the contour drawing model is set in the contour drawing model placement process, the placement matrix may be determined so that the placement matrix includes the expansion conversion. Conversely, when arranging the three-dimensional model, the three-dimensional model arrangement matrix may be determined such that the three-dimensional model arrangement matrix includes reduction conversion.

On the other hand, the material color of each polygon of the contour drawing model is generated by darkening the material color of each polygon of the corresponding stereo model. Note that, as described in the description of step S307 (FIG. 6), the color of the contour drawing model to be generated does not have to be defined at this point. Alternatively, the material color of each polygon of the contour drawing model may be the same as the material color of each polygon of the corresponding stereo model. In this case, the color of the contour drawing model is not taken into consideration during the drawing process of the contour drawing model, and the contour drawing model is drawn with a separately defined color such as black or a texture color that expresses fading. To be done.

Next, it is determined whether or not a texture expressing blur is mapped on the contour drawing model (step S209). When the contour drawing model is generated in step S205, the determination is performed based on the data of the corresponding stereo model. On the other hand, if the contour drawing model is read in step S207,
The determination is performed based on the read data of the contour drawing model. When it is determined that the texture expressing the blur is mapped, step S211 is performed.
At, the texture expressing the blur is mapped on the contour drawing model. That is, texture coordinates (U, V) are set at each vertex of the polygon.

As described above, the texture expressing the blur has a pattern including a change in brightness or transparency. FIG. 10 shows an example of a texture including a change in brightness. This is a texture with a pattern in which white diagonal lines are finely drawn on a black background. The brightness of the black part is low,
Since the brightness of the white part is high, the texture shown in FIG. 10 includes a change in brightness.

In the present invention, the contour is drawn by cutting out a part of the contour drawing model as a line. That is, when the contour drawing model to which the texture of FIG. 10 is mapped is drawn as a contour line, a line corresponding to the line cut out as a contour line from the contour drawing model is drawn from the texture. At this time, if a line is cut out from the texture in a substantially vertical direction or a substantially horizontal direction, each line will include a change in brightness. By drawing such a line as a contour line, a contour line including a change in brightness is drawn. That is, blurring of the contour line is expressed, and a more handwritten contour line is drawn.

With the texture shown in FIG. 10, no matter which direction the line is cut out, the line includes a change in brightness. However, the brightness may not change depending on the cutting direction. Since it is possible to adjust which part of the contour drawing model is drawn as a contour line in which direction, the texture of the blur is mainly adjusted according to the cutting direction.

When a contour line is drawn by a contour drawing model in which a texture having a pattern including a change in transparency is mapped, the contour line includes a change in transparency. A color close to the background color is drawn in the high transparency portion in accordance with the ratio, and a color close to the texture color such as black is drawn in the low transparency portion. As a result, a contour line including changes in shading is drawn, and blurring of the contour line is expressed.

When it is determined that the texture expressing the blur is not mapped, and when the texture mapping process is completed, the arithmetic processing unit 103 ends the contour drawing model acquisition process (step S21).
3).

[Contour Drawing Model Arrangement Processing] In step S3 of FIG. 8, the arrangement matrix of the contour drawing model is set, and the arrangement processing of the contour drawing model is performed.
The reference position of the normal contour drawing model is provided at a position corresponding to the reference position of the stereo model. Then, the arrangement matrix of the contour drawing model is set such that the reference position of the contour drawing model is arranged at the same position as or near the reference position of the stereo model.

Here, when the direction of the stereo model changes, the arrangement matrix including the rotation conversion is set so that the contour drawing model also corresponds to it. When the shape of the stereo model changes, the transformation process is performed so that the contour drawing model corresponds to it.

At this stage, if the contour drawing model has the same size as the corresponding stereo model, the contour drawing model is enlarged. Specifically, the arrangement matrix of the contour drawing model is set so that the vertices of the contour drawing model are enlarged and converted according to a predetermined enlargement ratio with the reference position of the contour drawing model as the center. Alternatively, conversely, the stereo model may be reduced. That is, in this case, the arranging matrix of the stereo model is set so that the vertices of the stereo model are reduced and converted with a predetermined reduction ratio with the reference position of the stereo model as the center.
When the contour drawing model is relatively larger than the corresponding stereo model, the stereo model arranging matrix can be used as it is as the contour drawing model arranging matrix.

In this way, the relatively large contour drawing model is finally arranged so as to include the stereo model. The contour drawing model may not completely include a stereo model depending on the arrangement position, direction, shape, and the like of both models. However, even in such a case, a contour line is drawn for the included portion.

At this stage, it is not always necessary to set the arrangement matrix, and the arrangement coordinates,
It suffices that each element necessary for the vertex conversion such as the direction and the enlargement / reduction ratio is fixed. Also in this case, the actual vertex conversion is performed at the stage of drawing processing of each model.

[Contour Drawing Model Drawing Processing] In FIG. 11 showing the drawing processing flow of the contour drawing model, the processing described below is repeated until all the vertices of the contour drawing model are processed (step S503). ). The first process that is repeatedly performed is a vertex conversion (enlargement / reduction / rotation / translation / perspective conversion) process for one vertex (step S505). Here, the arrangement matrix of the contour drawing model obtained in step S3 is also used in the vertex conversion.

For example, this processing is carried out by the geometric calculation section 207 instructed by the calculation processing section 103. It should be noted that the geometric calculation unit 20 is used for the contour drawing model.
7 is that light source calculation is not performed. This is because the contour line is drawn regardless of the position of the light source and it is useless to calculate the light source. For example, the material color of the contour drawing model may be ignored. Normally, this vertex transformation is performed based on the state specified in the virtual three-dimensional space. However, if the size of the contour drawing model is the same as the stereo model, according to the placement matrix set in the placement process. At this stage, the contour drawing model may be enlarged and converted.

Then, it is determined whether or not the polygon including the vertex is the front surface (step S507).
In the case of a triangular polygon, this determination is made based on which direction the triangular polygon composed of this apex and the two apexes processed before it is facing. FIG. 12 shows an example of triangular polygons forming a three-dimensional model for explaining the front / back determination method. In this example, the vertex number of the top vertex in the figure is 0, the vertex number of the lower left vertex is 1, and the vertex number of the lower right vertex is 2.
That is, the vertex numbers are given counterclockwise from the top vertex.

In the first embodiment, the surface in which the vertex numbers of the respective vertices of the triangular polygon appear to be given counterclockwise is defined as the front surface (so-called right-handed system). Therefore, the front side of the triangular polygon in FIG. 12 is the front side of the drawing. If there is a normal vector in the direction of the front surface, the front and back of the triangular polygon can be determined by the sign of the inner product of the normal vector and the line-of-sight vector. That is, if the sign of the inner product is negative, it means that the front surface is facing the viewpoint position, and if the sign of the inner product is positive, it means that the back surface is facing the viewpoint position. Become.

Actually, as shown in FIG. 13, the vector a projected from the vertex 0 to the vertex 1 is projected on the screen.
Then, the outer product a × b of the vector b from the vertex 0 to the vertex 2 projected on the screen is calculated, and it is determined whether or not the surface is the front surface in the direction of the vector n which is the result of this outer product. The vector n is parallel to the z-axis, and if the sign of the z component of the vector n is checked, it can be determined whether it is the front surface. That is,
Positive is hospitality, negative is back. On the left side of FIG. 13, the numbers of the vertices of the triangle are counterclockwise, and the vector n, which is the result of the outer product, is in the positive direction of the z axis, which is the front side. On the other hand, on the right side of FIG. 13, the numbers of the vertices of the triangle are clockwise, and the vector n resulting from the outer product is oriented in the negative direction of the z axis.

In the case of the contour drawing model in the first embodiment, the polygons of the contour drawing model are opposite to the polygons corresponding to the stereo model. FIG. 14 shows polygons corresponding to the polygons in FIG. 12 and whose front and back surfaces are reversed. To the vertices of the triangular polygon shown in FIG. 14, vertex numbers 0, 1 and 2 are given in the order of upper, lower right and lower left in the figure. That is, the corresponding triangular polygons are assigned vertex numbers in the reverse order of FIG. Therefore, in FIG. 14, the front side of the page is determined to be the back side. In the first embodiment, the front / back determination is performed at this stage, but the front / back determination may be performed before this stage.

If the polygon including the apex is the back surface, the process returns to step S503. If the polygon including the apex is the front surface, it is determined whether or not the texture expressing blur is mapped (step S509).

This means texture mapping for polygons. If the texture expressing the blur is mapped, the texture coordinate of the texture for expressing the blur is calculated for the vertex (step S511). When texture mapping is performed, the texture coordinates (U, V) have already been specified at the vertices of the polygon, but if the polygon is arranged diagonally to the screen, the texture will be distorted on the screen. May be displayed. In order to avoid this distortion, here, as the texture perspective processing, calculation of S = U × Q and T = V × Q is performed using Q = 1 / w (w is the depth from the screen). If the texture expressing the blur is not mapped, the process proceeds to step S513.

Then, for example, the triangle drawing processing unit 205 and the pixel color processing unit 209 shown in FIG. 2 are driven (step S513). As described above, the triangle drawing processing unit 205 interpolates the data of each vertex of the triangular polygon to generate the data of each pixel inside the triangular polygon. The data of each vertex is the color of the material, the screen coordinate value, and the texture coordinate value if step S511 is performed. The data in each pixel is the color of the material and the texel color if step S511 is performed.

However, it is possible to ignore the material color at this point and set the color of the contour line at each vertex.
It is also possible to set the color of the contour line in consideration of the color of the material. The pixel color processing unit 209 writes the display image in the frame buffer 213 using the data in each pixel inside the triangular polygon generated by the triangular drawing processing unit 205. At this time, the Z buffer 21
Use 1 to perform hidden surface removal.

Although an example in which the Z buffer 211 is used for hidden surface removal is shown, for a simple model as shown in FIG. 4, hidden surface removal processing such as the Z sort method is performed without using the Z buffer. You may implement. However, in a more complicated model, for example, when a person's hand is placed in front of the body, it is difficult to draw the contour line accurately unless hidden surface removal using the Z buffer is performed. is there.

[Stereoscopic Model Drawing Process] FIG. 15 shows a flow of the stereoscopic model drawing process in the first embodiment. First, initialization is performed (step S603). In this initial setting, the brightness range table (for example, FIG. 5 or FIG. 18) corresponding to the stereo model is acquired. Next, vertex conversion (enlargement / reduction / rotation / translation / perspective conversion) and light source calculation for one vertex are performed (step S60).
5). Here, the matrix for placement of the three-dimensional model is also used in the vertex conversion. This is executed by the geometric calculation unit 207 according to a command from the calculation processing unit 103, for example. The data of the stereo model is stored in the CD-R 131, for example.

The enlargement / reduction / rotation / translation / perspective transformation is basically based on the state in the virtual space set in step S3 of FIG. However, if the contour drawing model has the same size as the stereo model, the contour drawing model may be made relatively large by reducing the size of the stereo model. In this case, reduction conversion is performed in step S605.
It should be noted that the size can be easily reduced by moving each vertex along the normal line toward the center of the three-dimensional model. Here, the perspective transformation is to transform the coordinate value of each vertex of the polygon in the world coordinate system into the coordinate value in the screen coordinate system. The light source calculation is to calculate a shadow (luminance) generated by a virtual light ray emitted from the light source.

There are two methods for light source calculation in the stereo model drawing processing of the present invention. (A) A method that considers the color of the material defined in the polygon, and (B) a method that does not consider the color of the material. In the case of (A), it calculates by the following formula.

[0115]

[Equation 1]

However, Pn0, Pn1, Pn2, Nnx, Nny, N
n of nz, Pnr, Png, Pnb, Cnr, Cng, and Cnb has shown the nth vertex. Nnx is the x component of the normal at the nth vertex, Nny is the y component of the normal at the nth vertex,
Nnz is the z component of the normal at the nth vertex. Ligh
tMatrix is a matrix created by the normalized light source vector. The case where this can define up to three parallel light sources is shown below. Further, LColorMatrix has as a component the color of the light ray emitted from the light source, and the case where up to three light sources can be defined is shown below. M means the color of the material of the polygon, and rgb shows its component. The output in the case of (A) is Cnr, Cng, and Cnb.

[0117]

[Equation 2]

[0118] _{However, L 0x, L 0y, L} 0z is a component of the normalized light vector _{0, L 1x, L 1y,} L 1z are components of the normalized light vector _{1, L 2x, L 2y,} L _2z is a component of the normalized light source vector 2. Further, the color of the light ray of the light source vector 0 is LC _0r , LC _0g , LC _0b as a component, and the color of the light ray of the light source vector 1 is LC _1r , LC _1g , LC _1b as a component and the light ray vector 2 is The color of the light is LC _2r , L
C _2g and LC _2b are the components. Each color component is 0.0
It takes a value between 1 and 1.0. For example, when only the light source 0 exists and white light is used at an angle of 45 degrees with respect to the XYZ axes, the following matrix is obtained.

[0119]

[Equation 3]

In the case of (B), the following formula is used for the calculation.

[0121]

[Equation 4]

The results of the two formulas are naturally different,
The calculation result of (A) is correct. However, since the calculation amount of (B) is smaller than that of (A), the processing speed can be increased. The quality of the normal image does not change.

Next, it is determined whether or not the polygon including the vertex is the front surface (step S607). In the case of a triangular polygon, this judgment is performed depending on which direction the triangular polygon composed of this apex and the two apexes processed before is directed. For this determination, the method described in the drawing process of the contour drawing model can be used. In the first embodiment, the front / back determination is performed at this stage, but the front / back determination may be performed before this stage.

If the polygon including the apex is the back surface, the process returns to step S605. If the polygon including the apex is the front surface, the brightness at the apex subjected to the apex conversion and the light source calculation is calculated (step S609). YIQ conversion is performed in the brightness calculation. When the color at the apex is obtained by the method (A) described above, it is calculated by the following formula.

[0125]

[Equation 5]

When the color at the apex is obtained by the method (B) described above, it is calculated by the following equation.

[0127]

[Equation 6]

The matrix containing numerical values is the first row of the 3 × 3 matrix for conversion from RGB to YIQ. As a precaution, the 3 × 3 matrix (conversion matrix) is shown below.

[0129]

[Equation 7]

FIG. 16 shows the data structure of a stereo model before perspective transformation. FIG. 16A shows the data structure of the stereo model, and there are N triangular polygons in total. As shown in FIG. 16B, each triangular polygon has three material colors (YIQ) and three vertex data indexes (IDX). Here, the color of the material is assumed to have the YIQ system, but it may have the color of the RGB system. By using the vertex data IDX, it is possible to obtain information regarding the vertices from the vertex data table shown in FIG.

In the vertex data table, for each vertex data IDX, the three-dimensional coordinates (P _nx , P _ny ,
P _nz ) and a normal vector (Nnx, Nny, Nnz) are stored (n is a vertex number). When perspective transformation is performed, the data structure of the triangular polygon changes. FIG. 17 shows the one corresponding to FIG. For each vertex, the coordinate value (x, y, z) in the screen coordinate system, the color (r, g, b) and the α value at the vertex are stored. The brightness calculated in step S609 is stored in the area in which the α value is stored. Further, as will be described in detail below, when the triangle drawing processing unit 205 performs processing, drawing colors corresponding to the lightness range are stored in the three vertex common colors (r, g, b). Although the lightness is in the range of 0.0 to 1.0, the α value is an integer from 0 to 255, and therefore the lightness multiplied by 255 is used as the α value.

Returning to FIG. 15, the processing flow will be described. After step S609, the drawing color of the polygon including the vertices subjected to the vertex conversion and the light source calculation is calculated (step S609).
611). The drawing color of the polygon is calculated from the reference lightness corresponding to each lightness range stored in the lightness range table and the color of the polygon. For example, if the color of the polygon is YIQ
If the color is retained as a system color, IQ out of YIQ
Is calculated using the following formula using each reference lightness Tn.

[0133]

[Equation 8]

If there are three reference luminosities Tn (T1, T2, T
3), three drawing colors are obtained. The color of the polygon is Y
When not stored as IQ system colors, that is, RGB
When the color is held as a system color, the conversion from RGB to YIQ is performed using the conversion matrix shown above. Further, although the calculation result is different, when it is necessary to calculate the drawing color at high speed, the following calculation is performed.

[0135]

[Equation 9]

M means the color of the material of the polygon. Although the calculation results are different and the image quality is slightly different in the above two equations, the second one can obtain an almost similar image at high speed.

Next, one brightness range in the brightness range table is selected (step S613). In this embodiment, the lightness range table shown in FIG. 5 is used, but FIG.
It is also possible to use a brightness range table such as Figure 1
8 shows an example of a table in which the lightness range is specified by the upper limit and the lower limit. That is, the upper limit of the brightness range is 1.00
And a lower limit of 0.75, a reference brightness of 0.75, and an upper limit of the brightness range of 0.74 and a lower limit of 0.50, a reference brightness of 0.5.
A reference brightness of 0.25 is set to 0 and an upper limit of 0.49 and a lower limit of 0.00. When such a brightness range table is used, a brightness range including an upper limit and a lower limit can be randomly selected and set. However, FIG. 18 shows a case where the computer is effective up to the second decimal place. In the lightness comparison process described below, when the lightness of each pixel cannot be easily compared with the two lightness values of the upper limit and the lower limit, the lightness range is selected in order from the top of FIG. 18, for example. In this case, only the lower limit value is processed.

After that, the lightness at the vertex of this polygon is interpolated, and the lightness at each pixel inside the polygon (lightness distribution in the polygon) is calculated. The colors of the vertices are also interpolated, but the results are the same even if they are interpolated because the drawing colors are the same for all three vertices. If the lightness of the pixel is within the selected lightness range, the pixel is drawn with the drawing color corresponding to the selected lightness range (step S615). The brightness interpolation processing is executed by the triangle drawing processing unit 205 in FIG. 2, for example. The comparison process of whether the brightness of each pixel is within the selected brightness range is
For example, the pixel color processing unit 209 implements it. The steps S613 and S615 are repeated until processing is performed for all the brightness ranges (step S617).

For example, when the pixel color processing unit 209 cannot handle the two lightness values of the upper limit and the lower limit, the same effect can be obtained by using the Z buffer 211 together. The Z buffer 211 is used for hidden surface removal, but in the present embodiment, the combined use of the Z buffer 211 has the same effect as when compared with the upper limit value of brightness.

For example, when there is a brightness range table as shown in FIG. 5, the threshold value 0.75 is selected first. Then, in response to a command from the arithmetic processing unit 203, the triangle drawing processing unit 205 interpolates the brightness and coordinates (including Z value) of each vertex of the polygon, and calculates the brightness and coordinates (including Z value) of each pixel. Go. If the colors of all three vertices are set to the drawing color corresponding to the threshold value 0.75, the color of each pixel becomes the drawing color even if interpolation is performed.

In response to a command from the arithmetic processing unit 203, the pixel color processing unit 209 causes the pixel brightness and the threshold value 0.7.
5 and compares the Z value of the pixel obtained by interpolation with the Z value of the pixel stored in the Z buffer 211. If the brightness of the pixel is 0.75 or more and the Z value of the pixel obtained by interpolation is the Z buffer 211
If it is smaller than the Z value of the pixel stored in, the pixel color processing unit 209 writes the drawing color corresponding to the threshold value 0.75 as the color of the pixel in the frame buffer 213.

When this polygon is drawn for the first time and the lightness is 0.75 or more, the drawing color is written. FIG. 19
In (a), two examples of a triangular polygon 1011 and a triangular polygon 1012 are shown. Triangular polygon 10
The lightness of each of the vertices P11, P12, and P13 of 11 is set to 0.0, 1.0, and 1.0. The luminosity of the vertices P21, P22, and P23 of the triangular polygon 1012 are set to 0.0, 0.5, and 1.0, respectively. When the processing described above is executed, the painted portion in each triangular polygon is colored with the drawing color.

Next, the threshold value 0.5 is selected. Then, in accordance with a command from the arithmetic processing unit 203, the triangle drawing processing unit 2
05 calculates the brightness and coordinates (including Z value) of each pixel inside the polygon. In response to a command from the arithmetic processing unit 203, the pixel color processing unit 209 compares the brightness of the pixel with a threshold value of 0.5 and calculates the Z value of the pixel and the Z value of the pixel stored in the Z buffer 211. Compare the Z values. If the brightness of a pixel is greater than or equal to a threshold value of 0.5, and the calculated Z value of the pixel is the Z buffer 21.
If it is smaller than the Z value of the pixel stored in 1, the pixel color processing unit 209 writes the drawing color corresponding to the threshold value 0.5 as the color of the pixel in the frame buffer 213.

If the Z buffer 211 is not used,
As shown in FIG. 19B, the area of lightness of 1.0 to 0.5 is colored with the drawing color corresponding to the threshold value of 0.5. Z buffer 2 is used for areas with a brightness of 0.75 or higher.
Since the Z value stored in 11 and the Z value obtained by calculation are the same, the drawing color corresponding to the threshold value 0.5 is written in the frame buffer 213 in the area of lightness 0.75 or more. I can't. That is, as shown in FIG. 19C, a region having a brightness of 0.5 to 0.74 and a brightness of 0.
Different drawing colors are applied to 75 or more regions.

FIG. 20 shows the result of the same processing performed for the threshold value 0.0 in the example of FIG. The vertices P11, P12, P1 of the triangular polygon 1011 in FIG.
The respective lightnesses of 3 are set to 0.0, 1.0 and 1.0. Also, the vertex P21 of the triangular polygon 1011,
The brightness of P22 and P23 is 0.0, 0.5,
It is set to 1.0. The numbers surrounded by dotted lines, that is, 0.5 and 0.75 indicate the threshold value of brightness. In this way, each triangular polygon is divided into three areas and the drawing color is colored.

FIG. 21 shows the result of Gouraud shading. The vertex P of the triangular polygon 1021 in FIG.
The lightness of each of 31, 31, P32, and P33 is 0.0, 1.
It is set to 0 and 1.0. Also triangular polygon 10
The lightness of each of the vertices P41, P42, and P43 of 22 is set to 0.0, 0.5, and 1.0. That is, the brightness of each vertex is the triangular polygon 1 shown in FIG.
The same as 011 and 1012. However, in the Gouraud shading, the brightness changes smoothly by interpolation, whereas in FIG. 20, the brightness changes only at the boundaries of the divided areas. That is, it can be seen that there are three regions having a flat lightness, which is in a cel animation style.

In some cases, the smallest lower limit value in the brightness range table is not 0.0. In order to eliminate the part of the polygon that is not colored, step S in FIG.
In the repetition of 617, the lower limit value is set to 0.0 in the final repetition, and step S615 is executed.

The above steps S605 to S617 are repeated until all the vertices of the stereo model are processed, and as a result, all polygons are processed (step S619).

The lightness range table described above should be created optimally for each stereo model. However, the three-dimensional model may be divided into some categories, and the brightness range table may be prepared for each category. It is conceivable that the number of lightness ranges included in the lightness range table is set to 2 or 3 according to the actual cell animation. However, in the processing described above, the number of repetitions in step S617 of FIG. 15 only increases, so that any number of 2 or more can be easily performed. However, since the number of repetitions increases, the processing becomes slower as the number increases.

By carrying out the above-mentioned processing, all polygons of the stereo model are painted with the brightness of a predetermined level, and a cel animation-like image can be obtained for the stereo model. Further, in the contour drawing model introduced in the first embodiment, a part of the surface behind the stereo model that is not hidden by the stereo model is drawn, so that part is rendered as a contour line. In the first embodiment,
To draw the contour line, the contour line can be easily drawn by introducing a contour drawing model and performing the same processing as the normal rendering processing.

2. Second Embodiment Different from the first embodiment, the second embodiment does not calculate the drawing color for the stereo model in real time but calculates it in advance and stores it as data. By doing so, the processing speed becomes faster than in the first embodiment. In the first embodiment, since the color of the material of the polygon and the reference lightness of the lightness range table are used for calculation, it suffices if each polygon has data for one color.
In the second embodiment, it is necessary to hold color data for each polygon for the number of rows of the brightness range table. The processing of the contour drawing model is the same as in the first embodiment.

The outline of the second embodiment of the present invention will be described with reference to the functional block diagram of FIG. The rendering apparatus illustrated as the second embodiment includes a contour drawing model drawing unit 450 and a stereo model drawing unit 490. The contour drawing model drawing unit 450 includes a contour drawing model acquisition unit 300, a contour drawing model placement matrix setting unit 305, a contour drawing model processing unit 310, a blur expression texture mapping unit 315, and a stereoscopic model. A pixel processing unit 430 shared with the model drawing unit 490 is included. Each of these functions transfers data in the order described above. The contour drawing model acquisition unit 30
0 and the contour setting model placement matrix setting unit 30
5, the contour drawing model processing unit 310, and the blur expression texture mapping unit 315 are the same as those in the first embodiment.

The stereo model drawing unit 490 includes a vertex conversion / light source calculation unit 460, a brightness calculation unit 465, a brightness range table 475, a drawing color storage unit 470, a brightness range setting unit 480, and a contour. Model drawing unit 450 for drawing
And a pixel processing unit 430 that is shared with. The output of the vertex conversion and light source calculation unit 460 is input to the brightness calculation unit 465. The output of the brightness calculation unit 465 is input to the pixel processing unit 430. The brightness range table 475 is referred to by the brightness range setting unit 480. Drawing color storage unit 470
And the output of the brightness range setting unit 480 is the pixel processing unit 43.
Input to 0. The pixel processing unit 430 shared by the contour drawing model drawing unit 450 and the stereo model drawing unit 490 includes a brightness comparison unit 433 used in the stereo model drawing process.
And a hidden surface removal processing unit 437 used in both the contour drawing model drawing processing and the stereo model drawing processing.

The contour drawing model acquisition unit 300 is the same as that in the first embodiment and generates a contour drawing model corresponding to a stereo model composed of, for example, triangular polygons.
When the contour drawing model is generated in advance, the contour drawing model acquisition unit 300 reads the preliminarily generated contour drawing model composed of triangular polygons. It should be noted that the respective surfaces of the acquired contour drawing model are opposite to the corresponding surfaces of the stereo model. Also, the contour drawing model is larger than the stereo model,
It is defined by a predetermined color scheme for contour lines.

Although the contour drawing model must finally be relatively larger than the corresponding stereo model, the size of the contour drawing model at this stage may be the same as that of the stereo model. In this case, the contour drawing model is processed to be drawn relatively larger than the stereo model before the contour drawing model and the stereo model are drawn. Further, the color of the contour drawing model may inherit the color of the material of the corresponding stereo model as it is. In this case, the drawing color is specified separately.

The reference position of the contour drawing model is usually defined to be the same as or near the reference position of the corresponding stereo model.

Since the contour drawing model placement matrix setting unit 305, which is the same as in the first embodiment, places the contour drawing model reference position 530 in the virtual space at the same position as the stereo model reference position 520. Set the matrix for placement. That is, as shown in FIG. 4, the placement matrix of the contour drawing model 510 is set to the contour drawing model reference position 530 and the stereo model reference position 520.
The contour drawing model 510 is arranged at a position including the three-dimensional model 500 by setting the transformation to be performed in parallel with the coordinates of.

The contour drawing model processing unit 310 is the same as in the first embodiment, and performs vertex conversion (enlargement / reduction / rotation / translation / perspective conversion) for each vertex of the contour drawing model, and The front and back of each polygon of the contour drawing model is determined. The placement matrix described above is also used for this vertex transformation. Note that the light source calculation is not performed here. For example, in addition to enlarging / reducing / rotating / translating / perspective-transforming in accordance with a specified state in a virtual space that is a virtual three-dimensional space, the contour drawing model acquisition unit 300 may have a contour of the same size as the stereo model. When the drawing model is acquired, the contour drawing model processing unit 310 performs vertex conversion for enlarging the size of the contour drawing model. Even when enlarged here, the relationship between the stereo model and the contour drawing model is as shown in FIG.

Further, the front and back of a polygon can be determined by the camera 55.
This is performed in order to exclude the polygon whose direction of the front surface is the same as the direction of the line of sight 540 from 0 from the drawing target.

Faint expression texture mapping unit 315
Is the same as that of the first embodiment, and executes processing for mapping the blur expression texture on the contour drawing model so that the contour line drawn as a result becomes a faint line. Is. The texture for expressing blur is a texture having a pattern including a change in brightness or transparency. Since the outline does not necessarily have to be blurred, the blurred expression texture mapping unit 31
The process of 5 is selectively performed.

The vertex conversion and light source calculation unit 460 of the stereo model drawing unit 490 performs vertex conversion (enlargement / reduction / rotation / translation) for each vertex of the triangular polygons forming the surface of the stereo model arranged in the virtual three-dimensional space. -Perspective transformation) is performed (the area in which the triangular polygon is drawn is calculated), and the light source is calculated for each vertex of the transformed triangular polygon. Further, the vertex conversion of the stereo model drawing unit 490 and the light source calculation unit 460 also determine the front and back of each triangular polygon of the stereo model.

Also in the vertex conversion and light source calculation unit 460 of the three-dimensional model drawing unit 490, enlargement / reduction / rotation / translation / movement are performed in accordance with the state specified in the virtual three-dimensional space.
In addition to perspective conversion, the contour drawing model processing unit 31
When the contour drawing model after being processed with 0 is the same size as the stereo model, the size of the stereo model is reduced so that the stereo model becomes relatively smaller than the contour drawing model. Vertex transformation is performed.

The relationship between the stereo model 500 and the contour drawing model 510 is as shown in FIG. 4 even when the vertex conversion / light source calculation unit 460 performs the stereo model reduction processing. In addition, the surface front / back determination is the same as that of the contour drawing model processing unit 310, and excludes, from the drawing targets, the surface of the three-dimensional model whose direction, which is the same as the line-of-sight direction of the camera, is the front surface.

The lightness calculation unit 465 calculates the lightness from the color at each vertex of the triangular polygon calculated by the vertex conversion / light source calculation unit 460. Normal vertex conversion and light source calculation unit 4
Since 60 calculates the color in the RGB system, the lightness calculation unit 465 obtains the lightness Y by YIQ converting this RGB.
The brightness at each vertex of this triangular polygon is output to the pixel processing unit 430.

The lightness range table 475 is a table as shown in FIG. 5, for example, as in the first embodiment. In addition,
Here, the brightness takes a real value from 0 to 1. It is also possible to specify the range by the upper limit and the lower limit instead of the threshold value (see, for example, FIG. 18).

When using the lightness range table as shown in FIG. 5, the drawing color storage section 470 needs to store three drawing color data for each polygon. As shown in FIG. 23, for each polygon, a drawing color (r, g, b) corresponding to the first lightness range and a drawing color (r, g, b) corresponding to the second lightness range , The drawing color (r, g, b) corresponding to the third lightness range, the vertex data IDX of the vertex 0 forming the polygon, the vertex data IDX of the vertex 1 and the vertex data IDX of the vertex 2. But,
It is stored instead of the above-mentioned FIG. 16 (b). These data are prepared for the number of polygons of the three-dimensional model.

The pixel processing section 430 fetches a drawing color corresponding to the brightness range set by the brightness range setting section 480 from the drawing color storage section 470. Drawing color storage unit 4
Reference numeral 70 denotes a CD-R 131, which is stored as an element of the stereo model data 137.

The brightness range setting unit 480 selects one threshold value of the brightness range table 475 and sets it in the pixel processing unit 430. When the brightness range table 475 as shown in FIG. 5 is used as it is, the brightness range setting unit 480 sets the brightness range tables 475 one by one from the top. If upper and lower bounds are specified instead of thresholds,
It can be randomly selected and set.

The pixel processing unit 430 shared by the contour drawing model drawing unit 450 and the stereo model drawing unit 490 is
The color or brightness of each pixel in the triangular polygon is obtained by interpolating the color or brightness of each vertex of the triangular polygon. The interpolation method may be the Gouraud shading algorithm or the von shading algorithm.

The pixel processing unit 430, when processing a triangular polygon that is the drawing target of the contour drawing model, performs the hidden surface removal processing by using the hidden surface removal processing unit 437 while performing the contour drawing model. The color of each pixel in the triangular polygon that is the drawing target of is determined.

For example, in the case of FIG. 4, the three-dimensional model 500.
Two surfaces 501 and 502 that are closest to the camera 550 of the contour drawing model are drawn farther from the camera 550 of the contour drawing model.
Two faces 513, 514, 515 and 516 are drawn. Since these four surfaces of the contour drawing model 510 are projected to the left and right of the stereo model 500 when viewed from the camera 550, only the protruding portion is drawn without hidden surface removal. This protruding portion becomes the contour line. The pixel processing unit 430 determines the color in consideration of the color of the material of the contour drawing model. However, the color of the contour line (black or dark contour line color) may be set as the color of the contour drawing model by completely ignoring the color of the material.

On the other hand, when processing the triangular polygon which is the drawing target of the three-dimensional model, the pixel processing unit 430 first interpolates the brightness at each vertex of the triangular polygon output from the brightness calculation unit 465 to obtain the polygon. The brightness at each pixel inside is calculated.

Then, the brightness comparing section 433 compares the brightness of each pixel with the threshold value set by the brightness range setting section 480. If the brightness of the pixel is equal to or higher than the threshold value, the pixel processing unit 430 draws the pixel with a drawing color based on the reference brightness corresponding to the threshold value.
At the time of this drawing processing, hidden surface removal processing is also performed using the hidden surface removal processing unit 437. If the brightness at that pixel is less than the threshold, then this pixel is not drawn at this stage. The brightness range setting unit 480 sets the brightness range table 47.
All threshold values of 5 are set in the pixel processing unit 430,
Correspondingly, if the pixel processing unit 430 draws all the pixels in the triangular polygon, the inside of the triangular polygon is divided into three stages in the example of FIG. This process is executed for all triangular polygons of the three-dimensional model.

Next, the processing flow of the second embodiment will be described. The processing described below is performed by the arithmetic processing unit 103.
(FIG. 1) is a process executed by controlling other elements in the computer main body 101.

[CD-R Recording Processing] In the second embodiment, the contour drawing model generation processing performed in advance is shown in FIG.
Is the same as. When the processing is started, the data of the stereo model stored in the HDD 107 in advance is read (step S303) and is acquired as the conversion target model.

Next, the size of the model to be converted is enlarged so as to be slightly larger (step S30).
5). For example, in the normal direction of each vertex of the conversion target model,
The vertex is moved by a length of 2% of the total length of the conversion target model, and is expanded by about 2% as a whole. When the enlargement ratio is larger, the contour line is drawn thicker, and when the enlargement ratio is smaller, and the conversion target model is only slightly enlarged, the contour line is drawn thinner. Furthermore, if it is not uniform and a part is enlarged,
Only the outline of the enlarged portion is drawn thick. Since this size adjustment is usually performed by the creator of the three-dimensional model, it is possible to draw a contour line that reflects the creator's intention.

When the normal line of each vertex of the three-dimensional model is not defined, the normal line of the vertex which is obtained by interpolating the normal line of each surface sharing the vertex is used, Can also be moved in the direction of the normal to the vertex. Also, the surface can be moved in the normal direction of each surface of the three-dimensional model. However, when the surfaces are simply moved, a gap is created between the surfaces, and a separate process for filling the gap is required. Further, since the reference position is usually defined in the stereo model, each vertex of the conversion target model can be moved around the reference position of the corresponding conversion target model.

Next, the color of the material of each polygon of the model to be converted is set to a color having the same saturation but lower brightness (step S307). Note that all polygons may be set to a single color such as black. Further, the setting for mapping the texture for expressing blur may be performed. Since the color of the material is adjusted by the manufacturer, the contour line can be drawn in the color intended by the manufacturer.

Next, the front and back of each polygon of the conversion target model are reversed (step S309). Specifically, the order in which the vertices of each triangle forming the conversion target model are defined is changed at one place.

The data of the conversion target model converted so far is used as the contour drawing model data in the HDD 107.
(Step S311), and the contour drawing model generation process ends (step S313).

Next, various data including the stereo model data and the contour drawing model data stored in the HDD 107 is transferred to the CD-R 13 by the CD-R drive 113.
Written to 1. FIG. 7 schematically showing an example of data written in the CD-R 131 is the same in the second embodiment.

[Overall Processing Flow] The overall processing flow of the second embodiment is the same as that of the first embodiment as far as shown in FIG. First, when the process is started, initialization is performed (step S2). This initial setting processing includes data acquisition processing for the contour drawing model (FIG. 9), which will be described in detail later, and data acquisition processing for the stereo model to be drawn. Then, the state in the virtual space is set (step S3). This process is, for example, a process of changing the state in the virtual space in response to a change in the position of the viewpoint, a change in the position of the light source, a movement of the model, or a deformation of the model. By performing this processing, the position coordinates of the stereo model and the contour drawing model
The determination processing of the direction, the enlargement ratio, the reduction ratio, etc. is performed. More specifically, the process of determining the placement matrix (used in FIG. 11) for the stereo model and the contour drawing model is performed. Further, according to a key operation of the input device 161 (FIG. 1) or the like, the setting as to whether or not contour line drawing is performed in step S4 is performed in step S3.

Next, a judgment process as to whether or not to draw the contour line is performed (step S4). This is determined based on the setting by the key operation of the input device 161, or the setting by another program as described above. When it is determined that the contour line is drawn, the drawing process of the contour line drawing model is executed (step S5). This will be described later with reference to FIG. Then, the drawing process of the stereo model is executed regardless of whether the contour line is drawn or not (step S6). This processing will also be described later with reference to FIG. The steps S3 to S6 are repeatedly executed until the processing is completed (step S7). Whether or not the processing is ended is determined by whether or not an operation input indicating that the processing should be ended is performed.

[Contour Drawing Model Acquisition Processing] The contour drawing model acquisition processing shown in FIG. 9 is the same in the second embodiment. Here, first, it is determined whether or not a contour drawing model is generated (step S203). This is because there are cases where the contour drawing model is prepared in advance and cases where the contour drawing model is generated at this stage.
Here, this determination is performed by determining, for example, whether the contour drawing model corresponding to the stereo model is stored in the CD-R 131. If it is determined that the contour drawing model is stored, it is determined that the contour drawing model is not generated, and if it is determined that the contour drawing model is not stored, it is determined that the contour drawing model is generated.

If it is determined that the contour drawing model is not generated, the data of the contour drawing model stored in the CD-R 131 is read (step S20).
7). Each polygon of this contour drawing model is shown in FIG.
As described with reference to FIG. 6 and FIG. 6, the polygon corresponding to the three-dimensional model has its front and back reversed. Further, the size of the contour drawing model read out is defined to be slightly larger than that of the corresponding stereo model. Further, the color of the contour drawing model is defined as a color darker than the corresponding stereo model.

If it is determined that the contour drawing model is generated, the processing for generating the contour drawing model is performed (step S205). Similar to step S207, even when the contour drawing model is generated at this stage, each polygon of the contour drawing model is
As described above with reference to FIG. 4, the polygon corresponding to the stereo model is turned upside down.

The size of the contour drawing model is slightly larger than that of the corresponding stereo model. Step S30
As in the case of FIG. 5 (FIG. 6), for example, a contour drawing model is generated by moving the vertices of the stereo model in the normal direction of the vertices and enlarging the contour drawing model. When the contour drawing model is larger than the stereo model, the contour line is drawn thicker, and when the contour drawing model is slightly larger than the stereo model, the contour line is drawn thinner.

At this point, the size of the contour drawing model may be the same as the size of the corresponding stereo model. In this case, after the contour drawing model is acquired by the contour drawing model acquisition processing, the contour drawing model is arranged by the contour drawing model arrangement processing described later until the arrangement matrix of the contour drawing model is set. Is expanded. Alternatively, when the placement matrix of the contour drawing model is set in the contour drawing model placement process, the placement matrix may be determined so that the placement matrix includes the expansion conversion. Conversely, when arranging the three-dimensional model, the three-dimensional model arrangement matrix may be determined such that the three-dimensional model arrangement matrix includes reduction conversion.

On the other hand, the material color of each surface of the contour drawing model is generated by darkening the material color of each polygon of the corresponding stereo model. Note that, as described in step S307 (FIG. 6),
At this point, the color of the contour drawing model to be generated may not be defined. Alternatively, the material color of each polygon of the contour drawing model may be the same as the material color of each polygon of the corresponding stereo model. In this case, the color of the contour drawing model is not taken into consideration during the drawing process of the contour drawing model, and the contour drawing model is drawn with a separately defined color such as black or a texture color that expresses fading. To be done.

Next, it is judged whether or not the texture expressing the blur is mapped on the contour drawing model (step S209). When the contour drawing model is generated in step S205, the determination is performed based on the data of the corresponding stereo model. On the other hand, if the contour drawing model is read in step S207,
The determination is performed based on the read data of the contour drawing model. When it is determined that the texture expressing the blur is mapped, step S211 is performed.
At, the texture expressing the blur is mapped on the contour drawing model. That is, texture coordinates (U, V) are set at each vertex of the polygon.

As described above, the texture expressing the blur has a pattern including a change in brightness or transparency. FIG. 10 shows an example of a texture including a change in brightness. This is a texture with a pattern in which white diagonal lines are finely drawn on a black background. The brightness of the black part is low,
Since the brightness of the white part is high, the texture shown in FIG. 10 includes a change in brightness. By drawing such a line as a contour line, a contour line including a change in brightness is drawn. That is, blurring of the contour line is expressed, and a more handwritten contour line is drawn.

When it is determined that the texture expressing the blur is not mapped and when the texture mapping process is completed, the arithmetic processing unit 103 ends the contour drawing model acquisition process (step S21).
3).

[Contour Drawing Model Arrangement Processing] In step S3 of FIG. 8, the arrangement matrix of the contour drawing model is set, and the arrangement processing of the contour drawing model is performed.
The reference position of the normal contour drawing model is provided at a position corresponding to the reference position of the stereo model. Then, the arrangement matrix of the contour drawing model is set such that the reference position of the contour drawing model is arranged at the same position as or near the reference position of the stereo model.

Here, when the direction of the stereo model changes, the layout matrix including the rotation conversion is set so that the contour drawing model also corresponds to it. When the shape of the stereo model changes, the transformation process is performed so that the contour drawing model corresponds to it.

At this stage, if the contour drawing model has the same size as the corresponding stereo model, the contour drawing model is enlarged. Specifically, the arrangement matrix of the contour drawing model is set so that the vertices of the contour drawing model are enlarged and converted according to a predetermined enlargement ratio with the reference position of the contour drawing model as the center. Alternatively, conversely, the stereo model may be reduced. That is, in this case, the arranging matrix of the stereo model is set so that the vertices of the stereo model are reduced and converted with a predetermined reduction ratio with the reference position of the stereo model as the center.
When the contour drawing model is relatively larger than the corresponding stereo model, the stereo model arranging matrix can be used as it is as the contour drawing model arranging matrix.

In this way, the relatively large contour drawing model is finally arranged so as to include the stereo model. The contour drawing model may not completely include a stereo model depending on the arrangement position, direction, shape, and the like of both models. However, even in such a case, a contour line is drawn for the included portion.

At this stage, the arrangement matrix does not necessarily have to be set, and the arrangement coordinates,
It suffices that each element necessary for the vertex conversion such as the direction and the enlargement / reduction ratio is fixed. Also in this case, the actual vertex conversion is performed at the stage of drawing processing of each model.

[Contour Drawing Model Drawing Processing] The drawing processing flow of the contour drawing model shown in FIG. 11 is the same in the second embodiment. In FIG. 11, the process described below is repeatedly performed until all the vertices of the contour drawing model are processed (step S503). The first process that is repeatedly performed is a vertex conversion (enlargement / reduction / rotation / translation / perspective conversion) process for one vertex (step S505). Here, the arrangement matrix of the contour drawing model obtained in step S3 is also used in the vertex conversion.

For example, this processing is carried out by the geometric calculation section 207 instructed by the calculation processing section 103. It should be noted that the geometric calculation unit 20 is used for the contour drawing model.
7 is that light source calculation is not performed. This is because the contour line is drawn regardless of the position of the light source and it is useless to calculate the light source. For example, the material color of the contour drawing model may be ignored. Normally, this vertex transformation is performed based on the state specified in the virtual three-dimensional space. However, if the size of the contour drawing model is the same as the stereo model, according to the placement matrix set in the placement process. At this stage, the contour drawing model may be enlarged and converted.

Then, it is judged whether or not the polygon including the vertex is the front surface (step S507).
In the case of a triangular polygon, this determination is made based on which direction the triangular polygon composed of this apex and the two apexes processed before it is facing. If there is a normal vector in the direction of the front surface, the front and back of the triangular polygon can be determined by the sign of the inner product of the normal vector and the line-of-sight vector. That is, if the sign of the inner product is negative, it means that the front surface is facing the viewpoint position, and if the sign of the inner product is positive, it means that the back surface is facing the viewpoint position. Become. In the second embodiment, the front / backside determination is performed at this stage, but the front / backside determination may be performed before this stage.

If the polygon including the apex is a back surface, the process returns to step S503. If the polygon including the apex is the front surface, it is determined whether or not the texture expressing blur is mapped (step S509).

This means texture mapping for polygons. If the texture expressing the blur is mapped, the texture coordinate of the texture for expressing the blur is calculated for the vertex (step S511). When texture mapping is performed, the texture coordinates (U, V) have already been specified at the vertices of the polygon, but if the polygon is arranged diagonally to the screen, the texture will be distorted on the screen. May be displayed. In order to avoid this distortion, here, as the texture perspective processing, calculation of S = U × Q and T = V × Q is performed using Q = 1 / w (w is the depth from the screen). If the texture expressing the blur is not mapped, the process proceeds to step S513.

Then, for example, the triangle drawing processing unit 205 and the pixel color processing unit 209 shown in FIG. 2 are driven (step S513). As described above, the triangle drawing processing unit 205 interpolates the data of each vertex of the triangular polygon to generate the data of each pixel inside the triangular polygon. The data of each vertex is the color of the material, the screen coordinate value, and the texture coordinate value if step S511 is performed. The data in each pixel is the color of the material and the texel color if step S511 is performed.

However, it is possible to ignore the material color at this point and set the color of the contour line at each vertex.
It is also possible to set the brightness in consideration of the color of the material. The pixel color processing unit 209 uses the data in each pixel inside the triangular polygon generated by the triangular drawing processing unit 205 to use the frame buffer 213.
Write the display image to. At this time, the hidden surface is erased by using the Z buffer 211.

[Stereoscopic Model Drawing Process] FIG. 24 shows a flow of the stereoscopic model drawing process in the second embodiment. Figure 1
5 is different from the processing content of step S603 of FIG. 15 in the initial setting processing content of step S633 of FIG. 24 and that step S611 of FIG. 15 is replaced with step S641 of FIG. That is, the drawing color is calculated each time in FIG. 15, but since it is calculated and stored in advance in the second embodiment, the processing for loading the drawing model color data for the stereo model to be drawn in the memory 105 in step S633 in advance. Is required. Further, it is necessary to perform a process of reading the drawing color of the polygon in step S641. In addition, step S641 to step S6
35 and S639, after step S643, or in parallel with those steps. Since it is calculated and stored in advance, it may be read before it is actually used.

In FIG. 24, first, initial setting is performed (step S633). In this initial setting, the brightness range table (for example, FIG. 5 or FIG. 18) corresponding to the stereo model is acquired. Also, the color data for drawing the stereo model is acquired. Next, vertex conversion (enlargement / reduction / rotation / translation / perspective conversion) and light source calculation for one vertex are performed (step S635). Here, the matrix for placement of the three-dimensional model is also used in the vertex conversion. This is executed by the geometric calculation unit 207 according to a command from the calculation processing unit 103, for example. The data of the stereo model is stored in the CD-R 131, for example.

The enlargement / reduction / rotation / translation / perspective transformation is basically based on the state in the virtual space set in step S3 of FIG. However, if the contour drawing model has the same size as the stereo model, the contour drawing model may be made relatively large by reducing the size of the stereo model. In this case, reduction conversion is performed in step S635.
It should be noted that the size can be easily reduced by moving each vertex along the normal line toward the center of the three-dimensional model.

The two methods of light source calculation in the stereo model drawing processing of the present invention described in the first embodiment can be applied to the second embodiment as they are.

Next, it is judged whether or not the polygon including the vertex is the front surface (step S637). In the case of a triangular polygon, this judgment is performed depending on which direction the triangular polygon composed of this apex and the two apexes processed before is directed. For this determination, the method described in the drawing process of the contour drawing model can be used. In the second embodiment, the front / backside determination is performed at this stage, but the front / backside determination may be performed before this stage.

If the polygon including the apex is the back surface, the process returns to step S635. If the polygon including the apex is the front surface, the brightness at the apex subjected to the apex conversion and the light source calculation is calculated (step S639). YIQ conversion is performed in the brightness calculation.

Then, the drawing color of the polygon including the vertices subjected to the vertex conversion and the light source calculation is read from the memory 105, for example (step S641). The drawing color data to be read is calculated in advance, but the calculation method for this calculation may be either of the two methods described in the first embodiment, or another method. Good. Further, the drawing colors may be defined one by one. In the second embodiment, the drawing color is prepared in advance, so the execution speed is high, but it is not easy to change to a color other than the prepared drawing color. On the other hand, when the reference brightness defined in the brightness range table is used for calculation at the time of execution as in the first embodiment,
The drawing color can be changed as appropriate by simply changing the lightness range table or changing the reference lightness.

Next, one brightness range in the brightness range table is selected (step S643). In this embodiment, the lightness range table shown in FIG. 5 is used, but FIG.
It is also possible to use a brightness range table such as When such a brightness range table is used, a brightness range including an upper limit and a lower limit can be randomly selected and set. However, FIG. 18 shows a case where the computer is effective up to the second decimal place. In the lightness comparison process described below, if the lightness of each pixel cannot be easily compared with the two lightness values of the upper limit and the lower limit, the lightness range is set to, for example, FIG.
8 are selected in order from the top. In this case, only the lower limit value is processed.

After that, the lightness at the vertex of this polygon is interpolated, and the lightness at each pixel inside the polygon (lightness distribution in the polygon) is calculated. The colors of the vertices are also interpolated, but the results are the same even if they are interpolated because all three vertices have the same drawing color. Then, if the brightness of the pixel is within the selected brightness range, the pixel is drawn with the drawing color corresponding to the selected brightness range (step S645). The brightness interpolation processing is executed by the triangle drawing processing unit 205 in FIG. 2, for example. The comparison process of whether the brightness of each pixel is within the selected brightness range is
For example, the pixel color processing unit 209 implements it. The steps S643 and S645 are repeated until processing is performed for all the brightness ranges (step S647).

For example, when the pixel color processing unit 209 cannot handle the two lightness values of the upper limit and the lower limit, the Z buffer 211 can be used together to obtain the same effect. The Z buffer 211 is used for hidden surface removal, but in the present embodiment, the combined use of the Z buffer 211 has the same effect as when compared with the upper limit value of brightness.

In some cases, the smallest lower limit value of the brightness range table is not 0.0. In order to eliminate the portion of the polygon that is not colored, step S in FIG.
In the iteration of 647, the lower limit value is set to 0.0 in the last iteration, and step S645 is executed.

The above steps S635 to S647 are repeated until all vertices of the three-dimensional model are processed, and as a result, all polygons are processed (step S649).

By performing the above-described processing, all polygons of the stereo model are painted with the brightness of a predetermined level, and a cel animation-like image can be obtained for the stereo model. In particular, the second embodiment is faster than the first embodiment. Further, in the contour drawing model introduced in the second embodiment, a portion of the surface behind the stereo model that is not hidden by the stereo model is drawn, and thus that portion is rendered as a contour line. According to the second embodiment, the contour line can be easily drawn by introducing the contour drawing model and performing substantially the same processing as the normal rendering processing.

3. Third Embodiment In the third embodiment, unlike the first and second embodiments, the front and back of each polygon of the contour drawing model is not inverted with respect to the corresponding polygon of the stereo model. However, in the case of the contour drawing model, the reference of the process for determining the surface to be drawn is reversed to produce the same effect as in the first and second embodiments. The processing of the stereo model is the same as that of the first embodiment.

The outline of the third embodiment of the present invention will be described with reference to the functional block diagram of FIG. The rendering apparatus illustrated as the third embodiment includes a contour drawing model drawing unit 750 and a stereo model drawing unit 790. The contour drawing model drawing unit 750 includes a contour drawing model acquisition unit 700, a contour drawing model placement matrix setting unit 705, a contour drawing model processing unit 710, a blur expression texture mapping unit 715, and a stereoscopic model. A pixel processing unit 730 shared with the model drawing unit 790 is included. Each of these functions transfers data in the order described above.

The stereo model drawing unit 790 includes a vertex conversion / light source calculation unit 360, a brightness calculation unit 365, a brightness range table 375, a drawing color calculation unit 370, a brightness range setting unit 380, and a contour. Drawing model drawing unit 750
And a pixel processing unit 730 that is shared with. The vertex conversion / light source calculation unit 360, the brightness calculation unit 365, the brightness range table 375, the drawing color calculation unit 370, and the brightness range setting unit 380 are the same as those in the first embodiment. The output of the vertex conversion and light source calculation unit 360 is the brightness calculation unit 365.
Is input to. The output of the brightness calculation unit 365 is input to the pixel processing unit 330. The brightness range table 375 is referenced by both the drawing color calculation unit 370 and the brightness range setting unit 380. The outputs of the drawing color calculation unit 370 and the brightness range setting unit 380 are input to the pixel processing unit 730. Contour drawing model drawing unit 750 and stereo model drawing unit 790
The pixel processing unit 730 shared by the two includes a brightness comparison unit 733 used in the stereo model drawing process and a hidden surface removal processing unit 737 used in both the contour drawing model drawing process and the stereo model drawing process.

The contour drawing model acquisition section 700 generates a contour drawing model corresponding to a stereo model composed of, for example, triangular polygons. If the contour drawing model is generated in advance, the contour drawing model acquisition unit 7
00 reads out the preliminarily generated contour drawing model composed of triangular polygons. The third embodiment differs from the first and second embodiments in that each polygon of the acquired contour drawing model has the same front and back as the corresponding polygon of the stereo model. The contour drawing model is larger than the stereo model and is defined by a predetermined color scheme for contour lines.
Although the contour drawing model must be finally larger than the corresponding stereo model, the size of the contour drawing object at this stage may be the same as that of the stereo model. In this case, by the time the contour drawing model and the stereo model are drawn, the contour drawing model is processed to be drawn relatively larger than the stereo model.

Further, the color of the contour drawing model may inherit the color of the material of the corresponding stereo model as it is. In this case, the drawing color is specified separately. The reference position of the contour drawing model is usually defined to be the same as or near the reference position of the corresponding stereo model. For example, FIG. 26 shows a case where the contour drawing model 610 is defined to be slightly larger than the stereo model 600. In FIG. 26, the direction of the arrow on each surface indicates the front surface. Both the three-dimensional model 600 and the contour drawing model 610 have a front surface outside each hexagonal surface.

The stereo model reference position 620, which is the reference position of the stereo model 600, and the contour drawing model reference position 630, which is the reference position of the contour drawing model 610, are both defined at the center of each model. Also, the contour drawing model 610
Is defined to be slightly larger than the three-dimensional model 600 around the contour drawing model reference position 630.

Then, the contour drawing model placement matrix setting unit 705 (FIG. 25) creates a placement matrix for placing the contour drawing model reference position 630 in the virtual space at the same position as the stereo model reference position 620. Set. This arranging matrix is used for conversion such as translation, rotation, enlargement / reduction, etc. for each vertex of the corresponding model. That is, the arrangement matrix of the contour drawing model 610 is set so as to include a conversion for translating the contour drawing model reference position 530 to the coordinates of the stereo model reference position 520, so that the contour drawing is performed at a position including the stereo model 600. Model 610 is arranged.

The contour drawing model processing unit 710 carries out vertex conversion (enlargement / reduction / rotation / translation / perspective conversion) for each vertex of the contour drawing model, and the front and back of each polygon of the contour drawing model. Make a decision. The above-mentioned arrangement matrix is used for this vertex conversion. However, unlike the first and second embodiments, this reverse side front / back determination unit 713 performs this front / back determination. Further, the light source calculation is not performed here. For example, in addition to enlarging / reducing / rotating / translating / perspective converting in accordance with a specified state in a virtual space that is a virtual three-dimensional space, the contour drawing model acquisition unit 700 draws a contour of the same size as the stereo model. When the contour drawing model is acquired, the contour drawing model processing unit 710 performs vertex conversion for enlarging the contour drawing model. Even when enlarged here, the relationship between the stereo model and the contour drawing model is as shown in FIG.

Further, in the case of the contour drawing model of the third embodiment, the front surface is determined to be the back surface, and the back surface is determined to be the front surface. Therefore, in the example of FIG. 26, only the surfaces 613, 614, 615 and 616 in which the arrows point in the same direction as the line of sight 640 from the camera 650 are the drawing targets. Normally, this surface is not a drawing target because it is a back surface, but it is treated as a drawing target in the third embodiment. In this way, the surfaces 611 and 612, which are outside the stereo model 600 and close to the camera 650, are excluded from the drawing target, so that the stereo model 600 is drawn as usual. Since the hidden surface removal processing unit 435 of the pixel processing unit 430 performs hidden surface removal, not all of these surfaces are drawn even if they are drawing targets.

Faint expression texture mapping unit 715
Performs a process for mapping the blur expression texture on the contour drawing model so that the resulting contour line becomes faint. Since the contour line does not necessarily have to be blurred, the processing of the blurred expression texture mapping unit 715 is selectively performed.

The vertex conversion and light source calculation unit 360 of the stereo model drawing unit 790 performs the vertex conversion (enlargement / reduction / rotation / translation / perspective conversion) for each vertex of the triangular polygon of the stereo model arranged in the virtual three-dimensional space. Is performed (the area in which the triangular polygon is drawn is calculated), and the light source is calculated for each vertex of the triangular polygon whose vertex has been converted. In addition, the vertex conversion and light source calculation unit 360 of the stereo model drawing unit 790 also determines the front and back of each triangular polygon of the stereo model.

Also in the vertex conversion and light source calculation unit 360 of the stereo model drawing unit 790, enlargement / reduction / rotation / translation / movement are performed according to the specified state in the virtual three-dimensional space.
In addition to perspective transformation, the contour drawing model processing unit 71
When the contour drawing model after being processed with 0 is the same size as the stereo model, the size of the stereo model is reduced so that the stereo model becomes relatively smaller than the contour drawing model. Vertex transformation is performed.

FIG. 26 shows the relationship between the stereo model 600 and the contour drawing model 610 even when the vertex conversion / light source calculation unit 360 performs the stereo model reduction processing. Also,
Different from the contour drawing model processing unit 710, the surface front / back determination excludes, from the drawing target, the surface of the three-dimensional model whose front direction is the same direction as the line-of-sight direction of the camera.
In the example of FIG. 26, the rear surface 6 when viewed from the camera 650.
03, 604, 605 and 606 are excluded from the drawing targets.

The brightness calculation unit 365 calculates the brightness from the color at each vertex of the triangular polygon calculated by the vertex conversion / light source calculation unit 360. Normal vertex conversion and light source calculation unit 3
Since 60 calculates the color in the RGB system, the lightness calculation unit 365 obtains the lightness Y by YIQ converting this RGB.
The brightness at each vertex of this triangular polygon is output to the pixel processing unit 730.

The brightness range table 375 is the same as in the first embodiment.
As with No. 2 and No. 2, for example, a table as shown in FIG. Note that here, the brightness takes a real value from 0 to 1. It is also possible to specify the range by the upper limit and the lower limit instead of the threshold value (see, for example, FIG. 18). With reference to the brightness range table 375, the drawing color calculation unit 370 calculates the drawing color corresponding to each threshold value. The drawing color corresponding to each threshold value is calculated using the reference lightness corresponding to the threshold value and the color information preset for each triangular polygon of the stereo model. The drawing color calculation unit 370 outputs the calculated drawing color to the pixel processing unit 730.

The brightness range setting section 380 selects one threshold value of the brightness range table 375 and sets it in the pixel processing section 730. When using the brightness range table 375 as shown in FIG. 5 as it is, the brightness range setting unit 380 sets the brightness range table 375 one by one from the top. If upper and lower bounds are specified instead of thresholds,
It can be randomly selected and set.

The pixel processing unit 730 shared by the contour drawing model drawing unit 750 and the stereo model drawing unit 790 is
The color or brightness of each pixel in the triangular polygon is obtained by interpolating the color or brightness of each vertex of the triangular polygon. The interpolation method may be the Gouraud shading algorithm or the von shading algorithm.

The pixel processing unit 730, when processing a triangular polygon that is the drawing target of the contour drawing model, performs the hidden surface removal processing by using the hidden surface removal processing unit 737 while simultaneously performing the contour drawing model. The color of each pixel in the triangular polygon that is the drawing target of is determined.

For example, in the case of FIG. 26, the three-dimensional model 60
The two faces 601 and 602 closest to the zero camera 650
Is drawn and is far from the camera 650 of the contour drawing model.
Two surfaces 613, 614, 615 and 616 are drawn. Since these four surfaces of the contour drawing model 610 are projected to the left and right of the stereo model 600 when viewed from the camera 650, only the protruding portion is drawn without hidden surface removal. This protruding portion becomes the contour line. The pixel processing unit 730 determines the color in consideration of the color of the material of the contour drawing model. In some cases, the color of the material is completely ignored and the color of the contour line (color for black or dark contour line) is used as the color of the contour drawing model.

On the other hand, when processing the triangular polygon that is the drawing target of the three-dimensional model, the pixel processing unit 730 first interpolates the lightness at each vertex of the triangular polygon output from the lightness calculation unit 365 to obtain the polygon. The brightness at each pixel inside is calculated.

Then, the brightness comparing section 733 compares the brightness of each pixel with the threshold value set by the brightness range setting section 380. If the brightness of the pixel is greater than or equal to the threshold value, the pixel processing unit 730 draws the pixel with the drawing color based on the reference brightness corresponding to the threshold value.
At the time of this drawing processing, hidden surface removal processing is also performed using the hidden surface removal processing unit 737. If the brightness at that pixel is less than the threshold, then this pixel is not drawn at this stage. The brightness range setting unit 380 sets the brightness range table 37.
All threshold values of 5 are set in the pixel processing unit 730,
Correspondingly, if the pixel processing unit 730 draws all the pixels in the triangular polygon, the inside of the triangular polygon is divided into three stages in the example of FIG. This process is executed for all triangular polygons of the three-dimensional model.

Next, the processing flow of the third embodiment will be described. The processing described below is performed by the arithmetic processing unit 103.
(FIG. 1) is a process executed by controlling other elements in the computer main body 101.

[CD-R Recording Process] FIG. 27 shows a contour drawing model generation process which is performed in advance in the third embodiment. When the processing is started, the data of the stereo model stored in the HDD 107 in advance is read (step S353) and is acquired as the conversion target model.

Then, the size of the model to be converted is enlarged so as to be slightly larger (step S35).
5). For example, in the normal direction of each vertex of the conversion target model,
The vertex is moved by a length of 2% of the total length of the conversion target model, and is expanded by about 2% as a whole. That is, for example, if the conversion target model is a human type and the height thereof is equivalent to 1.8 m, each vertex is 0.0
It is moved by a length equivalent to 36 m. When the enlargement ratio is larger, the contour line is drawn thicker, and when the enlargement ratio is smaller, and the conversion target model is only slightly enlarged, the contour line is drawn thinner. Furthermore, if it is not uniform and a part is enlarged, only the outline of the enlarged portion is drawn thick. Since this adjustment is usually performed by the maker of the three-dimensional model, it is possible to draw a contour line that reflects the maker's intention.

When the normal line of each vertex of the three-dimensional model is not defined, the normal line of the vertex obtained by interpolating the normal line of each surface sharing the vertex is used to calculate the vertex line. Can also be moved in the direction of the normal to the vertex.

It is also possible to move the surface in the normal direction of each surface of the stereo model. However, when the surfaces are simply moved, a gap is created between the surfaces, and a separate process for filling the gap is required. Furthermore, since the reference position is usually defined in the stereo model, centering around the reference position of the corresponding conversion target model,
It is also possible to move each vertex of the conversion target model.

Next, the color of the material of each polygon of the model to be converted is set to the color with the same saturation but lower brightness (step S357). Note that each polygon may be set to a single color such as black. Further, the setting for mapping the texture for expressing blur may be performed. Since the color of the material is adjusted by the manufacturer, the contour line can be drawn in the color intended by the manufacturer.

In the third embodiment, since the process of reversing the front and back of each polygon of the conversion target model is not performed, the data of the conversion target model processed up to this point is stored in the HDD 107 as contour drawing model data. (Step S
361), and the contour drawing model generation processing ends (step S363).

Next, various data including the contour drawing model data stored in the HDD 107 is written in the CD-R 131 by the CD-R drive 113. Figure 7
At the level described in 1), the example of the data written in the CD-R 131 is the same as in the first and second embodiments.
That is, in the program area 132, the computer 1
000 stores a program for implementing the present invention. This program may not include the processing up to writing to the CD-R 131. The system data area 133 stores various data processed by the programs stored in the program area 132 described above.

In the image data area 134, a three-dimensional model
Data including the data 137 and the contour drawing model data 135 is stored. Here, the front and back of each polygon of the model indicated by the contour drawing model data is the same as the corresponding polygon of the stereo model. Further, in the contour drawing model acquisition process described later, when the contour drawing model is generated, it is not necessary to store the contour drawing model data 135. In the sound data area 136, the sound processing unit 109 shown in FIG.
Data for outputting sound from 25 is stored.

The size of the contour drawing model stored in the CD-R 131 may be defined as the same size as the size of the corresponding stereo model. In this case,
After the contour drawing model is acquired by the contour drawing model acquisition processing described below, the contour drawing model is expanded until the arrangement matrix of the contour drawing model is set by the contour drawing model arrangement processing described below. To be done. Alternatively, when the placement matrix of the contour drawing model is set in the contour drawing model placement processing, the placement matrix may be determined so that the placement matrix includes expansion conversion. Conversely, when arranging the three-dimensional model, the three-dimensional model arrangement matrix may be determined such that the three-dimensional model arrangement matrix includes reduction conversion.

The material color of each polygon of the contour drawing model stored in the CD-R 131 may be the same as the material color of each polygon of the corresponding stereo model. In this case, the contour drawing model is drawn in a separately defined color such as black during the drawing process of the contour drawing model described later.

[Overall Processing Flow] The processing flow at the level shown in FIG. 8 is the same as in the first and second embodiments. That is, initial setting is first performed (step S
2). This initial setting includes a contour drawing model data acquisition process (FIG. 28), which will be described in detail later. Then, the state in the virtual space is set (step S3). At this time, a process for determining the position coordinates of the contour drawing model is executed. Next, a process of determining whether or not to draw the contour line is performed (step S4). If the contour line is drawn, the drawing process of the contour line drawing model is executed (step S
5). This will be described later with reference to FIG. The stereo model is drawn regardless of whether the contour line is drawn or not (step S6). These steps S3 to S6 are repeated until the processing is completed (step S7).

[Contour Drawing Model Acquisition Processing] FIG. 28 shows the contour drawing model acquisition processing. First, it is determined whether or not the contour drawing model is generated (step S223). This is because there are cases where the contour drawing model is prepared in advance and cases where the contour drawing model is generated at this stage. Here, this determination is made by, for example, the CD-R for the contour drawing model corresponding to the stereo model.
It is implemented by determining whether or not it is stored in 131. If it is determined that the contour drawing model is stored, it is determined that the contour drawing model is not generated, and if it is determined that the contour drawing model is not stored, it is determined that the contour drawing model is generated.

If it is determined that the contour drawing model is not generated, the data of the contour drawing model stored in the CD-R 131 is read (step S22).
7). Each polygon of this contour drawing model is shown in FIG.
6 and FIG. 27, the first embodiment is described.
Unlike 2 and 2, the front and back sides are the same as the corresponding polygons of the stereo model. The size of the contour drawing model read out is defined to be slightly larger than that of the corresponding stereo model. Further, the color of the contour drawing model is defined as a color darker than the corresponding stereo model.

If it is determined that the contour drawing model is generated, the processing for generating the contour drawing model is performed (step S225). Similar to step S227, even when the contour drawing model is generated at this stage, each polygon of the contour drawing model has the same front and back as the corresponding polygon of the stereo model (see FIG. 26).

The size of the contour drawing model is slightly larger than that of the corresponding stereo model. Step S35
5 (FIG. 27), for example, a contour drawing model that is enlarged by moving the vertices in the normal direction of each vertex of the stereo model is generated. When the contour drawing model is larger than the stereo model, the contour line is drawn thicker,
If the contour drawing model is only slightly larger than the stereo model, the contour line is drawn thinner.

Further, as described in the description of step S355 (FIG. 27), even if the surface is moved in the normal direction of each surface of the stereo model, the enlarged contour drawing model is generated. Good. Further, the contour drawing model may be generated by moving each vertex of the stereo model around the reference position defined in the normal stereo model.

At this point of time, the size of the contour drawing model may be generated in the same size as the size of the corresponding stereo model. In this case, after the contour drawing model is acquired by the main contour drawing model acquisition processing, the contour drawing is performed until the arrangement matrix of the contour drawing model is set by the contour drawing model arrangement processing described later. The model will be expanded. Alternatively, when the placement matrix of the contour drawing model is set in the contour drawing model placement process, the placement matrix may be determined to include the enlargement conversion. Conversely, when arranging the three-dimensional model, the three-dimensional model arrangement matrix may be determined such that the three-dimensional model arrangement matrix includes reduction conversion.

On the other hand, the material color of each polygon of the contour drawing model is generated by a darker color of the material of each polygon of the corresponding stereo model. As described in step S357 (FIG. 27), the color of the contour drawing model to be generated may not be defined at this point. Alternatively, the material color of each polygon of the contour drawing model may be the same as the material color of each polygon of the corresponding stereo model. In this case, when drawing the contour drawing model,
The color of the contour drawing model is not taken into consideration, and the contour drawing model is drawn with a separately defined color such as black or the color of the texture expressing the blur.

Next, it is determined whether or not a texture expressing blur is mapped on the contour drawing model (step S229). When the contour drawing model is generated in step S225, this determination is performed based on the data of the corresponding stereo model. On the other hand, if the contour drawing model is read in step S227,
This judgment is performed based on the read data of the contour drawing model. If it is determined that the texture expressing the blur is mapped, step S231.
At, the texture expressing the blur is mapped on the contour drawing model. That is, texture coordinates (U, V) are set at each vertex of the polygon.

As described above, the texture expressing the blur is a texture having a pattern including a change in brightness or transparency, and is the texture shown in FIG. 10, for example. If it is determined that the texture expressing the blur is not mapped, and if the processing for mapping the texture has ended, the arithmetic processing unit 103 ends the contour drawing model acquisition processing (step S233).

[Contour Drawing Model Arrangement Processing] In step S3 of FIG. 8, the arrangement matrix of the contour drawing model is set, and the arrangement processing of the contour drawing model is performed.
The reference position of the normal contour drawing model is provided at a position corresponding to the reference position of the stereo model. Then, the reference position of the contour drawing model is arranged to be the same as or near the position where the reference position of the stereo model is arranged,
A layout matrix for the contour drawing model is set.

Here, when the direction of the stereo model changes, the layout matrix including the rotation conversion is set so that the contour drawing model also corresponds thereto. When the shape of the three-dimensional model changes, the deformation process is performed so that the contour drawing model corresponds to it.

At this stage, if the contour drawing model has the same size as the corresponding stereo model, the contour drawing model is enlarged. Specifically, the arrangement matrix of the contour drawing model is set so that the vertices of the contour drawing model are enlarged and converted according to a predetermined enlargement ratio with the reference position of the contour drawing model as the center. Alternatively, conversely, the stereo model may be reduced. That is, in this case, the arranging matrix of the stereo model is set so that the vertices of the stereo model are reduced and converted with a predetermined reduction ratio with the reference position of the stereo model as the center.
When the contour drawing model is relatively larger than the corresponding stereo model, the stereo model arranging matrix can be used as it is as the contour drawing model arranging matrix.

In this way, the relatively large contour drawing model is finally arranged so as to include the stereo model. The contour drawing model may not completely include a stereo model depending on the arrangement position, direction, shape, and the like of both models. However, even in such a case, the contour line is drawn for the included portion.

At this stage, the arrangement matrix does not necessarily have to be set, and the arrangement coordinates,
It suffices that each element necessary for the vertex conversion such as the direction and the enlargement / reduction ratio is fixed. Also in this case, the actual vertex conversion is performed at the stage of drawing processing of each model.

[Drawing Process of Contour Drawing Model] In FIG. 29 showing the drawing process flow of the contour drawing model, the process described below is repeated until all the vertices of the contour drawing model are processed (step S523). ). The first process that is repeatedly performed is vertex conversion (enlargement / reduction / rotation / translation / perspective conversion) for one vertex (step S525). For example, this processing is executed by the geometric calculation unit 207 instructed by the calculation processing unit 103.

Here, it should be noted that the light source calculation is not performed on the contour drawing model. This is because the contour line is irrelevant to the position of the light source and the light source calculation is useless (in some cases, the color of the material of the contour drawing model may be eventually ignored). Normally, this vertex transformation is performed based on the state specified in the virtual three-dimensional space. However, if the size of the contour drawing model is the same as the stereo model, according to the placement matrix set in the placement process. The contour drawing model may be enlarged and converted at this stage.

Then, it is judged whether or not the polygon including the apex is the back surface according to the normal judgment standard (step S527). Normally, only the front surface is the drawing target, but in the case of the contour drawing model of the second embodiment, the back surface is the drawing target according to the normal determination standard. In the case of a triangular polygon, the determination in this step is made depending on which direction the triangular polygon constituted by the two vertices processed before this vertex is oriented.

As shown in FIG. 12, for example, when each vertex of a triangular polygon is assigned a vertex number in a counterclockwise direction according to a normal criterion, it is defined that the front side of the page is the front side (so-called). Right hand system). In the third embodiment, the front and back determination criteria are reversed, and when the vertex numbers are given in the clockwise direction, it is determined that the front side of the page is the front side. Only the polygons determined to be the front surface by this reversed front / back determination criterion are the drawing targets. This is because, as a result, the front surface according to the determination criteria of the third embodiment is determined as the back surface according to the normal determination criteria.

FIG. 30 shows an example of triangular polygons to be judged. To the vertices of the triangular polygon shown in FIG. 30, vertex numbers 0, 1 and 2 are given in the order of upper, lower left and lower right in the figure. In the example of FIG. 30, the front side of the paper surface is the front surface in terms of how the vertex numbers are assigned, but the front side of the paper surface is the back surface according to the reversed determination criteria. In the case of the back surface based on the reversed determination criteria, this polygon is removed from the drawing target because it is usually the front surface. It should be noted that even in the third embodiment, the front / back determination is performed at this stage, but the front / back determination may be performed before this stage.

If the polygon including the apex is the front surface according to the normal judgment standard, the process returns to step S523. When the polygon including the apex is the back surface according to the normal determination standard, the determination process of whether or not to map the texture expressing the blur is performed (step S).
529). This means texture mapping for polygons. If the texture expressing the blur is mapped, the texture coordinate calculation processing for expressing the blur is performed on the vertex (step S531). texture·
As the perspective processing, here, Q = 1 / w
(W is the depth from the screen), S = U ×
The calculation of Q, T = V × Q is performed. If the texture expressing the blur is not mapped, step S
Move to 533.

Then, for example, the triangle drawing processing unit 205 and the pixel color processing unit 209 shown in FIG. 2 are driven (step S533). As described above, the triangle drawing processing unit 205 interpolates the data of each vertex of the triangular polygon to generate the data of each pixel inside the triangular polygon. The data of each vertex is the color of the material, the screen coordinate value, and the texture coordinate value if step S531 is performed. Further, the data in each pixel is the color of the material and the texel color if step S531 is performed. However, it is possible to ignore the material color at this point and set the color of the contour line at each vertex. It is also possible to set the color of the contour line in consideration of the color of the material. Pixel color processing unit 209
Writes the display image in the frame buffer 213 using the data in each pixel inside the triangular polygon generated by the triangular drawing processing unit 205. At this time, the hidden surface is erased by using the Z buffer 211.

[Stereoscopic Model Drawing Process] The flow of the stereoscopic model drawing process in the first embodiment shown in FIG. 15 is the same in the third embodiment. First, initialization is performed (step S603). In this initial setting, the brightness range table (for example, FIG. 5 or FIG. 18) corresponding to the stereo model is acquired. Next, vertex conversion (enlargement / reduction / rotation / translation / perspective conversion) and light source calculation for one vertex are performed (step S605). This is executed by the geometric calculation unit 207 according to a command from the calculation processing unit 103, for example. The data of the stereo model is, for example, CD-R131.
It is stored in.

The enlargement / reduction / rotation / translation / perspective transformation is basically based on the state in the virtual space set in step S3 of FIG. However, when the contour drawing model has the same size as the stereo model, the contour drawing model may be made relatively large by reducing the size of the stereo model. In this case, reduction conversion is performed in step S605. It should be noted that the size can be easily reduced by moving each vertex along the normal line toward the center of the three-dimensional model.

There are two methods for light source calculation in the stereo model drawing processing of the present invention, (A) a method considering the color of the material defined in the polygon, and (B) a method not considering the color of the material. These are also applicable to the third embodiment.

Next, it is judged whether or not the polygon including the vertex is the front surface (step S607). In the case of a triangular polygon, this judgment is performed depending on which direction the triangular polygon composed of this apex and the two apexes processed before is directed. For this determination, the method described in the drawing process of the contour drawing model can be used. In the third embodiment, the front / backside determination is performed at this stage, but the front / backside determination may be performed before this stage.

If the polygon including the vertex is a back surface, the process returns to step S605. If the polygon including the apex is the front surface, the brightness at the apex subjected to the apex conversion and the light source calculation is calculated (step S609). YIQ conversion is performed in the brightness calculation.

Then, the drawing color of the polygon including the vertices subjected to the vertex conversion and the light source calculation is calculated (step S61).
1). The drawing color of the polygon is calculated from the reference lightness corresponding to each lightness range stored in the lightness range table and the color of the polygon.

If the reference brightness Tn is three (T1, T2, T
3), three drawing colors are obtained. The color of the polygon is Y
When not stored as IQ system colors, that is, RGB
When the color is held as a system color, the conversion from RGB to YIQ is performed using the conversion matrix shown above. Further, the different calculation method described in the first embodiment can be applied to the third embodiment.

Next, one brightness range in the brightness range table is selected (step S613). In this embodiment, the lightness range table shown in FIG. 5 is used, but FIG.
It is also possible to use a brightness range table such as

Thereafter, the lightness at the vertex of this polygon is interpolated, and the lightness at each pixel inside the polygon (lightness distribution in the polygon) is calculated. The colors of the vertices are also interpolated, but the results are the same even if they are interpolated because all three vertices have the same drawing color. If the lightness of the pixel is within the selected lightness range, the pixel is drawn with the drawing color corresponding to the selected lightness range (step S615). The brightness interpolation processing is executed by the triangle drawing processing unit 205 in FIG. 2, for example. The comparison process of whether the brightness of each pixel is within the selected brightness range is
For example, the pixel color processing unit 209 implements it. The steps S613 and S615 are repeated until processing is performed for all the brightness ranges (step S617).

For example, when the pixel color processing unit 209 cannot handle the two lightness values of the upper limit and the lower limit, the same effect can be obtained by using the Z buffer 211 together. The Z buffer 211 is used for hidden surface removal, but in the present embodiment, the combined use of the Z buffer 211 has the same effect as when compared with the upper limit value of brightness.

In some cases, the smallest lower limit value of the brightness range table is not 0.0. In order to eliminate the part of the polygon that is not colored, step S in FIG.
In the repetition of 617, the lower limit value is set to 0.0 in the final repetition, and step S615 is executed.

The above steps S605 to S617 are repeated until all the vertices of the stereo model are processed, and as a result, all polygons are processed (step S619).

By carrying out the above processing, all polygons of the stereo model are painted with the brightness of a predetermined level, and a cel animation-like image can be obtained for the stereo model. Further, in the contour drawing model introduced in the third embodiment, a part of the surface behind the stereo model that is not hidden by the stereo model is drawn, so that part is rendered as a contour line. In the third embodiment,
To draw a contour line, a contour drawing model is introduced and the front and back determination of the contour drawing model is reversed, and the contour line can be drawn easily by performing almost the same processing as the normal rendering processing.

4. Fourth Embodiment A fourth embodiment is a case in which the processing of the third embodiment is used for the processing of the contour drawing model and the processing of the second embodiment is used for the processing of the stereo model.

An outline of the fourth embodiment of the present invention will be described with reference to the functional block diagram of FIG. The rendering apparatus illustrated as the fourth embodiment includes a contour drawing model drawing unit 850 and a stereo model drawing unit 890. The contour drawing model drawing unit 850 includes a contour drawing model acquisition unit 700, a contour drawing model placement matrix setting unit 705, a contour drawing model processing unit 710, a blur expression texture mapping unit 715, and a stereoscopic model. The pixel drawing unit 890 and the pixel processing unit 830 shared by the model drawing unit 890 are included. Each of these functions transfers data in the order described above.

Further, the stereo model drawing unit 890 includes a vertex conversion / light source calculation unit 460, a brightness calculation unit 465, a brightness range table 475, a drawing color storage unit 470, a brightness range setting unit 480, and a contour. Model drawing unit 450 for drawing
And a pixel processing unit 830 shared by The output of the vertex conversion and light source calculation unit 460 is input to the brightness calculation unit 465. The output of the brightness calculation unit 465 is input to the pixel processing unit 830. The brightness range table 475 is referred to by the brightness range setting unit 480. Drawing color storage unit 470
And the output of the brightness range setting unit 480 is the pixel processing unit 83.
Input to 0. The pixel processing unit 830 shared by the contour drawing model drawing unit 850 and the stereo model drawing unit 890 includes a brightness comparison unit 833 used in the stereo model drawing process.
And a hidden surface removal processing unit 837 used in both the contour drawing model drawing processing and the stereo model drawing processing.

The contour drawing model acquisition unit 700 generates a contour drawing model corresponding to a stereo model composed of, for example, triangular polygons. When the contour drawing model is generated in advance, the contour drawing model acquisition unit 70
0 reads out the contour drawing model composed of the triangular polygons which is generated in advance. The fourth embodiment differs from the first and second embodiments in that each polygon of the contour drawing model acquired has the same front and back as the corresponding polygon of the stereo model. The contour drawing model is larger than the stereo model and is defined by a predetermined color scheme for contour lines. Although the contour drawing model must be finally larger than the corresponding stereo model, the size of the contour drawing object at this stage may be the same as that of the stereo model. In this case, by the time the contour drawing model and the stereo model are drawn, the contour drawing model is processed to be drawn relatively larger than the stereo model.

In addition, the color of the contour drawing model may inherit the color of the material of the corresponding stereo model as it is. In this case, the drawing color is specified separately. The reference position of the contour drawing model is usually defined to be the same as or near the reference position of the corresponding stereo model.

Then, the contour drawing model placement matrix setting unit 705 (FIG. 25) places the contour drawing model reference position 630 in the virtual space at the same position as the stereo model reference position 620, as shown in FIG. Set the placement matrix to do this. This placement matrix is used to translate, rotate, and magnify each vertex of the corresponding model.
Used for conversion such as reduction. That is, the arrangement matrix of the contour drawing model 610 is set so as to include a conversion for translating the contour drawing model reference position 530 to the coordinates of the stereo model reference position 520, so that the contour drawing is performed at a position including the stereo model 600. Model 610 is arranged.

The contour drawing model processing unit 710 carries out vertex conversion (enlargement / reduction / rotation / translation / perspective conversion) for each vertex of the contour drawing model, and the front and back of each polygon of the contour drawing model. Make a decision. The above-mentioned arrangement matrix is used for this vertex conversion. However, unlike the first and second embodiments, this reverse side front / back determination unit 713 performs this front / back determination. Further, the light source calculation is not performed here. For example, in addition to enlarging / reducing / rotating / translating / perspective converting in accordance with a specified state in a virtual space that is a virtual three-dimensional space, the contour drawing model acquisition unit 700 draws a contour of the same size as the stereo model. When the contour drawing model is acquired, the contour drawing model processing unit 710 performs vertex conversion for enlarging the contour drawing model. Even when enlarged here, the relationship between the stereo model and the contour drawing model is as shown in FIG.

Further, in the case of the contour drawing model of the fourth embodiment, the front surface is determined to be the back surface, and the back surface is determined to be the front surface. Therefore, in the example of FIG. 26, only the surfaces 613, 614, 615 and 616 in which the arrows point in the same direction as the line of sight 640 from the camera 650 are the drawing targets. Normally, this surface is a back surface and thus is not a drawing target, but in the fourth embodiment, it is treated as a drawing target. In this way, the surfaces 611 and 612, which are outside the stereo model 600 and close to the camera 650, are excluded from the drawing target, so that the stereo model 600 is drawn as usual. Since the hidden surface removal processing unit 435 of the pixel processing unit 430 performs hidden surface removal, not all of these surfaces are drawn even if they are drawing targets.

Faint expression texture mapping unit 715
Performs a process for mapping the blur expression texture on the contour drawing model so that the resulting contour line becomes faint. Since the contour line does not necessarily have to be blurred, the processing of the blurred expression texture mapping unit 715 is selectively performed.

The vertex conversion and light source calculation unit 460 of the stereo model drawing unit 890 performs the vertex conversion (enlargement / reduction / rotation / translation / perspective conversion) for each vertex of the triangular polygon of the stereo model arranged in the virtual three-dimensional space. Is performed (the area in which the triangular polygon is drawn is calculated), and the light source is calculated for each vertex of the triangular polygon whose vertex has been converted. In addition, the vertex conversion of the stereo model drawing unit 890 and the light source calculation unit 460 also determine the front and back of each triangular polygon of the stereo model.

Also in the vertex conversion and light source calculation unit 460 of the stereo model drawing unit 890, enlargement / reduction / rotation / translation / movement are performed in accordance with the state specified in the virtual three-dimensional space.
Not only perspective transformation but also a contour drawing model processing unit 81
When the contour drawing model after being processed with 0 is the same size as the stereo model, the size of the stereo model is reduced so that the stereo model becomes relatively smaller than the contour drawing model. Vertex transformation is performed.

The relationship between the stereo model 600 and the contour drawing model 610 is as shown in FIG. 26 even when the vertex conversion and light source calculation unit 460 performs the stereo model reduction processing. Also,
Different from the contour drawing model processing unit 710, the surface front / back determination excludes, from the drawing target, the surface of the three-dimensional model whose front direction is the same direction as the line-of-sight direction of the camera.
In the example of FIG. 26, the rear surface 603 viewed from the camera,
604, 605 and 606 are excluded from the drawing target.

The brightness calculator 465 calculates the brightness from the color at each vertex of the triangular polygon calculated by the light source calculator 460. Since the normal light source calculation unit 460 calculates the colors in the RGB system, the brightness calculation unit 465 calculates the RGB as YI.
The brightness Y is obtained by Q conversion. The brightness at each vertex of this triangular polygon is output to the pixel processing unit 830.

The brightness range table 475 is the same as in the first embodiment.
Similar to Nos. 3 to 3, the table is, for example, as shown in FIG. Note that here, the brightness takes a real value from 0 to 1. It is also possible to specify the range by the upper limit and the lower limit instead of the threshold value (see, for example, FIG. 18).

When using the lightness range table as shown in FIG. 5, the drawing color storage section 470 needs to store three drawing color data for each polygon. As shown in FIG. 23, for each polygon, a drawing color (r, g, b) corresponding to the first lightness range and a drawing color (r, g, b) corresponding to the second lightness range , The drawing color (r, g, b) corresponding to the third lightness range, the vertex data IDX of the vertex 0 forming the polygon, the vertex data IDX of the vertex 1 and the vertex data IDX of the vertex 2. But,
It is stored instead of the above-mentioned FIG. 16 (b). These data are prepared for the number of polygons of the three-dimensional model.

The pixel processing section 830 takes out the drawing color corresponding to the brightness range set by the brightness range setting section 480 from the drawing color storage section 470. Drawing color storage unit 4
Reference numeral 70 denotes a CD-R 131, which is stored as an element of the stereo model data 137.

The brightness range setting unit 480 selects one threshold value of the brightness range table 475 and sets it in the pixel processing unit 430. When the brightness range table 475 as shown in FIG. 5 is used as it is, the brightness range setting unit 480 sets the brightness range tables 475 one by one from the top. If upper and lower bounds are specified instead of thresholds,
It can be randomly selected and set.

The pixel processing unit 830 shared by the contour drawing model drawing unit 850 and the stereo model drawing unit 890 is
The color or brightness of each pixel in the triangular polygon is obtained by interpolating the color or brightness of each vertex of the triangular polygon. The interpolation method may be the Gouraud shading algorithm or the von shading algorithm.

The pixel processing unit 830, in the case of processing a triangular polygon which is a drawing target of the contour drawing model, carries out the hidden surface removal processing by using the hidden surface removal processing unit 837, and at the same time, performs the contour drawing model. The color of each pixel in the triangular polygon that is the drawing target of is determined.

For example, in the case of FIG. 26, the three-dimensional model 60
The two faces 601 and 602 closest to the zero camera 650
Is drawn, and four surfaces 613, 614, 615, and 616 far from the camera 650 of the contour drawing model are drawn. Since these four surfaces of the contour drawing model 610 are projected to the left and right of the stereo model 600 when viewed from the camera 650, only the protruding portion is drawn without hidden surface removal. This protruding portion becomes the contour line. The pixel processing unit 830 determines the color in consideration of the color of the material of the contour drawing model. However, the color of the contour line (black or dark contour line color) may be set as the color of the contour drawing model by completely ignoring the color of the material.

On the other hand, when processing the triangular polygon which is the drawing target of the stereo model, the pixel processing unit 830 first interpolates the lightness at each apex of the triangular polygon output from the lightness calculation unit 465 to obtain the polygon. The brightness at each pixel inside is calculated.

Then, the brightness comparing section 833 compares the brightness of each pixel with the threshold value set by the brightness range setting section 480. If the brightness of the pixel is greater than or equal to the threshold value, the pixel processing unit 830 draws the pixel with a drawing color based on the reference brightness corresponding to the threshold value.
At the time of this drawing processing, hidden surface removal processing is also performed using the hidden surface removal processing unit 837. If the brightness at that pixel is less than the threshold, then this pixel is not drawn at this stage. The brightness range setting unit 480 sets the brightness range table 47.
All the threshold values of 5 are set in the pixel processing unit 830,
Correspondingly, if the pixel processing unit 830 draws all the pixels in the triangular polygon, the inside of the triangular polygon is divided into three stages in the example of FIG. This process is executed for all triangular polygons of the three-dimensional model.

Next, the processing flow for the fourth embodiment will be described. The processing described below is performed by the arithmetic processing unit 103.
(FIG. 1) is a process executed by controlling other elements in the computer main body 101.

[CD-R recording process] As shown in FIG.
The contour drawing model generation processing that is performed in advance is also performed in the fourth embodiment. When the processing is started, the data of the stereo model stored in the HDD 107 in advance is read (step S353) and is acquired as the conversion target model.

[0309] Next, the size of the model to be converted is enlarged so as to be slightly larger (step S35).
5). For example, in the normal direction of each vertex of the conversion target model,
The vertex is moved by a length of 2% of the total length of the conversion target model, and is expanded by about 2% as a whole. That is, for example, if the conversion target model is a human type and the height thereof is equivalent to 1.8 m, each vertex is 0.0
It is moved by a length equivalent to 36 m. When the enlargement ratio is larger, the contour line is drawn thicker, and when the enlargement ratio is smaller, and the conversion target model is only slightly enlarged, the contour line is drawn thinner. Furthermore, if it is not uniform and a part is enlarged, only the outline of the enlarged portion is drawn thick. Since this adjustment is usually performed by the maker of the three-dimensional model, it is possible to draw a contour line that reflects the maker's intention.

When the normal line of each vertex of the three-dimensional model is not defined, the normal line of the vertex which is obtained by interpolating the normal line of each surface sharing the vertex is used. Can also be moved in the direction of the normal to the vertex.

It is also possible to move the surface in the normal direction of each surface of the three-dimensional model. However, when the surfaces are simply moved, a gap is created between the surfaces, and a separate process for filling the gap is required. Furthermore, since the reference position is usually defined in the stereo model, centering around the reference position of the corresponding conversion target model,
It is also possible to move each vertex of the conversion target model.

Next, the color of the material of each polygon of the conversion target model is set to the color having the same saturation but lower lightness (step S357). Note that each polygon may be set to a single color such as black. Further, the setting for mapping the texture for expressing blur may be performed. Since the color of the material is adjusted by the manufacturer, the contour line can be drawn in the color intended by the manufacturer.

In the fourth embodiment, the processing of reversing the front and back of each polygon of the conversion target model is not performed as in the third embodiment. Therefore, the data of the conversion target model processed up to this point is used as the contour drawing model. .Data as HDD 107
(Step S361), and the contour drawing model generation process ends (step S363).

Next, various data including the stereo model data and the contour drawing model data stored in the HDD 107 is transferred to the CD-R 13 by the CD-R drive 113.
Written to 1. At the levels listed in Figure 7,
The example of the data written in the CD-R 131 is the same in the fourth embodiment.

[Overall Processing Flow] The processing flow at the level shown in FIG. 8 is the same as in the first to third embodiments. That is, initial setting is first performed (step S
2). This initial setting includes a contour drawing model data acquisition process (FIG. 28), which will be described in detail later. Then, the state in the virtual space is set (step S3). At this time, a process for determining the position coordinates of the contour drawing model is executed. Next, a process of determining whether or not to draw the contour line is performed (step S4). If the contour line is drawn, the drawing process of the contour line drawing model is executed (step S
5). This will be described later with reference to FIG. The stereo model is drawn regardless of whether the contour line is drawn or not (step S6). These steps S3 to S6 are repeated until the processing is completed (step S7).

[Contour Drawing Model Acquisition Processing] The contour drawing model acquisition processing shown in FIG. 28 is also implemented in the fourth embodiment. First, it is determined whether or not the contour drawing model is generated (step S223). This is because there are cases where the contour drawing model is prepared in advance and cases where the contour drawing model is generated at this stage. Here, this determination is performed by determining, for example, whether the contour drawing model corresponding to the stereo model is stored in the CD-R 131. If it is determined that the contour drawing model is stored, it is determined that the contour drawing model is not generated, and if it is determined that the contour drawing model is not stored, it is determined that the contour drawing model is generated.

If it is determined that the contour drawing model is not generated, the data of the contour drawing model stored in the CD-R 131 is read (step S22).
7). Each polygon of this contour drawing model is shown in FIG.
6 and FIG. 27, the first embodiment is described.
Unlike 2 and 2, the front and back sides are the same as the corresponding polygons of the stereo model. The size of the contour drawing model read out is defined to be slightly larger than that of the corresponding stereo model. Further, the color of the contour drawing model is defined as a color darker than the corresponding stereo model.

If it is determined that the contour drawing model is generated, the processing for generating the contour drawing model is performed (step S225). Similar to step S227, even when the contour drawing model is generated at this stage, each polygon of the contour drawing model is
The front and back are made the same as the corresponding polygon of the three-dimensional model (see FIG. 26).

The size of the contour drawing model is slightly larger than that of the corresponding stereo model. Step S35
5 (FIG. 27), for example, a contour drawing model that is enlarged by moving the vertices in the normal direction of each vertex of the stereo model is generated. When the contour drawing model is larger than the stereo model, the contour line is drawn thicker,
If the contour drawing model is only slightly larger than the stereo model, the contour line is drawn thinner.

Further, as described in the explanation of step S355 (FIG. 27), even if an enlarged contour drawing model is generated by moving the surface of the stereo model in the normal direction of the surface, the contour drawing model is generated. Good. Further, the contour drawing model may be generated by moving each vertex of the stereo model around the reference position defined in the normal stereo model.

At this point, the size of the contour drawing model may be generated in the same size as the size of the corresponding stereo model. In this case, after the contour drawing model is acquired by the main contour drawing model acquisition processing, the contour drawing is performed until the arrangement matrix of the contour drawing model is set by the contour drawing model arrangement processing described later. The model will be expanded. Alternatively, when the placement matrix of the contour drawing model is set in the contour drawing model placement process, the placement matrix may be determined to include the enlargement conversion. Conversely, when arranging the three-dimensional model, the three-dimensional model arrangement matrix may be determined such that the three-dimensional model arrangement matrix includes reduction conversion.

On the other hand, the material color of each polygon of the contour drawing model is generated by darkening the material color of each polygon of the corresponding stereo model. Note that, as described in the description of step S357 (FIG. 26), the color of the contour drawing model to be generated may not be defined at this point. Alternatively, the material color of each polygon of the contour drawing model may be the same as the material color of each polygon of the corresponding stereo model. In this case, when drawing the contour drawing model,
The color of the contour drawing model is not taken into consideration, and the contour drawing model is drawn with a separately defined color such as black or the color of the texture expressing the blur.

Next, it is determined whether or not a texture expressing blur is mapped on the contour drawing model (step S229). When the contour drawing model is generated in step S225, this determination is performed based on the data of the corresponding stereo model. On the other hand, if the contour drawing model is read in step S227,
This judgment is performed based on the read data of the contour drawing model. If it is determined that the texture expressing the blur is mapped, step S231.
At, the texture expressing the blur is mapped on the contour drawing model. That is, texture coordinates (U, V) are set at each vertex of the polygon.

As described above, the texture expressing blur is a texture having a pattern including a change in brightness or transparency, for example, the texture shown in FIG. If it is determined that the texture expressing the blur is not mapped, and if the processing for mapping the texture has ended, the arithmetic processing unit 103 ends the contour drawing model acquisition processing (step S233).

[Contour Drawing Model Arrangement Processing] In step S3 of FIG. 8, the contour drawing model arrangement matrix is set, and the contour drawing model arrangement processing is performed.
The reference position of the normal contour drawing model is provided at a position corresponding to the reference position of the stereo model. Then, the reference position of the contour drawing model is arranged to be the same as or near the position where the reference position of the stereo model is arranged,
A layout matrix for the contour drawing model is set.

When the direction of the stereo model changes, the layout matrix including the rotation conversion is set so that the contour drawing model also corresponds to it. When the shape of the three-dimensional model changes, the deformation process is performed so that the contour drawing model corresponds to it.

At this stage, if the contour drawing model has the same size as the corresponding stereo model, the contour drawing model is enlarged. Specifically, the arrangement matrix of the contour drawing model is set so that the vertices of the contour drawing model are enlarged and converted according to a predetermined enlargement ratio with the reference position of the contour drawing model as the center. Alternatively, conversely, the stereo model may be reduced. That is, in this case, the arranging matrix of the stereo model is set so that the vertices of the stereo model are reduced and converted with a predetermined reduction ratio with the reference position of the stereo model as the center.
When the contour drawing model is relatively larger than the corresponding stereo model, the stereo model arranging matrix can be used as it is as the contour drawing model arranging matrix.

In this way, the relatively large contour drawing model is finally arranged so as to include the stereo model. The contour drawing model may not completely include a stereo model depending on the arrangement position, direction, shape, and the like of both models. However, even in such a case, the contour line is drawn for the included portion.

At this stage, the arrangement matrix does not necessarily have to be set, and the arrangement coordinates,
It suffices that each element necessary for the vertex conversion such as the direction and the enlargement / reduction ratio is fixed. Also in this case, the actual vertex conversion is performed at the stage of drawing processing of each model.

[Drawing Process of Contour Drawing Model] The drawing process flow of the contour drawing model shown in FIG. 29 is also implemented in the fourth embodiment. Here, the process described below is repeated until all the vertices of the contour drawing model are processed (step S523). The first process that is repeated is the vertex conversion (expansion / enlargement) for one vertex.
Reduction / rotation / translation / perspective conversion) (step S)
525). For example, this processing is executed by the geometric calculation unit 207 instructed by the calculation processing unit 103.

It should be noted here that the light source calculation is not performed on the contour drawing model. This is because the contour line is irrelevant to the position of the light source and the light source calculation is useless (in some cases, the color of the material of the contour drawing model may be eventually ignored). Normally, this vertex transformation is performed based on the state specified in the virtual three-dimensional space. However, if the size of the contour drawing model is the same as the stereo model, according to the placement matrix set in the placement process. The contour drawing model may be enlarged and converted at this stage.

Then, the polygon including the apex is subjected to a determination process as to whether or not it is a back face according to a normal determination criterion (step S527). Normally, only the front surface is the drawing target, but in the case of the contour drawing model of the fourth embodiment, the back surface is the drawing target according to the normal determination standard. In the case of a triangular polygon, the determination in this step is made based on which direction the triangular polygon composed of this apex and the two apexes processed before is directed.

As shown in FIG. 12, for example, when each vertex of a triangular polygon is provided with a vertex number in a counterclockwise direction according to a normal criterion, it is defined that the front side of the drawing is the front side (so-called). Right hand system). In the fourth embodiment, the front and back determination criteria are reversed, and when the vertex numbers are given clockwise, it is determined that the front side of the page is the front side. Only the surface that is determined to be the front surface by this reversed front / back determination criterion is set as the drawing target. This is because, as a result, the front surface according to the determination criteria of the fourth embodiment is determined as the back surface according to the normal determination criteria.

FIG. 30 shows an example of triangular polygons to be judged. To the vertices of the triangular polygon shown in FIG. 30, vertex numbers 0, 1 and 2 are given in the order of upper, lower left and lower right in the figure. In the example of FIG. 30, the front side of the paper surface is the front surface in terms of how the vertex numbers are assigned, but the front side of the paper surface is the back surface according to the reversed determination criteria. In the case of the back surface based on the reversed determination criteria, this polygon is removed from the drawing target because it is usually the front surface. It should be noted that even in the fourth embodiment, the front / back determination is performed at this stage, but the front / back determination may be performed before this stage.

If the polygon including the apex is the front surface according to the normal judgment criteria, the process returns to step S523. When the polygon including the apex is the back surface according to the normal determination standard, the determination process of whether or not to map the texture expressing the blur is performed (step S).
529). This means texture mapping for polygons. If the texture expressing the blur is mapped, the texture coordinate calculation processing for expressing the blur is performed on the vertex (step S531). texture·
As the perspective processing, here, Q = 1 / w
(W is the depth from the screen), S = U ×
The calculation of Q, T = V × Q is performed. If the texture expressing the blur is not mapped, step S
Move to 533.

Then, for example, the triangle drawing processing unit 205 and the pixel color processing unit 209 shown in FIG. 2 are driven (step S533). As described above, the triangle drawing processing unit 205 interpolates the data of each vertex of the triangular polygon to generate the data of each pixel inside the triangular polygon. The data of each vertex is the color of the material, the screen coordinate value, and the texture coordinate value if step S531 is performed. Further, the data in each pixel is the color of the material and the texel color if step S531 is performed. However, it is possible to ignore the material color at this point and set the color of the contour line at each vertex. It is also possible to set the color of the contour line in consideration of the color of the material. Pixel color processing unit 209
Writes the display image in the frame buffer 213 using the data in each pixel inside the triangular polygon generated by the triangular drawing processing unit 205. At this time, the hidden surface is erased by using the Z buffer 211.

[Stereoscopic Model Drawing Process] The stereoscopic model drawing process flow in the second embodiment shown in FIG. 24 is the same in the fourth embodiment. First, initialization is performed (step S603). In this initial setting, the brightness range table corresponding to the stereo model (for example, FIG. 5 or 18)
Is obtained. Further, the color data for drawing the stereo model is acquired. Next, vertex conversion (enlargement / reduction / rotation / translation / perspective conversion) and light source calculation for one vertex are performed (step S635). This is executed by the geometric calculation unit 207 according to a command from the calculation processing unit 103, for example. The data of the stereo model is stored in the CD-R 131, for example.

The enlargement / reduction / rotation / translation / perspective transformation is basically based on the state in the virtual space set in step S3 of FIG. However, when the contour drawing model has the same size as the stereo model, the contour drawing model may be made relatively large by reducing the size of the stereo model. In this case, reduction conversion is performed in step S635. It should be noted that the size can be easily reduced by moving each vertex along the normal line toward the center of the three-dimensional model.

The two methods of light source calculation in the stereo model drawing processing of the present invention described in the first embodiment can be applied to the fourth embodiment as they are.

Next, it is determined whether or not the polygon including the vertex is the front surface (step S637). In the case of a triangular polygon, this judgment is performed depending on which direction the triangular polygon composed of this apex and the two apexes processed before is directed. For this determination, the method described in the drawing process of the contour drawing model can be used. In the fourth embodiment, the front / back determination is performed at this stage, but it is also possible to perform the front / back determination before this stage.

If the polygon including the apex is the back surface, the process returns to step S635. If the polygon including the apex is the front surface, the brightness at the apex subjected to the apex conversion and the light source calculation is calculated (step S639). YIQ conversion is performed in the brightness calculation.

Then, the drawing color of the polygon including the vertices subjected to the vertex conversion and the light source calculation is read from the memory 105, for example (step S641). The drawing color data to be read is calculated in advance, but the calculation method for this calculation may be either of the two methods described in the first embodiment, or another method. Good. Further, the drawing colors may be defined one by one. In the fourth embodiment, drawing colors are prepared in advance, so the execution speed is faster than in the third embodiment.

Next, one brightness range in the brightness range table is selected (step S643). In this embodiment, the lightness range table shown in FIG. 5 is used, but FIG.
It is also possible to use a brightness range table such as When such a brightness range table is used, a brightness range including an upper limit and a lower limit can be randomly selected and set. However, FIG. 18 shows a case where the computer is effective up to the second decimal place. In the lightness comparison process described below, when it is not easy to compare the lightness of each pixel with two lightness values of the upper limit and the lower limit, the lightness range is set to, for example, FIG.
Select from top 8 in order. In this case, only the lower limit value will be processed.

Thereafter, the lightness at the vertex of this polygon is interpolated, and the lightness at each pixel inside the polygon (lightness distribution in the polygon) is calculated. The colors of the vertices are also interpolated, but the results are the same even if they are interpolated because all three vertices have the same drawing color. Then, if the brightness of the pixel is within the selected brightness range, the pixel is drawn with the drawing color corresponding to the selected brightness range (step S645). The brightness interpolation processing is executed by the triangle drawing processing unit 205 in FIG. 2, for example. The comparison process of whether the brightness of each pixel is within the selected brightness range is
For example, the pixel color processing unit 209 implements it. The steps S643 and S645 are repeated until processing is performed for all the brightness ranges (step S647).

For example, when the pixel color processing unit 209 cannot handle the two lightness values of the upper limit and the lower limit, the Z buffer 211 can be used together to obtain the same effect. The Z buffer 211 is used for hidden surface removal, but in the present embodiment, the combined use of the Z buffer 211 has the same effect as the case of comparing with the upper limit value of brightness.

The smallest lower limit value in the brightness range table may not be 0.0. In order to eliminate the portion of the polygon that is not colored, step S in FIG.
In the iteration of 647, the lower limit value is set to 0.0 in the last iteration, and step S645 is executed.

The above steps S635 to S647 are repeated until all the vertices of the stereo model are processed, and as a result, all the polygons are processed (step S649).

By carrying out the above-mentioned processing, all polygons of the stereo model are painted with the brightness of a predetermined level, and a cel animation-like image can be obtained for the stereo model. In particular, the fourth embodiment is faster than the third embodiment. Further, in the contour drawing model introduced in the fourth embodiment, a part of the surface behind the stereo model that is not hidden by the stereo model is drawn, and thus the part is rendered as a contour line. In the fourth embodiment, a contour drawing model is introduced to draw a contour line, and the contour drawing model is simply drawn by performing the same processing as the normal rendering processing only by reversing the front / back determination of the contour drawing model. become able to.

5. Other Embodiments (1) In the first and third embodiments, the process of calculating the drawing color of a polygon is step S61 after steps S605 to S609 in FIG. 15 showing the stereo model drawing process.
It is supposed to be implemented as 1, but step S6
There is no problem if it is calculated before the drawing color is used in 15. Therefore, step S611 is replaced by step S60.
9 or S605, may be performed in parallel with steps S609 and S605, may be performed after step S613, or may be performed in parallel with step S613.

(2) In step S609 in FIG. 15, which represents the stereo model drawing processing of the first and third embodiments, Y is selected.
The lightness Y of the vertex of the polygon after the light source calculation is calculated by IQ conversion. It is faster to not calculate because I and Q which are the results of YIQ conversion are not used. However, if a routine or the like for performing YIQ conversion already exists, I and Q are used.
And Q may be calculated.

The same applies to the second and fourth embodiments.

(3) In step S615 in FIG. 6, which represents the stereo model drawing processing of the first and third embodiments, the data of the vertices of the polygon are interpolated to generate the data of the pixels inside the polygon. Since this processing does not change even if different lightness ranges are selected and set by repeating step S617 once performed, the result may be stored and used.

The same applies to the second and fourth embodiments.

(4) In step S605 in FIG. 15 showing the stereo model drawing processing of the first embodiment, perspective transformation and light source calculation are performed, but perspective transformation may be performed up to step S615. However, step S61
Putting it outside the loop of 7 eliminates the need for perspective transformation many times. Therefore, if it is executed at the timing of step S605, the calculation amount can be reduced.

The same applies to the second and fourth embodiments.

(5) In the first to fourth embodiments, the stereo model and the contour drawing model are composed of a plurality of triangular polygons, but they may be composed of a plurality of polygons including polygon polygons of quadrangle or more. . Further, both models may be composed of a plurality of surfaces including curved surfaces, and each surface may be processed by being approximated by one or a plurality of polygons. In this case, the inversion of each polygon of the contour drawing model in Embodiments 1 and 2 may be performed by the inversion of each surface of the contour drawing model. Since each surface is approximated by one or a plurality of polygons before being drawn, if the front and back of each surface are reversed, the front and back of each polygon will be reversed.

(6) In FIG. 4 and FIG. 26, the surface of the contour drawing model and the surface of the stereo model are one-to-one.
It is also possible to reduce the number of faces of the contour drawing model.
This is because if the number of faces is reduced, the processing speed is increased. However, the surface of the contour drawing model has a surface corresponding to that of the stereo model.

(7) The order of steps S4 and S6 of the processing flow shown in FIG. 8 can be interchanged.

(8) Change of Hardware Used In the embodiment described above, the graphics processing unit 111 executes a part of the stereo model and contour drawing model drawing processing. However, the graphics processing unit 111 or the arithmetic processing unit 103 may perform the entire stereo model and contour drawing model drawing process.

Further, FIG. 1 is an example, and various changes can be made. For example, in the case of a game device, the interface unit 117 may be provided with a memory card read / write interface for storing data. Further, whether or not the communication interface 115 is provided is arbitrary. Since the present invention is not directly related to sound processing, it is not necessary to include the sound processing unit 109.

The CD-R is an example of a recording medium and includes an internal memory such as RAM, a floppy disk,
It may be another recording medium such as a magnetic disk or a DVD-RAM. In that case, the CD-R drive 113 needs to be a readable / writable drive on a corresponding medium. Further, according to the present invention, the process up to writing to the recording medium and FIG.
The processing shown in 1) is independent, and each can be operated by different computers. Figure 8
The computer shown in FIG. 8 may be equipped with a drive that can only read the programs and data stored in the medium, because the process shown in FIG. Good. That is, the recording medium further includes an internal memory such as a ROM and a CD.
A read-only recording medium such as -ROM, DVD-ROM, and memory cartridge may be used. In that case CD
-R drive 113 must be a drive that can read the corresponding media.

Further, the case where the present invention is implemented by a computer program has been described above.
It is also possible to mount the program and a dedicated device such as an electronic circuit in combination, or only a dedicated device such as an electronic circuit. At that time, FIG. 8, FIG. 9, FIG.
5, FIG. 8, FIG. 9, FIG. 15 and FIG. 24, FIG. 8, FIG. 28, FIG. 29 and FIG.
The device may be configured for each function shown in each step of step 4, or the device may be configured for each part or a combination thereof.

Although the present invention has been specifically described based on the embodiments above, the present invention is not limited to the above-mentioned embodiments and can be appropriately modified without departing from the gist thereof. For example, in the above embodiment, the case where the present invention is realized by using a normal computer as a platform has been described, but the present invention may be realized by using a home game machine, an arcade game machine, or the like as a platform.
In some cases, personal digital assistants, car navigation
It may be possible to implement the system as a platform.

The programs and data for implementing the present invention are not limited to the form provided by a recording medium such as a CD-R that can be attached to and detached from a computer or a game machine. That is, as the programs and data for implementing the present invention, the programs and data are recorded in the memory of another device on the network 151 connected through the communication line 141 by the communication interface 115 shown in FIG. , This program and data communication line 141
Alternatively, it may be stored in the memory 105 and used in sequence as required.

[Display Example] FIG. 32 shows a display example of an image when the present invention is not used, that is, when the color of each pixel in the polygon is interpolated by the color of the vertex of the polygon. For example, it can be seen that the brightness naturally changes from the ear area of the face of the person in the center of the image to the right side of the person's face.
On the other hand, FIG. 33 shows a display example of an image when the brightness range table in which two brightness ranges (threshold values) are defined is used and drawn by the algorithm of the present invention. The brightness range table used in FIG. 33 is shown in FIG. Here, the reference brightness is set to 0.75 with respect to the threshold value 0.3125, and the reference brightness is set to 0.60 with respect to the threshold value 0. It can be seen that, unlike FIG. 32, FIG. 33 is colored in two levels of brightness from the ear area of the person's face in the center of the image to the right side of the person's face. Furthermore, you can see that the contour lines of hair, body and bag are also drawn.

By using the algorithm of the present invention, it is possible to obtain a cell animation-like image with a contour drawn by a simple process. In the present invention, the process of introducing a contour drawing model and comparing the α value storing the lightness with the lightness range when writing to the frame buffer and determining whether or not the predetermined drawing color can be written is unnecessary. Since it is only performed in step 1, it is possible to easily switch the generation of the image as shown in FIG. 32 by the conventional technique and the generation of the image as shown in FIG. In addition, when the cel animation is drawn by a human hand, it takes a lot of work to create images of various states of the character, for example. Also, in a game in which cel animation-style game characters are displayed, it is not possible to create images of characters from many angles for the same reason. However, by using the present invention, it is possible to easily obtain cel animation-like images in a large number of states, and the number of steps can be greatly reduced.

[0367]

As described above, according to the present invention, a three-dimensional model arranged in a virtual space is drawn with a color determined based on a typical lightness assigned to a lightness level, and based on the three-dimensional model. The inside of the contour drawing model that is generated and includes the stereo model is drawn with a predetermined color scheme. As a result, a rendering method and device for rendering a stereoscopic image of a stereo model by drawing a predetermined color on the stereo model on which the contour line is drawn, and a computer-readable recording medium storing a rendering program are provided. Could be provided.

[Brief description of drawings]

FIG. 1 is a block diagram showing an example of a computer that executes a program according to the present invention.

FIG. 2 is a block diagram showing an example of a graphics processing unit in FIG.

FIG. 3 is a functional block diagram of the first embodiment.

FIG. 4 is a schematic diagram for explaining a positional relationship between a camera, a stereo model, and a contour drawing model in the first and second embodiments. The direction in which the front surfaces of the stereo model and the contour drawing model face is indicated by an arrow.

FIG. 5 is a table showing an example of a brightness range table.

FIG. 6 is a flowchart of contour drawing model generation processing according to the first and second embodiments.

FIG. 7 is a schematic diagram showing an example of data written in a CD-R 131.

FIG. 8 is a flowchart showing the overall processing of the present invention.

FIG. 9 is a flowchart of contour drawing model acquisition processing according to the first and second embodiments.

FIG. 10 is an example of a texture for expressing blur.

FIG. 11 is a flowchart of contour drawing model drawing processing according to the first and second embodiments.

FIG. 12 is a schematic diagram for explaining front / back determination of a triangular polygon.

FIG. 13 is a schematic diagram for explaining a method for determining the front and back sides.

FIG. 14 is a schematic diagram for explaining front / back determination of triangular polygons in the first and second embodiments.

FIG. 15 is a flowchart of a stereo model drawing process according to the first and third embodiments.

FIG. 16 is a schematic diagram showing a data structure of a polygon model. (A) shows the data structure of the entire stereo model,
(B) shows the data structure of the triangular polygon before perspective transformation, and (c) shows the data structure of the vertex data table.

FIG. 17 shows a data structure of a triangular polygon after perspective transformation corresponding to FIG.

FIG. 18 is a table showing an example of a brightness range table.

FIG. 19 is an example of an image for explaining each step of processing when a triangular polygon is drawn by the algorithm of the present invention. (A) shows an area to be drawn when a threshold value of 0.75 is set, and (b) shows a threshold value of 0.5 when the Z buffer is not used in the first embodiment. FIG. 6C shows a range to be drawn in this case, and FIG. 7C shows a range to be drawn when the threshold value 0.5 is set when the Z buffer is used in the first embodiment.

FIG. 20 is an example of an image when a triangular polygon is drawn by the algorithm of the present invention.

FIG. 21 is an example of an image when a triangular polygon is drawn by a conventional technique.

FIG. 22 is a functional block diagram of the second embodiment.

FIG. 23 is a schematic diagram of a data structure of a triangular polygon used in the second and fourth embodiments. This corresponds to FIG. 16 (b).

FIG. 24 is a flowchart of a stereo model drawing process according to the second and fourth embodiments.

FIG. 25 is a functional block diagram of the third embodiment.

FIG. 26 is a schematic diagram for explaining the positional relationship between the camera, the stereo model, and the contour drawing model in the third and fourth embodiments. The direction in which the front surfaces of the stereo model and the contour drawing model face is indicated by an arrow.

FIG. 27 is a flowchart of contour drawing model generation processing in the third and fourth embodiments.

FIG. 28 is a flowchart of contour drawing model acquisition processing according to the third and fourth embodiments.

FIG. 29 is a flowchart of contour drawing model drawing processing according to the third and fourth embodiments.

FIG. 30 is a schematic diagram for explaining front / back determination of triangular polygons according to the third and fourth embodiments.

FIG. 31 is a functional block diagram of the fourth embodiment.

FIG. 32 is a display example of an image rendered by a conventional technique.

FIG. 33 is a display example of an image rendered using the present invention.

34 is a table showing an example of a brightness range table used in the rendering of FIG. 33. FIG.

[Explanation of symbols]

1000 computer 101 computer main body 103 arithmetic processing unit 105 memory 107 HDD 109 sound processing unit 111 graphics processing unit 113 CD-R drive 115 communication interface 117 interface unit 119 internal bus 121 display device 125 sound output device 131 CD-R 141 communication medium 151 network 161 input device 201 bus control unit 205 triangle drawing processing unit
207 Geometric calculation unit 209 Pixel color processing unit 211 Z buffer 213 Frame buffer 300, 700 Contour drawing model acquisition unit 305, 705 Contour drawing model placement matrix setting unit 310, 710 Contour drawing model processing unit 315, 715 Expression texture mapping unit 330, 430, 730, 830 Pixel processing unit 333, 433, 733, 833 Brightness comparison unit 337, 437, 737, 837 Hidden surface removal processing unit 360, 460 Vertex conversion and light source calculation unit 365, 465 Brightness Calculation unit 370 Drawing color calculation unit 375, 475 Lightness range table 380, 480
Brightness range setting unit 470 Drawing color storage units 350, 450, 750, 850 Contour drawing model drawing units 390, 490, 790, 890 Stereo model drawing unit

Continuation of the front page (56) Reference JP 2001-84394 (JP, A) JP 2001-79261 (JP, A) JP 2001-70634 (JP, A) JP 2000-251094 (JP, A) Kai 2000-172880 (JP, A) Mitsuru Kaneko, Algorithm for generating border lines of 3D CG images, 1994 Autumn Meeting of the Institute of Electronics, Information and Communication Engineers, Japan, The Institute of Electronics, Information and Communication Engineers, September 5, 1994, No. 6, 245 Mitsuru Kaneko, Non-photorealistic animation generation, IPSJ research report (Graphics and CAD 76-4), Japan, IPSJ, August 17, 1995, Vol. 95, No. 78,23-30 Masayuki Nakajima, Animation and Computer, Information Processing, Japan, Information Processing Society of Japan, July 15, 1998, Vol. 39 No. 7,613-620 The world of non-photoreal rendering, Nikkei CG, Japan, Nikkei BP, May 8, 1998, No. 140,110-149 Stefan Schlechtwe gl et al, Surfaces to Lines: Renderin g Rich Line Drawin gs, Proceedings of WSCG'98, The Sixth I nternational Confe rence in Central E urope on Computer Graphics an, Centra l Europe on Comput er Graphics and Vi sulizat, February 9, 1998, vol. 2,356-361 R.I. Raskar, Image Precision Silhouette Edges, Proceedings of the 1999 symposium on Interactive 3D graphics, USA, ACM SIGGRAPH, April 26, 1999, w / wp, 135-140, 135-140. cs. unc. edu / ￣rask ar / NPR / sil-i3d99. pdf f Kaneko Isamu, AnimBody, as of October 30, 2002, URL, http: // homepage. nifty.com/kaneko/anibody. htm Craig W. Reynolds, Stylized Deposition in Computer Graphics, as of October 30, 2002, UR L, http: // www. red3 d. com / cwr / npr / (58) Fields investigated (Int.Cl. ⁷ , DB name) G06T 15/00 G06T 15/50 200 G06T 15/70 JISST file (JOIS)

Claims (12)

(57) [Claims] 1. A rendering method for rendering a stereo model, which is arranged in a virtual space and is composed of a plurality of polygons that represent the outside of an object to be represented, said rendering method corresponding to said stereo model and of said stereo model. A first step of acquiring a contour drawing model in which the front and back of polygons corresponding to each polygon are reversed and storing it in a storage device, and a second step of arranging the contour drawing model at a position including the stereo model. Of the contour drawing model, a third step of drawing only polygons facing the predetermined viewpoint position with a predetermined color scheme and storing them in a storage device, and a level division within a certain range of brightness And a representative lightness is assigned to each level, and the first lightness distribution, which is the lightness distribution of the region where the stereo model is to be drawn, is set to the lightness level. A color distribution is generated based on a second lightness distribution replaced with the representative lightness assigned to each and a color preset in the three-dimensional model, and the three-dimensional model is drawn with the color distribution. A fourth step of storing in a storage device, and a rendering method comprising: 2. A rendering method for rendering a stereo model, which is arranged in a virtual space and is composed of a plurality of polygons that represent the outside of an object to be represented, wherein a contour drawing model corresponding to the stereo model is used. Acquired,
A first step of storing in a storage device; a second step of arranging the contour drawing model at a position including the stereo model; and a back side facing a predetermined viewpoint position of the contour drawing model. The third step of drawing only the polygons in a predetermined color scheme and storing in the storage device, and the brightness is divided into levels within a certain range, and typical brightness is assigned to each level, and the three-dimensional model is drawn. Based on a second lightness distribution obtained by replacing the first lightness distribution, which is the lightness distribution of the region to be processed, with the representative lightness assigned to each lightness level, and a color preset in the stereo model. A fourth step of generating a color distribution by using the color distribution, drawing the three-dimensional model with the color distribution, and storing the three-dimensional model in a storage device. 3. The rendering method according to claim 1, wherein the third step includes a step of mapping a texture having a pattern including a change in brightness or transparency. 4. A computer-readable recording medium that stores a program for rendering a stereoscopic model that is arranged in a virtual space and is composed of a plurality of polygons that represent the outside of an object to be represented. A first step of acquiring, in the computer, a contour drawing model corresponding to the stereo model and having polygons corresponding to the polygons of the stereo model, the front and back sides of which are reversed, and storing the model in a storage device; And a second step of arranging the contour drawing model at a position including a line, and drawing and storing only a polygon of the contour drawing model facing a predetermined viewpoint position in a predetermined color scheme. The third step of storing in the device is that the brightness is divided into levels within a certain range, and a typical brightness is assigned to each level. A second lightness distribution obtained by replacing the first lightness distribution, which is the lightness distribution of the area where the three-dimensional model is to be drawn, with the representative lightness assigned to each lightness level, and preset in the three-dimensional model. A computer-readable program for executing a fourth step of generating a color distribution based on the colors and drawing the three-dimensional model with the color distribution and storing the same in a storage device. Recording medium. 5. A computer-readable recording medium that stores a program for rendering a stereoscopic model that is arranged in a virtual space and is composed of a plurality of polygons that represent the outside of an object to be represented. Acquires a contour drawing model corresponding to the stereo model in the computer,
A first step of storing in a storage device; a second step of arranging the contour drawing model at a position including the stereo model; and a back side facing a predetermined viewpoint position of the contour drawing model. The third step of drawing only the polygons in a predetermined color scheme and storing in the storage device, and the brightness is divided into levels within a certain range, and typical brightness is assigned to each level, and the three-dimensional model is drawn. Based on a second lightness distribution obtained by replacing the first lightness distribution, which is the lightness distribution of the region to be processed, with the representative lightness assigned to each lightness level, and a color preset in the stereo model. A fourth step of generating a color distribution by drawing the three-dimensional model with the color distribution and storing it in a storage device. A readable recording medium. 6. The computer-readable recording medium according to claim 4, wherein the third step includes a step of mapping a texture having a pattern including a change in brightness or transparency. 7. A rendering device that renders a stereo model that is arranged in a virtual space and is composed of a plurality of polygons that represent the outside of an object to be represented, the rendering device corresponding to the stereo model and that of the stereo model. A unit for obtaining a contour drawing model in which the front and back of polygons corresponding to each polygon are reversed, a unit for arranging the contour drawing model at a position including the stereo model, and a predetermined unit of the contour drawing model. A means for drawing only polygons facing the table with respect to the viewpoint position in a predetermined color scheme, and the brightness is divided into levels within a certain range, and representative brightness is assigned to each level. The second lightness distribution in which the first lightness distribution, which is the lightness distribution of the area to be drawn, is replaced with the representative lightness assigned to each lightness level And a unit configured to generate a color distribution based on a color preset in the stereo model and draw the stereo model with the color distribution. 8. A rendering device for rendering a stereo model, which is arranged in a virtual space and is composed of a plurality of polygons representing the outside of an object to be represented, wherein a contour drawing model corresponding to the stereo model is provided. A means for acquiring, a means for arranging the contour drawing model at a position including the stereo model, and only a polygon of the contour drawing model whose back is turned to a predetermined viewpoint position are predetermined. The contour drawing model drawing means for drawing with a color scheme, and the lightness is divided into levels within a certain range, and a typical lightness is assigned to each level, which is a lightness distribution of an area in which the stereo model is to be drawn. Based on a second lightness distribution obtained by replacing the lightness distribution of 1 with the representative lightness assigned to each lightness level and a color preset in the stereo model. And a means for generating a color distribution and drawing the three-dimensional model with the color distribution. 9. The rendering apparatus according to claim 7, wherein the contour drawing model drawing means includes means for mapping a texture having a pattern including a change in brightness or transparency. 10. A game device for rendering a three-dimensional model, which is arranged in a virtual space and is composed of a plurality of polygons representing the outside of an object to be represented, and stores a computer and a program to be executed by the computer. did,
A computer-readable recording medium, the program, in the computer, to obtain a contour drawing model in which the front and back of polygons corresponding to the three-dimensional model and corresponding to each polygon of the three-dimensional model are reversed, A function of storing in a storage device, a function of arranging the contour drawing model at a position including the stereo model, and only a polygon of which the table is directed to a predetermined viewpoint position of the contour drawing model in advance A model drawing function for contour drawing that draws in a predetermined color scheme and stores it in a storage device, and the brightness is divided into levels within a certain range and representative brightness is assigned to each level, and the stereo model is drawn. A second lightness distribution obtained by replacing the first lightness distribution, which is the lightness distribution of the region to be replaced, with the representative lightness assigned to each lightness level; Serial game device to generate a color distribution based on a preset color three-dimensional model, characterized in that it perform a function of storing in the storage device to draw the three-dimensional model in the color distribution. 11. A game device for rendering a three-dimensional model, which is arranged in a virtual space and is composed of a plurality of polygons that represent the outside of an object to be represented, and stores a computer and a program to be executed by the computer. did,
A computer readable recording medium, the program, in the computer, to obtain a contour drawing model corresponding to the stereo model,
A function of storing in a storage device, a function of arranging the contour drawing model at a position including the stereo model, and only a polygon of the contour drawing model whose back is turned to a predetermined viewpoint position in advance. A model drawing function for contour drawing that draws in a predetermined color scheme and stores it in a storage device, and the brightness is divided into levels within a certain range and representative brightness is assigned to each level, and the stereo model is drawn. Based on a second lightness distribution obtained by replacing the first lightness distribution, which is the lightness distribution of the area to be replaced, with the representative lightness assigned to each lightness level, and a color preset in the stereo model. And a function of generating a color distribution, drawing the three-dimensional model with the color distribution, and storing the three-dimensional model in a storage device. 12. The contour drawing model drawing function includes a function of mapping a texture having a pattern including a change in brightness or transparency.
Alternatively, the game device according to item 11.