JPH0973547A - Rendering device/mapping device and rendering method/ mapping method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To generate the picture of high quality at high speed by deciding a picture element including a plotting point on a polygon projected on a plane and executing rendering calculation based on a sample point having a fixed position for the picture element. SOLUTION: The arithmetic part 101 of an address generation part in a mapping device calculates the parameter used for plotting point calculation and corresponding point calculation in the polygon by an increased algorithm by using the position of a polygon apex and the corresponding point. An edge generation part 102a executes calculation for deciding the positions of the plotting point and the corresponding point on a polygon edge by using the parameter, and a correction part 103a executes a correction processing for the corresponding point (texture picture) which the edge generation part 102a generates by using the parameter and the plotting point which the edge generation part 102a decides. Namely, the picture element including the plotting point on the polygon projected on the plane having the plural picture elements is decided, rendering calculation is executed based on the sample point having the fixed point for the picture element and a calculated result is outputted.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a rendering device and method in computer graphics (CG) processing, and more particularly to a mapping device and mapping method for performing texture mapping to generate a high quality image. Is. Further, the present invention can be applied to various CG calculations such as illuminance calculation, opacity calculation, and bump calculation, which are performed when generating a CG.

[0002]

2. Description of the Related Art In recent years, computer graphics and
In the field of game machines and the like, polygon rendering processing using texture mapping is often performed. Usually, in the texture mapping method called inverse map, the coordinates of the corresponding texture pattern (u, v) are calculated from the screen coordinates (Xs, Ys), and the color of the texture point (corresponding point) at the calculated coordinates is converted into the screen coordinates. (Xs, Ys)
To reflect.

This corresponding point calculation is performed by the incremental algorithm (Increment Algorithm) from the viewpoints of high speed processing, simplification of constituent circuits, etc.
al Algorithm: In each step, incremental calculation is performed based on the calculation result of the previous step) (George Wolberg, “Digital Image Warpin
g, ”pp189-pp204).

FIGS. 37 and 38 are diagrams showing the texture mapping processing by the incremental algorithm. Here, it is assumed that the relationship of (Equation 1) and (Equation 2) is established between the point (x, y) on the polygon of the screen coordinates and the point (u, v) on the texture of the texture coordinates. Hereinafter, the calculation for obtaining (u, v) by (Equation 1) and (Equation 2) will be referred to as corresponding point calculation.

[0005]

## EQU1 ## u = F (x, y) = a × x + b × y + c

[0006]

## EQU00002 ## v = G (x, y) = d.times.x + e.times.y + f In FIG. 37, the part surrounded by the grid is 1 of the screen.
It represents a pixel. The triangles in the figure indicate polygons projected on the screen. Point 1401,
1402 and 1403 are vertices that form this polygon. The point 1401 is the 0th vertex, the point 1402 is the first vertex, and the point 1403 is the second vertex, and the respective screen coordinates and the corresponding texture coordinates are (8,0,0,0), (0,
6,0,16) and (11,9,16,16). The side 1404 is the 0th
Edge 0 that is connected by the vertex 1401 and the first vertex 1402 of
1, the side 1405 is the first vertex 1402 and the second vertex 14
The side 12 and the side 1406 connected by 03 are the second vertices 140.
The side 20 is connected to 3 and the 0th vertex 1401. Furthermore, points 1002, 1004, and 1006 represent drawing points on the screen calculated by the incremental algorithm. The drawing point is set on the polygon projected on the screen when the polygon is drawn. The pixel value to be given to each pixel is determined based on the coordinate value of the drawing point.
An incremental algorithm is often used as a method of determining a drawing point. The incremental algorithm determines a drawing point based on a point on the polygon edge (boundary of the polygon).
In particular, the mapping device determines a corresponding point corresponding to each drawing point on a mapping image (for example, a texture image) and reflects the pixel value of the corresponding point on a pixel on the screen to perform mapping. The determination of the corresponding points is performed by calculating the corresponding points.

Details of calculation of corresponding points at polygon edges will be described below. In FIG. 38, the portion surrounded by the grid is one pixel on the screen. FIG. 38 shows a part of the side 1404 in FIG. 37 and 38
Points 1002, 1004, and 1006 in represent the same drawing point. The edge 1001 of the polygon drawn by texture mapping is a part of the side 1404.
A polygon to be drawn exists on the right side of 001. The drawing point on the polygon edge obtained in the previous step of the incremental algorithm is set to 1002, and the pixel to which the pixel value of the edge drawing point 1002 is reflected is set to 1003. For the polygon edge, the edge drawing point found in the current step is 1004.
Then, the pixel to which the pixel value of the edge drawing point 1004 is reflected is set to 1005.

The drawing points on the polygon edge and the corresponding points corresponding to the drawing points are calculated by sequentially changing the scan line, and when the scan line scans the entire polygon, the processing relating to the polygon edge is ended. Here, the scan line is set as a horizontal line on the screen, and the entire polygon is scanned by changing the scan line from the top to the bottom of the screen. In other words, polygon scanning is synonymous with adding y values of scan lines one by one.

At this time, the screen coordinates of the edge drawing point 1002 and the texture coordinates of the corresponding point are set to (x0, y0, u
0, v0) and the screen coordinates of the edge drawing point 1004 and the texture coordinates of the corresponding points are (x1, y1, u1, v1), these can be expressed by the following equation (Equation 3).

[0010]

[Expression 3] (x1, y1, u1, v1) = (x0 + dx / dy, y0 + 1, u0 + du / dy, v0 + dv / dy) Similarly, (Expression 4) corresponds to the edge texture mapping. In the general formula for point calculation, the subscript indicates the scan line number.

[0011]

(Xn + 1, yn + 1, un + 1, vn + 1) = (xn + dx / dy, yn + 1, un + du / dy, vn + dv / dy) ((expression 1) , (U and v can be differentiated from y from (Equation 2)) By performing the processing of (Equation 4) from the start point to the end point of the polygon edge, the position of the drawing point (screen coordinate value) and the position of the corresponding point with respect to the polygon edge (Texture coordinate value) can be calculated.

Here, the parameters (DDA parameters) required to realize the processing by the incremental algorithm
Is dx / dy, du / dy, dv / dy, respectively

[0013]

[Expression 5] dx / dy = (0-8) / (6-0) = -8/6 du / dy = (0-0) / (6-0) = 0/6 dv / dy = (16- 0) / (6-0) = 16/6.

Calculation of screen coordinates and texture coordinates at the polygon edge in FIG. 37 can be realized as follows. Here, the calculation is performed for the side 1404.

(8.00, 0.00, 0.00, 0.00) ……… (initial value) ↓ + (-8/6, 1, 0, 16/6) (6.67, 1.00, 0.00, 2.67) ↓ + (-8 / 6, 1, 0, 16/6) (5.33, 2.00, 0.00, 5.33) ↓ + (-8/6, 1, 0, 16/6) (4.00, 3.00, 0.00, 8.00) ↓ + (-8 / 6, 1, 0, 16/6) (2.67, 4.00, 0.00, 10.67) ↓ + (-8/6, 1, 0, 16/6) (1.33, 5.00, 0.00, 13.33) ↓ + (-8 / 6, 1, 0, 16/6) (0.00, 6.00, 0.00, 16.00) ... (End) Next, the corresponding point calculation in the polygon span (internal polygon point) will be described. The point of the polygon span obtained in the previous step (hereinafter referred to as the span point) is the span point 10
04, and the pixel in which the pixel value of the span point 1004 is reflected is pixel 1005. The point of the polygon span to be obtained in the current step is set to the span point 1006 and the span point 10
A pixel in which the pixel value of 06 is reflected is defined as a pixel 1007.
The polygon span point 1004 is the same as the polygon edge point. Note that the span points include edge points, and the point 1004 is also referred to as a span point.

The drawing points and corresponding points in the polygon span are calculated by sequentially changing the drawing target pixel (hereinafter referred to as the drawing pixel), and when the entire polygon span is drawn, the processing concerning the polygon span is ended. . Here, it is assumed that the drawing pixel is changed from left to right and that the screen coordinate value x is added one by one.

At this time, the screen coordinates of the span point 1004 and the texture coordinates of the corresponding point corresponding to the span point 1004 are set to (x1, y1, u1, v1), and the span point 1006
Let (x2, y2, u2, v2) be the texture coordinates of the screen coordinates of and the corresponding point corresponding to the span point 1006,
These can be expressed by the following equation (Equation 6).

[0018]

(Equation 6) (x2, y2, u2, v2) = (x1 + 1, y1, u1 + ∂u / ∂x, v1 + ∂v / ∂x) Similarly, (Equation 7) is the corresponding point of the texture mapping regarding the polygon span. In the general formula of calculation, the subscript represents the number of the drawing pixel.

[0019]

(Xn + 1, yn + 1, un + 1, vn + 1) = (xn + 1, yn, un + ∂u / ∂x, vn + ∂v / ∂x) ((Equation 1), ( From equation 2), u and v can be partially differentiated by x) By performing the processing of equation 7 from the start point to the end point of the polygon span, the coordinate value of the drawing point (screen coordinate value) and the coordinate of the corresponding point with respect to the polygon span The value (texture coordinate value) can be calculated.

Here, the parameters (DDA parameters) required to realize the processing by the incremental algorithm
Is ∂u / ∂x, ∂v / ∂x, respectively

[0021]

[Equation 8] ∂u / ∂x =-{(6-0) (16-0)-(9-6) (0-0)} / {(0-8) (9-6)-(11- 0) (6-0)} = 96/90 ∂v / ∂x =-{(6-0) (16-16)-(9-6) (16-0)} / {(0-8) ( 9-6)-(11-0) (6-0)} = -48/90.

Calculation of screen coordinates and texture coordinates in the polygon span shown in FIG. 37 can be realized as follows. Here, the span including the span point 1004 is calculated.

(5.33, 2.00, 0.00, 5.33) ……… (initial value) ↓ + (1, 0, 96/90, -48 / 90) (6.33, 2.00, 1.07, 4.80) ↓ + (1, 0 , 96/90, -48 / 90) (7.33, 2.00, 2.13, 4.27) ↓ + (1, 0, 96/90, -48 / 90) (8.33, 2.00, 3.20, 3.73) ……… (End) The calculation of the span point is continued until the edge on the right side of the figure is crossed.

With the incremental algorithm described above, accurate corresponding point calculation in the texture mapping process can be performed at high speed. Calculated from this (u, v)
Texture mapping can be realized by referring to the texture image with a value and reflecting it in the drawing pixel. With respect to the drawing point (x, y) that is normally calculated, a point that is rounded down to the right of the decimal point is set as a drawing pixel. Here, the pixel 1003 is a pixel value that refers to the texture image with the texture coordinate value (u0, v0). The texture coordinate value (u
1, v1), the pixel value that refers to the texture image
In 07, the pixel value that refers to the texture image by the texture coordinate value (u2, v2) is reflected.

[0025]

In the corresponding point calculation by the incremental algorithm, the calculation is performed based on the calculation result of the previous step, so that a high speed calculation can be realized with a simple structure. However, there is a problem that an error occurs in calculation of corresponding points.

Here, a case where the corresponding points for the drawing pixels are not accurately calculated will be described. FIG. 39 shows FIG.
Similarly, the calculation of corresponding points by the incremental algorithm will be described, and the same reference numerals are attached to indicate the same.

The polygon span point 1006 (x2, y2, u2, v0) calculated from the polygon edge point 1002 (x0, y0, u0, v0) in FIG. 39 by the incremental algorithm.
Texture coordinate (u2, v2) of 2) becomes (Equation 9) (from (Equation 1), (Equation 2), (Equation 6)), and the drawing pixel 1007
Is drawn with a value obtained based on the texture coordinates of (Equation 9).

[0028]

(U2, v2) = (u1 + ∂u / ∂x, v1 + ∂v / ∂x) = (u0 + du / dy + ∂u / ∂x, v0 + dv / dy + ∂v / ∂x) ) = (u0 + (a × dx / dy + b) + a, v0 + (d × dx / dy + e) + d) = (u0 + a × (1 + dx / dy) + b, v0 + d × ( 1 + dx / dy) + e) However, regarding the texture coordinates, the texture coordinate of the drawing pixel 1007 with respect to the drawing pixel 1003 is (Equation 10).
Becomes The coordinates of the drawing pixel 1003 are polygon edge points 1002 (x0, y0, u0, v0). A position obtained by shifting the edge point 1002 by one pixel in parallel with the y-axis is a point 1111 (x ', y', u ', v'), and this point is a coordinate in the drawing pixel 1007.

[0029]

(U2 ', v2') = (u0 + ∂u / ∂y, v0 + ∂v / ∂y) = (u0 + b, v0 + e) where the true value and the increment algorithm are used for the corresponding point u. When E is the difference from the obtained value, Eq.

[0030]

[Equation 11] E = u2′−u2 = −a × (1 + dx / dy) Therefore, the range of E is (Equation 12).

[0031]

[Equation 12] -a <E <a (∵ From FIG. 38, -2 <dx / dy <0) Next, the error of (Equation 12) will be described in terms of the difference between the sampling point and the drawing point. Here, the sample points are points that have the same positional relationship in the area of each pixel. For example, the center point of the pixel area can be defined as a sample point. That is, if the pixels are regularly arranged in a grid pattern, the sample points are also arranged in a grid pattern. On the other hand, since the drawing points are determined by the positional relationship between the drawing target polygons (particularly polygon edges) and the pixels, the drawing points are not always arranged in a regular grid.

In FIG. 39, the sample point in each pixel is set as the point closest to the origin in the pixel (upper left of the pixel shown in the figure), and a circle is marked. In addition, drawing points (edge points 1002, 10 calculated by the incremental algorithm,
04 and the span point 1006) and the distances 1108, 1109, 111 from the pixel sampling points in the pixels 1003, 1005, 1007 where these drawing points are reflected.
Let 0 be L1, L2, and L3, respectively. Here (Equation 7)
Therefore, L2 and L3 have the same value.

At this time, with respect to the corresponding point u, the errors E1, E2, E3 between the value at the sample point and the value at the drawing point calculated by the increment algorithm can be expressed by (Equation 13).

[0034]

[Equation 13] E1 = L1 × ∂u / ∂x E2 = L2 × ∂u / ∂x E3 = L3 × ∂u / ∂x In addition, if the error is E '(Equation 1
4).

[0035]

[Equation 14] E '<∂u / ∂x (the distance between the ∵ sample point and the drawing point is less than 1) <a In this way, the corresponding point calculation in the incremental algorithm shown in It can be said that the error is caused by the difference between the sample point of the pixel and the drawing point calculated by the increment algorithm. FIG. 40 shows an example of image quality deterioration due to this calculation error. FIG. 40 is a diagram showing that a texture mapped to a polygon due to a calculation error is discontinuous, and a texture image to be mapped to the polygon and an image when mapped by a conventional method,
And an image by ideal mapping.
It can be seen from FIG. 40 that image quality degradation occurs when mapping is performed by the conventional method.

The reason why the errors expressed by (Equation 12) and (Equation 14) are different is that (Equation 12) is an arbitrary point within a pixel as a sample point. Hereinafter, regarding the error of the corresponding point calculation, the sampling point is fixed. That is, the corresponding point error is (Equation 14).

Although the corresponding point u has been described here, the corresponding point v can be expressed as (Equation 2), and the illuminance (R,
G and B) and opacity A (Equation 15),
It can be expressed as (Equation 16). From these formulas,
It can be seen that the same error as the calculation error described here also occurs in the calculation of the corresponding point v, the illuminance, the opacity, and the like.

[0038]

R = Fr (x, y) = (ar × x + br × y + cr) G = Fg (x, y) = (ag × x + bg × y + cg) B = Fb (x, y) y) = (ab x x + bb x y + cb)

[0039]

## EQU16 ## A = Fa (x, y) = (aa × x + ba × y + ca) That is, an incremental algorithm was used for calculation of corresponding points, illuminance, opacity, etc. for displaying a polygon. In some cases, an error will occur.

Furthermore, the relationship between the polygon and the texture is not limited to (Equation 1) and (Equation 2), but occurs in corresponding point calculation by the incremental algorithm. The relational expression of polygon (x, y) and texture (u, v), illuminance (R, G, B), opacity A can be differentiated by y, partially differentiated by x. Can be done. For example, this correspondence point relationship is expressed by (Equation 17), (Equation 1)
8) can also be used.

[0041]

U = F (x, y) = (a × x + b × y + c) / (g × x + h × y + i)

[0042]

V = G (x, y) = (d × x + e × y + f) / (g × x + h × y + i) [coefficient a of (Equation 17), (Equation 18), b, c, d, e,
f is different from the coefficients of (Equation 1) and (Equation 2)] However, the relational expression of the polygon and the texture is (Equation 17),
Even with (Equation 18), a calculation error occurs as in the above description.

In other words, no matter what relational expression is used, the occurrence of calculation errors such as corresponding points, illuminance and opacity due to the incremental algorithm cannot be avoided.

As described above, the calculation of the illuminance, the opacity, the bump, the displacement, etc., including the corresponding point for the drawing pixel, cannot be accurately performed by the conventional incremental algorithm.

The error in the corresponding point calculation by the incremental algorithm is caused by the shift of 1 pixel at the maximum on the screen. If the size of the texture is about the same as that of the polygon or less than that, the result is (Equation 12),
The error represented by (Equation 14) becomes small. However,
When the texture is sufficiently enlarged, the error increases in proportion to the enlargement ratio, and when the texture mapping image is generated, the image quality is significantly deteriorated. However, in a device that maps an image captured in real time to an arbitrary polygon in real time, it is impossible to predetermine the enlargement ratio of the texture. Therefore, when an arbitrary texture is mapped to an arbitrary polygon, the deterioration in image quality, which is a problem here, cannot be avoided.

Therefore, paying attention to the error generated by the incremental algorithm, the error is effectively eliminated, and theoretically accurate corresponding point calculation is realized. Further, even in the realization, the configuration is extremely simple, and the high speed and the ease of configuration are not impaired. Furthermore, to realize the corresponding point calculation in mapping accurately and at high speed,
It is very important not only in the field of computer graphics, but also in multimedia devices such as home-use game machines and karaoke, and its applications are wide-ranging. In addition, the demand for real-time mapping processing is expected to increase in the future.

The error contained in the coordinates (u, v) calculated by the conventional DDA will be described below with reference to FIG.

In FIG. 6, solid lines represent coordinates in screen space, and broken lines represent coordinates in texture space.
The numbers circled indicate the coordinate values in the texture space.

It is also assumed that the screen pixel 603 is located at the screen coordinate (0,2) and the texture pixel 604 is located at the texture coordinate (2,3).

Reference numeral 605 indicates a polygon to be processed. The polygon 605 has coordinates (x, y, u, v)
= (7,0,0,0), (0,5,0,4), (3,
9, 4, 4) and (10, 4, 4, 0).

Further, reference numerals 601-a, 601-b,
601-c, 601-d and 601-e represent DDA in the edge direction, and reference numerals 602-a, 602-b,
602-c, 602-d, 602-e and 602-f
Is the DD in the span direction at the scan line of y = 3
Represents A.

The end points of the arrows representing these respective DDAs are points generated by the DDA in the span direction (this is called the DDA arrow).

In the texture mapping process, 601,
The pixels in the texture space indicated by the arrow of DDA 602 are reflected in the pixels in the screen space indicated by the arrows of DDA 601 and 602.

Specifically, the scan line at y = 3 has the following correspondence.

DDA Arrow Screen Pixel Texture Pixel 601-c (2,3) (0,2) 602-a (3,3) (0,2) 602-b (4,3) (0,1) 602 -C (5,3) (1,1) 602-d (6,3) (1,0) 602-e (7,3) (2,0) 602-f (8,3) (2,0 However, in the screen pixel (6, 3), the texture pixels to be reflected are (1, 0), (1, 1), (2, 1), depending on at which point in the pixel the texture is referred to.
It becomes (2, 0).

Which of these four texture pixels is DD
What happens in A depends on at which point of the screen pixel the texture is referenced.

That is, the image generated by the texture mapping differs depending on which point of the screen pixel refers to the texture.

If the reference points to the texture in the screen pixels are scattered within the polygon, the quality of the generated image will be deteriorated. In particular, when there is this variation between spans, the image quality is greatly degraded.

The above phenomenon will be described using mathematical expressions.

In the texture mapping, the texture coordinates (u, v) and the polygon coordinates (x, y) are expressed as in (Equation 19) and (Equation 20).

[0061]

U = f (x, y)

[0062]

[Expression 20] v = g (x, y) Here, although description will be made regarding (Expression 19), the same applies to (Expression 20).

At this time, the polygon coordinates that refer to the texture are (x0, y0). Also, polygon coordinates (x
When the screen coordinates obtained by quantizing (0, y0) are (X, Y),

[0064]

X 0 = X + ex

[0065]

It can be expressed as y0 = Y + ey.

That is, the coordinate value u of the texture is (equation 2)
3). It can be seen that the texture coordinate values vary due to the quantization error.

[0067]

U = f (X + ex, Y + ey) As described above, in the conventional method, the position of the screen pixel that refers to the texture has not been determined, so that the deterioration of the image is significant.

The present invention has been made in view of the above problems, and an object of the present invention is to provide a rendering device and a mapping device capable of generating a high quality image at high speed in a rendering process of computer graphics. It is in. Particularly, in the mapping process, it is to realize the corresponding point calculation in the mapping process at high speed and accurately with a simple configuration. In addition, for mapping processing, including texture mapping, illuminance,
It includes mapping processing for opacity, bumps, displacement, etc.

[0069]

A rendering device of the present invention comprises means for determining a pixel including a drawing point on a polygon projected on a plane having a plurality of pixels, and a fixed position for the pixel. And means for performing a rendering calculation based on the sample points having and outputting the calculation result.

Another rendering apparatus of the present invention corresponds to a means for performing rendering calculation based on a drawing point of a polygon projected on a plane having a plurality of pixels and outputting the calculation result, and the drawing point. Means for determining a plurality of pixels, means for determining a weighting coefficient for each of the corresponding plurality of pixels based on the position of the drawing point,
Means for distributing the calculation result to each of the plurality of corresponding pixels based on the weighting coefficient determined for each of the plurality of corresponding pixels, thereby achieving the above object. It

The mapping apparatus of the present invention generates a parameter corresponding to a polygon based on the positions of the vertices of the polygon projected on a plane having a plurality of pixels and the corresponding points given to the vertices of the polygon. A calculation unit, and an edge generation unit that determines a position of a first drawing point on the edge of the polygon and a position of a first corresponding point corresponding to the first drawing point based on the parameter, A drawing pixel corresponding to the first drawing point is determined, a sample point corresponding to the drawing pixel is determined, and the first pixel is determined based on the sample point.
The position of the second drawing point inside the polygon is determined based on the correction unit that corrects the position of the corresponding point, and the parameter and the sample point, and the second drawing point corresponding to the second drawing point is determined. And a span generator that determines the position of the corresponding point, thereby achieving the above object.

The correction unit sets 1 for the first drawing point.
The dimension may be corrected.

The edge generation unit defines the point where the edge of the polygon intersects the upper edge of the drawing pixel as the first drawing point, and the correcting unit selects the point closest to the origin within the drawing pixel. It may be a sample point.

The bit precision expressing the parameter may be set to a predetermined value, and the correction section may include a shifter and an adder.

Another mapping apparatus of the present invention determines a parameter corresponding to a polygon based on the position of the vertex of the polygon projected on a plane having a plurality of pixels and the corresponding point given to the polygon vertex. A calculation unit for generating the edge, an edge generation unit for determining the position of the first drawing point on the edge of the polygon and the first corresponding point corresponding to the first drawing point based on the parameter, The position of the second drawing point, which is a point inside the polygon, is determined based on the parameter and the first drawing point, and the second drawing point corresponding to the second drawing point is determined.
Of the corresponding drawing points, a generated image storage section that stores values corresponding to each of the plurality of pixels of the plane, the position of the first drawing point, or the second drawing And a pixel storage processing unit that stores a pixel value generated based on the position of a point as a value corresponding to one pixel or a plurality of adjacent pixels in the generated image storage unit. Is achieved.

The pixel storage processing unit has one pixel on the plane or a plurality of adjacent pixels of the area of the region on the plane defined by the position of the first drawing point or the second drawing point. The area ratio calculation unit that distributes the generated pixel value to one pixel or a plurality of adjacent pixels on the plane based on the ratio of the area occupied by the pixel to the area, and each of the distributed values, Generated image storage unit 1
It may be provided with a pixel value distribution unit that stores a value corresponding to one pixel or a plurality of adjacent pixels.

When the position of the first drawing point or the second drawing point is (x, y), the area ratio calculating unit
A first multiplier for multiplying (1-x) and (1-y);
a second multiplier for multiplying x by (1-y), and (1-
x) and y, and a third multiplier for multiplying x and y, and a fourth multiplier for multiplying x and y, and the pixel value distribution unit outputs the output from the first multiplier. A fifth multiplier for multiplying the generated pixel value, a sixth multiplier for multiplying the output from the second multiplier by the generated pixel value, and a third multiplier from the third multiplier. It may have a seventh multiplier that multiplies an output by the generated pixel value and an eighth multiplier that multiplies the output from the fourth multiplier by the generated pixel value. Good.

The bit precision expressing the position (x, y) of the first drawing point or the second drawing point or the generated pixel value is set to a predetermined value, and the first to eighth values are set. Each of the multipliers may include a shifter and an adder.

The edge generation unit defines the point at which the edge of the polygon intersects with the upper edge of the corresponding pixel as the first drawing point, whereby the second multiplier,
The fourth multiplier, the sixth multiplier, and the eighth multiplier may be omitted.

Another mapping apparatus of the present invention determines a parameter corresponding to a polygon based on the positions of the vertices of the polygon projected on a plane having a plurality of pixels and the corresponding points given to the vertices of the polygon. The position of the first candidate point corresponding to the first drawing point on the edge of the polygon and the position of the first corresponding point corresponding to the first candidate point on the basis of the calculation unit and the parameter. A first edge generating unit that determines the position of a second candidate point corresponding to the first drawing point on the edge of the polygon, and a second edge point corresponding to the second candidate point based on the parameter. A second edge generator that determines the position of two corresponding points; a selector that selects one of the output of the first edge generator and the output of the second edge generator; And the output selected by the selector And a span generation unit that determines the position of the second drawing point inside the polygon and determines the position of the third corresponding point corresponding to the second drawing point, thereby achieving the above object. To be done.

While either one of the first edge generating section and the second edge generating section is operating, the other edge generating section may not be operating.

The calculation unit calculates the inclination of the edge of the polygon on the plane, the selection unit cumulatively adds the decimal parts of the inclination, and determines whether the result of the cumulative addition exceeds a predetermined value. Accordingly, one of the output of the first edge generator and the output of the second edge generator may be selected.

According to another mapping apparatus of the present invention, based on the positions of the vertices of a polygon projected on a plane having a plurality of pixels and the corresponding points given to the vertices of the polygon, at least 2 corresponding to the polygon is obtained. A computing unit for generating a set of parameters, and one of the at least two sets of parameters
An increment switching unit for selecting a set of parameters, a position of a sample point corresponding to a first drawing point on the edge of the polygon, and the first drawing point based on the selected one set of parameters An edge generation unit that determines the position of the first corresponding point corresponding to, and the position of the second drawing point inside the polygon based on the selected set of parameters and the position of the sample point. A span generation unit that determines the position of the second corresponding point corresponding to the second drawing point,
This achieves the above object.

The calculation unit calculates the inclination of the edge of the polygon on the plane, and the increment switching unit cumulatively adds the fractional parts of the inclination, and whether or not the cumulative addition result exceeds a predetermined value. According to the above, one set of parameters may be selected from the at least two sets of parameters.

Another mapping apparatus of the present invention determines a parameter corresponding to a polygon based on the position of the vertex of the polygon projected on a plane having a plurality of pixels and the corresponding point given to the polygon vertex. Based on the calculation unit and the parameter, the position of the first corresponding point corresponding to the first drawing point on the edge of the polygon and the position of the sample point corresponding to the first drawing point are calculated. Based on the edge generation unit that determines, a correction unit that corrects the position of the sample point when the distance between the first drawing point and the sample point exceeds a predetermined value, and the parameter and the sample point. And a span generation unit that determines the position of the second drawing point inside the polygon and the position of the second corresponding point corresponding to the second drawing point. To be achieved.

The correction unit is included in the span generation unit, and the span generation unit corrects the position of the sample point, the position of the second drawing point, and the second corresponding point. Both of the processing for determining the position of s may be executed.

The calculation unit calculates the inclination of the edge of the polygon on the plane, the span generation unit cumulatively adds the fractional parts of the inclination, and whether or not the result of the cumulative addition exceeds a predetermined value. Depending on, it may be determined whether or not to correct the position of the sample point.

The polygon vertex has an attribute value for expressing the material of the polygon, and values corresponding to the first corresponding point and the second corresponding point are generated based on the attribute value. You may further have a means.

The polygon vertices have coordinate values of bumps or displacements, and may further include means for calculating bumps or displacements inside the polygons.

The mapping device may further have means for performing anti-aliasing processing.

Each of the plurality of pixels on the plane may have a plurality of corresponding subpixels, and the first drawing point may be determined corresponding to the subpixel positioning.

The rendering method of the present invention is based on a step of determining a pixel including a drawing point on a polygon projected on a plane having a plurality of pixels, and a sample point having a fixed position with respect to the pixel. And performing rendering calculation and outputting the calculation result, thereby achieving the above object.

Another rendering method of the present invention corresponds to a step of performing rendering calculation based on a drawing point of a polygon projected on a plane having a plurality of pixels and outputting the calculation result, and the drawing point. Determining a plurality of pixels, determining a weighting coefficient for each of the corresponding plurality of pixels based on the position of the drawing point, and determining for each of the corresponding plurality of pixels Allocating the calculation result to each of the corresponding plurality of pixels based on the calculated weighting coefficient, thereby achieving the above object.

The mapping method of the present invention generates a parameter corresponding to a polygon based on the positions of the vertices of the polygon projected on a plane having a plurality of pixels and the corresponding points given to the vertices of the polygon. The first step on the edge of the polygon based on the step and the parameter;
Determining the position of the drawing point and the position of the first corresponding point corresponding to the first drawing point, determining the drawing pixel corresponding to the first drawing point, and further corresponding to the drawing pixel Determining a sample point to be corrected and correcting the position of the first corresponding point based on the sample point; and determining the position of the second drawing point inside the polygon based on the parameter and the sample point. And determining the position of the second corresponding point corresponding to the second drawing point, whereby the above object is achieved.

According to another mapping method of the present invention, the parameter corresponding to the polygon is calculated based on the position of the vertex of the polygon projected on the plane having a plurality of pixels and the corresponding point given to the polygon vertex. A first step on the edge of the polygon based on the generating step and the parameter;
Determining the position of the drawing point and the first corresponding point corresponding to the first drawing point, and determining a second inside point of the polygon based on the parameter and the first drawing point. The position of the second drawing point and the position of the second corresponding point corresponding to the second drawing point, and the position of the first drawing point or the position of the second drawing point. A value generated based on one of the plurality of pixels on the plane, or
And a step of setting values corresponding to a plurality of adjacent pixels, thereby achieving the above object.

The operation will be described below.

According to the rendering device and the mapping device of the present invention, in order to correct the error contained in the pixel value generated based on the coordinate value of the drawing point on the polygon edge of the polygon projected on the screen, Generate a high-quality image by performing rendering processing based on the coordinate values of the sample points corresponding to the pixels including the drawing points or by calculating the corresponding points to the mapping image and generating the pixel values of each pixel. You can More specifically, it is as follows.

First, in the first configuration, the edge drawing portion calculates the edge drawing point in accordance with the y coordinate of the sample point of the pixel defined in advance. At this time, the defined sample points and the calculated edge drawing points include the calculation error shown in (Equation 14). For the calculation error of (Equation 14),
If the distance between the sample point and the edge drawing point on the x-axis is L,
With respect to the corresponding points, the processing of (calculation of texture coordinates (u, v)) (Equation 24) is performed by the correction unit. This makes it possible to accurately calculate the corresponding points of the texture.

[0099]

[Equation 24] u = u-L x ∂u / ∂x v = v-L x ∂v / ∂x Also, the same is applied to parameters other than texture coordinates (Equation 24) such as illuminance and opacity. be able to.

If the edge drawing point, which is the initial value of the span drawing point, matches the sample point of the pixel, all the span drawing points included in the span match the sample point of the pixel. That is, it is not necessary to perform the parameter correction process, and the corresponding points of the span drawing points can be calculated correctly.

As described above, by correcting the parameters for calculating the corresponding points at the edge drawing points, the corresponding points can be accurately calculated for all the pixels including the polygon.

L shown in the correction processing (Equation 24) is a value after the decimal point of the x coordinate value of the edge drawing point calculated by the edge generation unit. When the x coordinate value of the drawing point is x, (Equation 25) Can be given.

[0103]

L = x- [x] Further, since ∂u / ∂x and ∂v / ∂x are used in the span generation unit, there is no need to newly calculate them. Furthermore, (Equation 24)
Regarding the precision of the multiplication shown in, it is not necessary to have a precision higher than the precision below the decimal point of the x-coordinate value of the edge point calculated by the edge generation unit, and the precision has several bits more than the resolution of the generated pixel. It's enough. in this way,
The configuration of the correction unit is relatively simple, and the advantage of simplification of the configuration by the incremental algorithm remains for the entire apparatus.

Regarding the processing speed, the processing speed of the incremental algorithm is not impaired. The processing in the interpolation unit shown here is one multiplication and one addition (Equation 24),
The delay due to the correction process is extremely small. Further, the edge generation unit, the interpolation unit, the span generation unit and the like can be operated in a pipeline manner in the entire apparatus, and the high speed performance of the entire processing is not impaired at all.

Further, in the anti-aliasing processing for the edge, it is necessary to sample the point on the screen coordinates in the area where the polygon does not exist. in this case,
It is possible to calculate the corresponding point from the parameter of the sample point where the polygon exists by using (Formula 24), where L is the distance from the sample point where the polygon exists and is closest to the x-axis direction. .

As described above, the texture mapping apparatus of the present invention realizes accuracy of calculation of corresponding points of texture mapping, simplification of the configuration, and high speed processing.

In the second configuration, the sampling point in the first configuration is set to, for example, the point closest to the origin inside the pixel, thereby reducing the amount of calculation required for the correction process and simplifying the circuit configuration. A mapping device can be realized with.

In the first and second configurations, the error in the corresponding point calculation is eliminated by correcting the calculation parameter of the corresponding point to be calculated as a result, but the third configuration is the memory for storing the generated image. The pixel value obtained from the texture image is distributed to (frame memory) according to the area covered by the texture pixel and stored.
In the incremental algorithm, the deviation between the calculated drawing point and the pixel sampling point is a factor of image quality deterioration. When the texture pixel referred to by the calculated drawing point is written in the frame memory, ideally, the texture pixel affects a maximum of 4 pixels in the frame memory. With respect to the four pixels of the affected frame memory, the pixel value according to the area ratio shared by the texture pixel and the frame pixel is reflected in each frame pixel. As a result, the same effect as that described above can be obtained. That is, high-quality images can be obtained by various mappings.

In the third structure, it is not necessary to perform the parameter correction process. On the other hand, the access amount to the frame memory increases. However, the distribution storage in the frame memory is local, and an increase in access amount can be hidden by using a small amount of intermediate buffer with the frame memory.

In the fourth structure, the two edge generators calculate the data relating to the polygon edge, and the selector appropriately selects these data.
It is possible to ideally calculate screen coordinate values and texture coordinate values regarding polygon edges.

In this configuration, it is necessary to provide two edge generating units, but the correction processing can be realized in the data selection. Since this selector can be realized by a selector, it can be executed faster than the correction process. Also, the circuit of this selection unit can be realized very easily.

Furthermore, it is possible to stop the operation of either of the two edge generation sections by the control of this selection section. As a result, the power consumption of the entire device can be suppressed.

In the fifth structure, the two edge generators provided in the fourth structure are unified.
This is realized by appropriately selecting the increment value of the increment processing in the edge generation unit. The circuit can be realized with a very simple structure.

Further, in the sixth configuration, the increment value calculated by the calculation section and used by the edge generation section is set to a value that simplifies the correction processing in the correction section. That is, the edge generation unit performs normal increment processing. The correction processing of the data generated here can be performed very easily in the correction unit.

As a result, the correction unit can be realized with a very simple structure. Also, regarding the increment value calculated by the calculation unit, the calculation amount is not much different from the conventional one. Therefore, it becomes possible to calculate the ideal screen coordinate value of the polygon and the ideal texture coordinate corresponding thereto at high speed with a simple configuration.

In the seventh configuration, the processing in the correction unit and the span generation unit in the sixth configuration is realized by the same circuit. This is achieved by using the same circuit, and does not slow down the processing speed. That is, it is possible to realize the same function with a simpler configuration than the sixth configuration.

The mapping apparatus of the present invention pays attention to the fact that the relative positional relationship between the drawing point calculated by the incremental algorithm and each pixel corresponding thereto is not always constant, and the positional information of the sample point representing each pixel is calculated. By using each pixel value to determine, the image quality is improved.

Further, in another mapping device according to the present invention, each pixel value determined by the position information of each drawing point calculated by the incremental algorithm is converted into a relative positional relationship between each drawing point and each pixel value. Based on this, by allocating to a plurality of adjacent pixels, the same effect as that of the above mapping device is obtained.

Further, the problem of error due to the incremental algorithm not only affects the mapping process, but is also a problem common to general rendering algorithms such as shading. Therefore, the present invention is widely applicable to rendering processing in general.

The deterioration of the image is caused because the position of the screen pixel which refers to the texture is not fixed. That is, it is possible to prevent deterioration of image quality by making the position in the screen pixel that refers to the texture constant in the screen coordinate space.

The relational expression in texture mapping is
It is expressed as in (Equation 19) and (Equation 20), but here, description will be made using (Equation 26).

[0122]

[Mathematical formula-see original document] u = f (x) = Axx + B This is for the purpose of simplifying the explanation, but the same applies in principle to (Formula 19) and (Formula 20).

The quantized value of x is X, and the quantization error is ex.
Then, the value x is represented by (Equation 27).

[0124]

X = X + ex In addition, letting E be the error between u calculated with a value before quantization and u calculated with a value after quantization, the error E is expressed by (Equation 28). To be done.

[0125]

E = f (x) −f (X) = f (X + ex) −f (X) = f (ex) That is, if the quantization error is eliminated in the x space,
An error is not included in the value of u before and after the quantization.

Further, by making the quantization error constant, E becomes constant as a result. This is because the calculated u will have an error uniformly and there will be no relative error in the u space.

As a result, in all the screen space pixels, the coordinate value (x, y) for calculating the texture is made constant within the pixel, so that the texture coordinate to be calculated has no error in the texture space. it can.

Therefore, the coordinates of the polygon and the texture can be calculated without any error, and thus the image quality deterioration of the image generated by the texture mapping can be prevented.

[0129]

Although the principle of the mapping apparatus of the present invention can be widely applied to rendering processing in general, the mapping apparatus will be described in each of the following embodiments for simplification of description. When the present invention is applied to a rendering process other than mapping, instead of performing the rendering operation using the coordinate values of each drawing point, the coordinate value of the sample point corresponding to each drawing point and the attribute value given to each polygon. The desired rendering operation may be performed using The attribute value is a value for expressing the material of the polygon, and is a value indicating the reflectance, the transmittance, the refractive index, etc. of the polygon.

(First Embodiment) A first embodiment of the mapping apparatus of the present invention will be described below with reference to the drawings.

FIG. 1 shows an example of the configuration of the mapping device of the present invention. The mapping device includes a main processor 1, a main memory 2, a rendering processor 3, a pixel memory 4, and a DAC (digital-analog converter).
5, a monitor 6, and a system bus 7.

The main processor 1 performs geometrical operations for three-dimensional graphic processing for texture mapping, controls the rendering processor 3 and the DAC 5, and controls the entire system.

The main memory 2 stores a program executed by the main processor 1 and data necessary for each processing such as polygon data.

Under the control of the main processor 1, the rendering processor 3 draws polygons on the pixel memory 4.

The pixel memory 4 stores the image drawn by the rendering processor 3 and the texture to be mapped.

The system bus 7 interconnects an input / output device (not shown) with the main processor 1, the main memory 2, the rendering processor 3, and the pixel memory 4.

The generated image is output to the monitor 6 via the DAC 5.

FIG. 2 shows a configuration example of the rendering processor 3. The rendering processor 3 includes a DDA coefficient calculation unit 201, a mapping address generation unit 208, and an original image pixel processing unit 207.

The DDA coefficient calculation unit 201 uses the DDA used in the mapping address generation unit 208 from the relationship between polygons and texture vertices input via the system bus 7.
Generate the coefficients. The DDA coefficient includes a differential value of y (vertical DDA) and a partial differential value of x (horizontal DDA).

The mapping address generator 208 uses the DD
The rendering processing is performed using A, and the edge generation unit 202, the correction unit 203, and the span generation unit 204
And

As shown in FIG. 3, the edge generator 202 includes an x DDA processor 301 and a y DDA processor 3.
02, u DDA processing unit 303, and v DDA processing unit 304.

The DDA processing units 301 to 304 calculate the coordinate points (x, y, u, v) of the polygon edge by using the differential value of y calculated by the DDA coefficient calculation unit 201. Specifically, the process shown in (Equation 29) is performed.

In (Equation 29), (x (n + 1), y
(N + 1), u (n + 1), v (n + 1)) is (x
(N), y (n), u (n), v (n)) shows the coordinates at the polygon edge on the scan line immediately below. Also, dx / dy and du / d shown in (Equation 29)
The values of v and dv / dy are calculated by the DDA coefficient calculator 201 and given to the edge generator 202.

[0144]

(X (n + 1), y (n + 1), u (n + 1), v (n + 1)) = (x (n), y (n), u (n), v (n)) + (dx / Dy, 1, du / dy, dv / dy) Further, as shown in FIG. 4, the correction unit 203 includes an x correction unit 411, a y correction unit 412, and a u correction unit 413.
v correction unit 414 is included. Correction units 411 to 41
Each of 4 performs the processing (correction processing) shown in (Equation 32). Further, the DDA coefficient calculation unit 201 calculates du / dy, ∂u / ∂x, etc., which are necessary for the correction.

Further, as shown in FIG. 5, the span generation section 204 includes an x DDA processing section 511, a u DDA processing section 512, and a v DDA processing section 513.
In span processing, y value does not change, so DD of y
A processing unit is not required.

The DDA processing units 511 to 513 calculate the coordinate points (x, y, u, v) of the polygon edge by using the differential value of y calculated by the DDA coefficient calculation unit 201. Specifically, the process shown in (Equation 30) is performed.

In (Equation 30), (x (n + 1), y
(N + 1), u (n + 1), v (n + 1)) is (x
The coordinates at a point of 1 pixel in the span direction with respect to (n), y (n), u (n), and v (n) are shown.

[0148]

(X (n + 1), y (n + 1), u (n + 1), v (n + 1)) = (x (n), y (n), u (n), v (n)) + (1 , 0, ∂u / ∂x, ∂v / ∂x) Further, the pixel memory 4 includes a generated image storage unit 205,
The original image storage unit 206 is included. Generated image storage unit 2
05 stores the image drawn by the rendering processor 3. The original image storage unit 206 stores the original image (texture) to be drawn by the rendering processor 3.

FIG. 7 is a block diagram showing an address generator in the mapping device according to the first embodiment of the present invention. The calculation unit 101 uses the positions of the polygon vertices and the corresponding points given to the polygon vertices to calculate the parameters used for the drawing point calculation and the corresponding point calculation inside the polygon by the incremental algorithm. The edge generation unit 102a uses the parameters calculated by the calculation unit 101 to perform a calculation for determining the positions of the drawing point and the corresponding point on the polygon edge. The correction unit 103a uses the parameters calculated by the calculation unit 101 and the drawing points determined by the edge generation unit 102a to perform correction processing on the corresponding points (texture image) generated by the edge generation unit 102a. The span generation unit 104 uses the parameters calculated by the calculation unit 101 and the data generated by the correction unit 103a to calculate drawing points of the polygon span and corresponding points, and outputs the calculation results.

The operation of the address generator of the mapping device configured as described above will be described below with reference to the drawings.

FIG. 8 is an enlarged view of the pixels that the side 1404 in FIG. 37 crosses. The pixels in FIG. 8 are (x, y) = (2,4) (2,5) (3, 5) It corresponds to the part surrounded by (3, 4). 201 indicates a pixel on the screen, 202 indicates an edge of the polygon, and the polygon exists on the right side of the edge. Reference numeral 203 denotes an edge drawing point (x0, y0) before correction processing which is generated by the edge generating unit 102a. Reference numeral 204 is a sample point (X, Y) of the pixel 201. Reference numeral 205 denotes a point (x ′, for explaining the correction processing in the correction unit 103a.
y '), located on the polygon edge 202, and y' = Y
And Further, the point (u, v) on the texture image corresponding to the point (x, y) on the polygon is (Equation 1), (Equation 2)
Assuming that the relationship shown in is established, the corresponding point of the texture image at the edge drawing point 203 is (u0, v0), and the sample point 2
The corresponding point of the texture image in 04 is (U, V), point 2
The corresponding point of the texture image in 05 is (u ′, v ′).

At this time, these corresponding points are (equation 3)
1), the relationship shown in (Equation 32) holds ((Equation 4),
(See (Equation 7)).

[0153]

(U ′, v ′) = (u0 + du / dy × (Y−y0),
v0 + dv / dy × (Y-y0))

[0154]

(32) (U, V) = (u '+ ∂u / ∂x × (X-x'),
v '+ ∂v / ∂x × (X-x'))

[0155]

Since x ′ = x0 + dx / dy, (u0, v0) and (U, V) are given by (Equation 3
The relationship of 4) is established.

[0156]

(34) (U, V) = (u0 + du / dy × (Y-y0) + ∂u
/ ∂x × (X-x0-dx / dy), v0 + dv / dy × (Y-y
0) + ∂v / ∂x × (X-x0-dx / dy)) Specifically, the sample point is the center point of the pixel and (X, Y) =
(2.5,4.5), if the y coordinate of the drawing point to be calculated = 4.20, then (x0, y0, u0, v0) = (2.40, 4.20, 0,00, 11.20), and (Equation 34) and (Equation 5) , (Equation 8), (U, V) = (1.53, 11.24). Here, the target polygon is the same as that described in the related art, and dx / dy, du / dy,
The same values can be used for dv / dy, ∂u / ∂x, and ∂v / ∂x. By calculating the pixel value of the texture with this (U, V) value, it becomes possible to reflect an accurate pixel value on the screen pixel. That is, the edge drawing point 20 generated by the edge generation unit 102a
By correcting the corresponding point (u0, v0) of 3 (x0, y0) to the corresponding point (U, V) of the sampling point 204 (X, Y) by the correction unit 103a (see (Equation 34)), An ideal corresponding point of 201 can be calculated. Furthermore, by ideally calculating the corresponding points of the first pixel of the span, the corresponding points of the pixels included in this span can also be ideally calculated (see (Equation 7)).

Here, a configuration example of the correction unit 103a will be described. Since the correction process for the texture coordinate v is similar to the correction process for u, the correction unit 103 in FIG.
Only the configuration for the texture coordinate u in a is shown. In (Equation 34), (Y-y0) and (X-x0) do not exceed 1, and are defined by the pixel sampling points. That is, here, Y-y0 = fractional part of Y-fractional part of y0 = 0.5-fractional part of y0 X-x0 = fractional part of X-fractional part of x0 = 0.5-fractional part of x0 . In FIG. 9, a multiplier 1601
Is a first multiplier that calculates the second term of (Equation 34) representing U. The adder 1602 is a first adder that calculates (X−x0 + dx / dy) of the third term of (Expression 34) representing U. The multiplier 1603 is a second multiplier that calculates the third term of (Equation 34) representing U. The adder 1604 is U
The first term, the second term and the third term of (Equation 34)
It is a second adder for calculating. Further, the corresponding point v can also be realized with a similar configuration. With this configuration, it is possible to realize the correction process according to (Equation 34), that is, the correction unit 103a according to the present embodiment.
Can be realized.

As described above, according to this embodiment, by providing the correction unit 103a, the edge generation unit 102a
Corresponding points corresponding to the edge drawing points generated in step 3 are corrected. Accordingly, it is possible to calculate an ideal corresponding point for the pixel including the edge drawing point. Further, if the head of the span drawing point, that is, the edge drawing point is an ideal corresponding point, the ideal corresponding point can be calculated for the pixels included in the span. That is, the ideal corresponding points can be calculated for all the pixels included in the polygon, and the image quality of the mapping image can be improved.

Furthermore, the variable (Equation 34) used in the correction processing performed by the correction unit 103a is the same as the variable used in the conventional incremental algorithm, and it is not necessary to newly generate it. Further, although a delay occurs in the entire processing due to the correction processing, all the constituent elements shown in FIG. 7 can be operated in a pipeline manner, and the high speed of the entire processing is not impaired at all.

In this embodiment, the relationship between the polygon and the texture image is set to (Equation 1) and (Equation 2), but the relationship is not limited to (Equation 1) and (Equation 2). For example, even if this relationship is set to (Equation 17) and (Equation 18), accurate corresponding point calculation can be performed.

In other words, this embodiment can correspond to all corresponding point expressions.

In the present embodiment, the calculation of the corresponding points on the image to be mapped on the polygon has been described, but the same processing can be applied to the illuminance calculation and the opacity calculation. Also, regarding these, similarly to the corresponding points, it is possible to correspond to all relational expressions.

In this embodiment, the method in which the anti-aliasing process is not performed has been described, but the same process can be performed even if the anti-aliasing process is included.

Even when sub-pixel positioning is performed, the same correction processing as in this embodiment can be performed.

Although the address generator for determining the position (address) of the corresponding point on the texture image has been described with reference to FIG. 7, if this is used for texture mapping, the configuration of FIG. 10 is obtained. The address generation unit 1701 generates the address (x, y) of the polygon on the screen and the address (u, v) of the texture corresponding to the input of the correspondence between the polygon and the texture by this device. A texture image is stored in the texture memory 1702. The frame memory 1703 stores the image to be generated. The address generation unit 1701 draws polygon coordinates and the corresponding texture coordinates (u, v).
Is output. The pixel value read from the texture memory 1702 by the output texture coordinates (u, v)
(R, G, B) is stored in the frame memory 1703. This storage address is determined from (x, y) output from the address generator 1701.

Illuminance mapping (shading)
If it is used for, the configuration shown in FIG. 11 is obtained. Illuminance calculator 1
801 is the address (x, y) of the polygon on the screen with the correspondence between the polygon and the illuminance of the polygon as input.
And the pixel value that represents the color in the corresponding polygon
Generate (R, G, B) and. The frame memory 1803 is a frame memory that stores an image to be generated. The illuminance calculation unit 1801 outputs the drawing coordinates of the polygon and the pixel values (R, G, B) corresponding thereto. Output pixel value (R,
G, B) are stored in the frame memory 1803. This storage address is output from the illuminance calculation unit 1801 (x, y)
Is determined from

Further, mapping other than those described here such as opacity, bump, and displacement mapping can also be realized by the same configuration.

The same applies to the following embodiments, and the present invention can be applied to mapping of texture, illuminance, opacity, bumps, displacement and the like.

(Second Embodiment) A second embodiment of the mapping apparatus of the present invention will be described below with reference to the drawings. The configuration is the same as that of the first embodiment. In the present embodiment, a point closest to the origin in a pixel is defined as a sample point, and an edge drawing point and its corresponding point are calculated in accordance with the y axis of the sample point, so that the correction process in the correction unit is performed in one dimension. The correction can be simplified as compared with the first embodiment.

FIG. 12 is an enlarged view of the pixel which the side 1404 in FIG. 37 crosses. The pixel of FIG.
It corresponds to the part surrounded by (x, y) = (2,4) (2,5) (3,5) (3,4) in 7. 401 indicates one pixel on the screen,
Reference numeral 402 denotes an edge of the polygon. The polygon to be drawn exists on the right side of the polygon edge. Reference numeral 403 is an edge drawing point (x0, y0) before the correction processing generated by the edge generation unit 102a.

Reference numeral 404 is the sample point (X, Y) of the pixel 401.
It is. Further, it is assumed that the relationship shown in (Equation 1) and (Equation 2) is established between the point (x, y) on the polygon and the point (u, v) on the texture image. The corresponding point is (u0, v0), and the corresponding point of the texture image at the sample point 404 is (U, V).

At this time, these corresponding points are (equation 3)
1), the relationship shown in (Equation 32) holds ((Equation 4),
(See (Equation 7)).

Here, the sample point of the screen pixel is the point closest to the origin in that pixel (see FIG.
The upper left point of the pixel in 2) and the edge generation unit 102a calculate the edge drawing point and its corresponding point in accordance with the y-axis of the sample point of the screen pixel defined in advance. As a result, edge points as shown in FIG. 12 can be generated.

At this time, these corresponding points are (equation 3)
The relationship shown in 5) is established (see (Equation 4) and (Equation 7)).

[0175]

(35) (U, V) = (u0-∂u / ∂x × (x0-X), v0-
∂v / ∂x × (x0-X)) Since the sample point 404 is the point closest to the origin in the pixel 401,

[0176]

(X, Y) = ([x0], y0), and (Equation 35) can be (Equation 37).

[0177]

(U, V) = (u0-∂u / ∂x × (x0- [x0]), v
0-∂v / ∂x × (x0- [x0])) Specifically, the sample point is located at the upper left of the pixel (X, Y) = (2.0,
4.0), when the y coordinate of the drawing point to be calculated is adjusted to the sample point, (x0, y0, u0, v0) = (2.67, 4.00, 0.00, 10.67), and (Equation 34), (Equation 5), (Equation 5) From (8), (U, V) = (-0.71, 11.03). Here, the calculated U has a negative value. This can be said to be an ideal correspondence point calculation in which the sample points are outside the polygon and the sample points are negative.

Here, the target polygon is the same as that described in the prior art, and dx / dy, du / d.
The same values can be used for y, dv / dy, ∂u / ∂x, and ∂v / ∂x. By calculating the pixel value on the texture image using this (U, V) value, it becomes possible to reflect the accurate pixel value for the pixel.

That is, the corresponding point (u0, v0) of the edge drawing point 403 (x0, y0) generated by the edge generating unit 102a is
Corresponding point of sample point 404 (X, Y) in the correction unit 103a
By correcting to (U, V) (see (Equation 37)), an ideal corresponding point of the pixel 401 can be calculated. Furthermore, by ideally calculating the corresponding points of the first pixel of the span, the corresponding points of the pixels included in this span can also be ideally calculated (see (Equation 7)).

Next, the correction unit 103 according to the present embodiment.
A configuration example of a will be described. Since the correction process regarding the texture coordinates u and v can be considered in the same manner, the correction process shown in FIG.
3 shows a configuration regarding the texture coordinate u in the correction unit 103a. In (Equation 37), (x0- [x0]) represents the decimal point part of x0. In FIG. 13, the multiplier 1901 calculates the second term of (Equation 37) representing U. The adder 1902 is an adder that calculates the U by adding the first term and the second term of (Equation 37). Also, the corresponding point V
Can be realized with a similar configuration. With this configuration, it is possible to realize the correction processing according to (Equation 37), that is, the correction unit 1 according to the present embodiment.
03a can be realized.

The corresponding point (u0, v0) of the edge drawing point 403 (x0, y0) generated by the edge generating section 102a is converted into the corresponding point (U, V) of the sample point 404 (X, Y) by the correcting section 103a. By correcting (Equation 37), the ideal corresponding point of the pixel 401 can be calculated. Furthermore, by ideally calculating the corresponding points of the first pixel of the span, the corresponding points of the pixels included in this span can also be ideally calculated (see (Equation 7)).

As described above, according to the present embodiment, the edge generation unit 102a generates the edge drawing point and the corresponding point in accordance with the y coordinate of the sample point of the pixel, and the correction unit 10a.
In 3a, the parameter relating to the polygon edge generated by the edge generating unit 102a is corrected. As a result, similar to the first embodiment, it is possible to calculate ideal corresponding points for all pixels included in the polygon, and it is possible to improve the image quality of the texture mapping image.

The correction process can be realized by one multiplication and one addition for one parameter from the configuration diagram of the correction unit 103a shown in FIG. 13, and (x0- [x
0]) can be realized with a high-order bit mask,
The circuit can be simplified as compared with the case of the first embodiment. That is, the delay of the entire process due to the correction process can be made extremely small. Further, in the entire apparatus, all the constituent elements can be operated in a pipeline, and the high speed of the entire processing is not impaired at all.

In the present embodiment, the sample point in the pixel is the point closest to the origin in the pixel, but the position of this sample point can be another position. For example, the point farthest from the origin may be used as the sample point.

In the present embodiment, the relationship between the points on the polygon and the points on the texture image is (Equation 1), (Equation 2)
However, the relationship is not limited to (Equation 1) and (Equation 2). For example, even if this relationship is set to (Equation 17) and (Equation 18), accurate corresponding point calculation can be performed. In other words, it is possible to deal with all corresponding point expressions.

In the present embodiment, the calculation of the corresponding points of the polygon has been described, but the same processing can be realized for the illuminance calculation and the opacity calculation. Also, regarding these, similarly to the corresponding points, it is possible to correspond to all relational expressions.

In the present embodiment, the method in which the anti-aliasing processing is not performed has been described, but the same processing can be performed even if the anti-aliasing processing is included.

Even if sub-pixel positioning is performed, the same correction processing as in this embodiment can be performed.

(Embodiment 3) Hereinafter, a third embodiment of the present invention will be described with reference to the drawings.

FIG. 14 is a block diagram of a mapping device according to the third embodiment of the present invention. The calculation unit 501 uses the positions of the polygon vertices and the corresponding points given to the polygon vertices to calculate the drawing points inside the polygon by the incremental algorithm and the parameters used to calculate the corresponding points. The edge generation unit 502 uses the data generated by the calculation unit 501 to calculate drawing points and corresponding points of polygon edges. The span generation unit 503 uses the data generated by the calculation unit 501 and the corresponding points generated by the edge generation unit 502 to calculate the drawing points of the polygon span and the corresponding points. The original image storage unit 504 stores the original image (mapping image). The generated image storage unit 505 stores the generated image. The pixel storage processing unit 506 is the span generation unit 5
One pixel data stored in the original image storage unit 504 is stored in one or more adjacent pixel data in the generated image storage unit 505 by using the data from 03.

The operation of the mapping device configured as described above will be described below with reference to the drawings.

FIG. 15 is an enlarged view of a pixel crossed by the side 1404 in FIG. 37. The pixel in FIG.
It corresponds to the part surrounded by (x, y) = (2,4) (2,6) (4,6) (4,4) in 7. Pixels 601a, 601b, 601c, 601
d is a pixel on the screen, and the coordinates of the point (upper left) closest to the origin in each pixel is (x, y) =
They are (2,4), (3,4), (2,5) and (3,5). The edge 602 indicates the edge (side 1404) of the polygon, and it is assumed that the polygon exists on the right side of the edge 602. The drawing point 603 is the drawing point (x0, y0) = (2.4
0, 4.20).

Here, the sample point of the screen pixel is the point closest to the origin in that pixel (see FIG.
5, the point 604a is the sample point (X, Y) = (2,4) of the pixel 601a, the point 604b is the sample point (X + 1, Y) = (3,4) of the pixel 601b, and the point 604c is The sample point (X, Y + 1) = (2,5) of the pixel 601c, and the point 604d is the pixel 601.
The sample point of d is (X + 1, Y + 1) = (3, 5). Pixel 60
Reference numeral 5 denotes a texture pixel corresponding to the drawing point 603, which represents a state of being projected on the screen. This texture pixel is read from the original image storage unit 504 and has the same size as the screen pixel.

When texture mapping is performed,
As described above, the corresponding point (u0, v0) of the drawing point 603 generated by the span generation unit 503 does not match the corresponding point of the pixel 601a.

Here, assuming that the corresponding points at the drawing points and the corresponding points at the pixel sample points do not match, the data of the pixels to be drawn on the screen are stored in a plurality of pixels on the screen, and the corresponding points are stored. Resolve the discrepancy problem.

That is, the texture pixel 605 represented by the corresponding point (u0, v0) of the drawing point 603 (x0, y0) is read from the original image storage unit 504. The pixel 605 is distributed to the pixels 601a, 601b, 601c, and 601d on the screen and then stored to solve the conventional problem. The distribution is performed according to the area shared by the determined texture pixel 605 and each screen pixel. That is, a weighting coefficient is calculated for each screen pixel according to the shared area, and a value obtained by multiplying the pixel value of the texture pixel 605 by each weighting coefficient is distributed to each screen pixel value. For example, the shared area ratio can be used as the weighting coefficient.

In FIG. 15, 606a represents a portion (shared area) shared by the texture pixel 605 and the screen pixel 601a, and its area (shared area) is Sa. Similarly, 606b represents a shared area with the screen pixel 601b, and its shared area is Sb. 606c
Represents a shared area with the screen pixel 601c, and the shared area is Sc. 606d is a screen pixel 6
A shared area with 01d is shown, and the shared area is Sd. These shared areas can be expressed as in (Equation 38) to (Equation 41).

[0198]

(38) Sa = (1- (x0-X)) x (1- (y0-Y))

[0199]

Sb = (x0-X) × (1- (y0-Y))

[0200]

(40) Sc = (1- (x0-X)) x (y0-Y)

[0201]

## EQU00004 ## Sd = (x0-X) × (y0-Y) Specifically, the shared area of each is shown below (see FIG. 16). FIG. 16 is an enlarged view of the four shared areas in FIG.

Sa = (1-2.4-2) * (1-4.2-4) = 0.6 * 0.8 = 0.48 Sb = (2.4-2) * (1-4.2-4) = 0.4 * 0.8 = 0.32 Sc = ( 1-2.4-2) * (4.2-4) = 0.6 * 0.2 = 0.12 Sd = (2.4-2) * (4.2-4) = 0.4 * 0.2 = 0.08 At this time, the pixel value of the texture pixel 605 is (R, G, B) = (1
92, 64, 32), pixel 601a, pixel 601b,
The pixel 601c and the pixel 601d have (R, G, B) = (9
2.16, 30.72, 15.36), (61.44, 20.48, 10.24), (23.
04, 7.68, 3.84) and (15.36, 5.12, 2.56) are stored.

These processes are performed by the image storage processing unit 506, the texture data (pixel value) of the original image storage unit 504 is read out, and the above-described processes are performed on the generated image storage unit 505. The original image storage unit 504 and the generated image storage unit 505 are realized by, for example, a memory, and the image storage processing unit 506 is realized by the configuration shown in FIG.

In FIG. 17, the multiplier 2101 calculates the shared area Sa, the multiplier 2102 calculates the shared area Sc, the multiplier 2103 calculates the shared area Sb, and the multiplier 2
104 calculates the shared area Sd. The multiplier 2105 calculates the pixel value to be stored in the pixel 601a, and the multiplier 2106
Calculates the pixel value to be stored in the pixel 601c, and the multiplier 21
07 calculates the pixel value stored in the pixel 601b, and the multiplier 2108 calculates the pixel value stored in the pixel 601d.
With such a configuration, the image storage processing unit 506 can be realized.

As described above, according to the present embodiment, by providing the image storage processing unit 506, the value of the texture pixel to be reflected on the screen pixel is adjusted according to the deviation between the drawing point and the sample point generated in the increment algorithm. , Distribute and store. That is, accurate pixel values in texture mapping can be reflected in all pixels on the screen, and accurate texture mapping can be realized on the screen. This can be said that the present embodiment corrects the drawing pixel, while the first and second embodiments correct the position of the drawing point.

The processing performed in the image storage processing unit 506 added in the configuration of the present embodiment is based on the calculation processing of the shared area of the texture pixel and the screen pixel ((Equation 38) to (Equation 41)) and the area thereof. The process of distributing the texture pixel value by the process and the process of storing the distributed texture pixel value can be divided into three.

The calculation processing of the shared area and the distribution of the pixel values performed by the multiplier of FIG. 17 can be realized by the shifter and the adder. In other words, the multiplier shown in FIG. 17 is not a so-called multiplier but a multiplier composed of shift and addition. Further, by suppressing the precision required in these processes, it is possible to suppress the number of adders and further suppress the delay due to the addition process. That is, these processes do not have a high load and do not lead to a decrease in processing speed. Further, regarding the processing of storing the distributed texture pixel values, the pixels to be stored are always adjacent to each other. Therefore, the access itself to the generated image storage unit 505 is operated in parallel, and the processing performed by the image storage processing unit 506 is performed. It is also possible to operate in parallel with other components. That is, the configuration of the image storage processing unit 506 is not complicated, and its processing speed does not deteriorate.

In this embodiment, the sample point in the pixel is the point closest to the origin in the pixel, but the position of the sample point is arbitrary.

In the present embodiment, the edge generator 5
Although the y-axis direction value of the drawing point generated in 02 and the y-axis direction value of the pixel sample point do not match, the edge generating unit 502 generates the drawing point that matches these values. The amount of access to the generated image storage unit 505 can be reduced.

In the present embodiment, the calculation of the corresponding points of the polygon has been described, but the same processing can be realized for the illuminance calculation and the opacity calculation. Also, regarding these, similarly to the corresponding points, it is possible to correspond to all relational expressions.

In the present embodiment, the method in which the anti-aliasing process is not performed has been described, but the same process can be performed including the anti-aliasing process.

Even if the sub-pixel positioning is performed, the same correction processing as in the present embodiment can be performed.

In this embodiment, the texture pixel has four pixels on the screen and a common area.
By generating drawing points according to the y-coordinate values of sample points, the maximum number of shared areas is two. As a result, it is possible to dramatically reduce access to the generated image storage unit realized by the memory.

(Embodiment 4) Hereinafter, a fourth embodiment of the present invention will be described with reference to the drawings.

FIG. 18 is a block diagram of the address generator of the mapping apparatus according to the fourth embodiment of the present invention. The calculation unit 701 calculates the drawing points inside the polygon and the parameters used for calculating the corresponding points using the positions of the polygon vertices and the corresponding points given to the polygon vertices. The first edge generation unit 702a uses the parameters generated by the calculation unit 701 to calculate candidate points and corresponding points corresponding to drawing points of polygon edges. The second edge generation unit 702b uses the parameters generated by the calculation unit 701 to calculate candidate points and corresponding points corresponding to drawing points of polygon edges. The selection unit 703 is the first edge generation unit 702a.
Or the second edge generation unit 70
The result generated in 2b is selected, and the selected result is output to the first edge generating unit 702a and the second edge generating unit 702b. The span generation unit 704 is the calculation unit 70.
Using the parameters generated in 1 and the result selected by the selection unit 703, the drawing point of the polygon span and the corresponding point are calculated, and the calculation result is output.

The operation of the address generator of the mapping device configured as described above will be described below with reference to the drawings.

FIG. 19 shows pixels on the screen which the edges of the drawn polygon cross. Reference numeral 801 denotes an edge of a polygon, and it is assumed that the polygon exists on the right side of the polygon, and the relationship shown in (Equation 1) and (Equation 2) is established between the polygon and the texture. Further, this polygon edge passes through P0 (6,0) and P4 (0,4) on the screen, and the first edge generation unit 702a and the second edge generation unit 702a define screen pixels defined in advance. The candidate points and corresponding points corresponding to the edge drawing points are calculated according to the y-axis of the sample points of. Here, the sample point of a pixel is the point closest to the origin in that pixel. Points 802a, 802b, 802c, 802d, 80
Let 2e be the drawing points P0, P1, P2, P3, and P4 calculated in the scan line direction by the incremental algorithm.
Pixels 803a, 803b, 803c, 803d, 803
e is the drawing points 802a, 802b, 802c,
It is assumed that pixels 802d and 802e are reflected. Point 804
Let a, 804b, 804c, 804d, and 804e be sample points of the pixels 803a, 803b, 803c, 803d, and 803e. Furthermore, arrows indicated by 805a to 805d indicate transitions of pixel sample points drawn along polygon edges. Here, in the polygon to be texture-mapped, the screen coordinates and texture coordinates of each vertex are (x, y, u, v) = (6,0,0,0) (0,4,0,
16) is a triangle that is (8,5,16,16).

FIG. 20 is an enlarged view near the drawing point P0 (802a). 20, the same reference numerals as those in FIG. 19 denote the same parts as those in FIG. The point 901 is the closest sample point existing in the left direction in the figure with respect to the drawing point 802b (same as 804b), and the point 902 is the closest sample point existing in the right direction in the figure with respect to the drawing point 802b. To do. Here, it is assumed that the sample points 901 and 902 are candidate points for the drawing point 802b. Also, the arrow indicated by 903 is the sample point 8
It is assumed that the change from 04a to the sample point 901 and the arrow 904 indicates the change from the sample point 804a to the sample point 902.

At this time, sampling points 901 and 902
From the drawing point 802b and the corresponding point u at 903
Change (du / dy (L)) is (Equation 42), and the change (du / dy (R)) of the corresponding point u in 904 is (Equation 43) (correspondence from sample point 804a to drawing point 802b) The change of the point u is du / dy).

[0220]

[Equation 42] du / dy (L) = du / dy-∂u / ∂xx
(Dx / dy- [dx / dy])

[0221]

[Expression 43] du / dy (R) = du / dy + ∂u / ∂xx
(1−dx / dy + [dx / dy]) The change of these two corresponding points u is represented by the sampling point 804a,
804b, 804c, 804d, and 804e, the pixels 803a and 803 can be selected appropriately.
The ideal corresponding points of b, 803c, 803d, and 803e can be calculated. In FIG. 19, 805a, 80
5c is du / dy (L), and 805b and 805d are du.
/ Dy (R) will be used to calculate the corresponding points.

The change of the two corresponding points u (du / dy
Selection of (L) and du / dy (R)) is performed as follows. Error E of x on the x-axis between the drawing point and the sampling point
(Distance) is calculated by repeating the incremental algorithm to E
(Equation 46) is accumulated. However, the cumulative error does not exceed 1. Similar to the corresponding point u, there are two increments of x and v, respectively, and these are selected by the same procedure. Two increments of the drawing point x and the corresponding point v are dx / dy (L),
dx / dy (R), dv / dy (L), dv / dy
Represented as (R). These are (Formula 44) and (Formula 45)
It is calculated by the calculation unit 701 according to the formula shown in.

[0223]

Dx / dy (L) = [dx / dy] dx / dy (R) = [dx / dy] +1

[0224]

[Expression 45] dv / dy (L) = dv / dy-∂v / ∂xx
(Dx / dy- [dx / dy]) dv / dy (R) = dv / dy + ∂v / ∂x × (1 − dx
/ Dy + [dx / dy]) The two increments are selected by cumulatively adding the difference (expressed as an error) between the drawing point and the sampling point shown in (Equation 46). If this cumulative error is 1 or more, dx / Dy (R), du / dy
(R) and dv / dy (R) are used (increment of (R)). If the cumulative error is less than 1, dx / dy
(L), du / dy (L), and dv / dy (L) are used (increment of (L)).

[0225]

[Equation 46] E = dx / dy- [dx / dy] In FIG. 19, the sampling point 804a is changed to the sampling point 80.
When changing to 4b, the cumulative error E0 of x is (Equation 47)
And the increment for the drawing point x and the corresponding points u and v is (L)
Increment of.

[0226]

[Equation 47] E0 = dx / dy- [dx / dy] = 1.5-[1.5] = 0.5 <1.0 When changing the sampling point 804b to the sampling point 804c, the cumulative error E1 of x becomes (Equation 48) (R) is used for the increment regarding the point x and the corresponding points u and v.

[0227]

E1 = E0 + dx / dy- [dx / dy] = 0.5 + 1.5-[1.5] = 1.0> = 1.0 When the sampling point 804c is changed to the sampling point 804d, the cumulative error E2 of x is 49), and (L) is used as the increment for the drawing point x and the corresponding points u and v.

[0228]

[Equation 49] E2 = E1 + dx / dy-[dx / dy] = 0 + 1.5-[1.5] (∵ E1 exceeds 1, so E1 = E1-
1) = 0.5 <1.0 When the sampling point 804d is changed to the sampling point 804e, the cumulative error E3 of x is (Equation 50), and (R) is used for the increment regarding the drawing point x and the corresponding points u and v. .

[0229]

E3 = E2 + dx / dy- [dx / dy] = 0.5 + 1.5- [1.5] = 1.0> = 1.0 Specifically, the corresponding points are calculated as follows. First, the relationship between the screen coordinates of the polygon to be texture-mapped and the texture coordinates ((x, y, u, v) = (6,0,0,0) (0,
4,0,16) (8,5,16,16) triangular polygon), dx / dy, du / dy, dv / d of three sides forming the polygon
y, dx / dy, dx / dy (L), du / dy
(L), dv / dy (L), dx / dy (L), dx /
dy (R), du / dy (R), dv / dy (R), d
x / dy (R), ∂u / ∂x, ∂v / ∂x are calculated by the arithmetic unit 70.
Calculate with 1. Here, the edge 801 used for the description is calculated (Equation 51).

[0230]

Dx / dy = -6/4 = -1.5 du / dy = -0/4 = 0 dv / dy = 16/4 = 4 ∂u / ∂x =-((y1-y0) (u2- u1)-(y2-y1) (u1-u0)) / ((x1-x0) (y2-y1)-(x2-x1) (y 1-y0)) =-((4-0) (16- 0)-(5-4) (0-0)) / ((0-6) (5-4)-(8-0) (4-0)) = 64/38 ∂v / ∂x =-( (y1-y0) (v2-v1)-(y2-y1) (v1-v0)) / ((x1-x0) (y2-y1)-(x2-x1) (y 1-y0)) =-( (4-0) (16-16)-(5-4) (16-0)) / ((0-6) (5-4)-(8-0) (4-0)) = -16 / 38 dx / dy (L) = [dx / dy] = -2.0 du / dy (L) = 0 -∂u / ∂x * ((-1.5)-(-2.0)) = 0 -64 / 38 * ( 0.5) = -0.84 dv / dy (L) = 0 -∂v / ∂x * ((-1.5)-(-2.0)) = 4-(-16/38) * (0.5) = 4.21 dx / dy ( R) = [dx / dy] +1 = -1.0 du / dy (R) = 0 + ∂u / ∂x × (1-(-1.5) + (-2.0)) = 0 + (64/38) × (0.5) = 0.84 dv / dy (R) = 0 + ∂v / ∂x × (1-(-1.5) + (-2.0)) = 4 + (-16/38) × (0.5) = 3.79 E = dx / dy- [dx / dy] = -1.5-(-2.0) = 0.5 Shown in (Equation 51) calculated by the arithmetic unit 701. Based on the parameter, the first edge generation section 702a, a second edge generation section 702b, the processing of the selection unit 703 will be described.

The first edge generator 702a includes an arithmetic unit 7
The increment of (L) calculated in 01 (see (Equation 51)) and the screen coordinate value and texture coordinate value (6,0,0,0) of the point P0 are given as initial values. Second edge generation unit 702b
Is the increment of (R) calculated by the calculation unit 701 and ((Equation 5
1)), and the screen coordinate value and the texture coordinate value (6,0,0,0) of the point P0 are given as initial values. Selector 703
Is given a difference E (see (Equation 51)) between the drawing point and the sample point calculated by the calculation unit 701.

In the edge generators 702a and 702b, the initial value is (x, y, u, v) given to the polygon vertex,
After that, each increment value is added to (x, y, u, v) which is the output from the selection unit 703. The first edge generation unit 702a uses the increment value of (L), and the second edge generation unit 702b uses the increment value of (R).
Calculate (x, y, u, v).

The selecting unit 703 sequentially adds E (= dx / dy- [dx / dy]) when calculating the edge points to calculate the cumulative error. When the accumulated error is less than 1, (x, y, u, v) output from the first edge generation unit 701a is output from the second edge generation unit 702b when the accumulated error is 1 or more (x , y, u, v) is selected.

FIG. 21 shows the flow of processing in the first edge generating section 702a, the second edge generating section 702b, and the selecting section 703.

In FIG. 21, processing 2201 and 220
Reference numerals 3, 2205, and 2207 denote the first edge generation unit 702.
This is the process in a. Process 2202, 2204, 22
06 and 2208 are processes in the second edge generation unit 702b. Processing 2211, 2212, 2213, 2
Reference numeral 214 denotes a process in the selection unit 703, which selects data according to this value. 2221, 2222, 222
Reference numerals 3 and 2224 indicate how data is selected by the selection unit 703. Further, in the figure, the shaded numerical values (data) are values calculated by the calculation unit 701. Further, the data output from the selection unit 703 is underlined.

Next, the calculation of screen coordinate points and texture coordinate points in this target polygon (step 0)
To (step 4).

At each step, the y coordinate value is 0, 1, 2, 3, 4.

(Step0) Point P0 (6,0,0,
0) is the initial value in the subsequent processing.

This (6,0,0,0) is the output value from the selection unit 703 and the span generation unit 704.

(Step 1) The initial value of the increment algorithm is the point P0 (6,0,0,0) which is the vertex of the polygon.

In the first edge generation unit 702a and the second edge generation unit 702b, the process 220 is performed.
1 and 2202 are performed. These are processes for adding the increment values given by the arithmetic unit 701.

At the same time, the selection unit 703 performs the processing 2211. This is a process of cumulatively adding E given from the calculation unit 701. Process 220
After completion of steps 1, 2202, and 2211, the selection unit 703 selects data from the value of the cumulative error E (sum (E) in the figure). Here, since it is less than 1, (x, y, u, v) output from the first edge generation unit 702a is selected. That is, (4.00,1.00, -0.84,4.21) is selected (2221)
Selection of). This is the incremental calculation of 805a shown in FIG.

(Step2) Next, the first edge generator 702
In a and the second edge generation unit 702b, processes 2203 and 2204 are performed, respectively. Data previously selected by the selection unit 703 (4.00,1.00, -0.84,4.21)
Again, the increment value is added in the same manner as above.

At the same time, in the selecting unit 703, a process 2412 of accumulatively adding E given from the arithmetic unit 701.
Perform After the processing 2403, 2404, and 2412 is completed, the selection unit 703 determines the cumulative error E (sum (E) in the figure).
Data is selected from the value of. Here, since it is 1 or more, it is output from the second edge generation unit 702b.
Select (x, y, u, v) (selection 2222). That is,
Select (3.00,2.00,0.00,8.00). This is the incremental calculation of 805b shown in FIG. Further, since the cumulative error E has exceeded 1 here, the process of subtracting 1 from this is simultaneously performed (error correction).

(Step3) Next, the first edge generator 7
02a and the second edge generation unit 702b perform processing 2205 and 2206, respectively. Selector 70
In 3, the previously selected data (3.00,2.00,0.00,8.
The increment value is added to 00) as before.

At the same time, in the selecting unit 703, a process 2413 for accumulatively adding E given from the arithmetic unit 701.
Perform After the processes 2405, 2406, and 2413 are completed, the selection unit 703 selects the cumulative error E (sum (E) in the figure).
Data is selected from the value of. Here, since it is less than 1, it is output from the first edge generation unit 702a.
Select (x, y, u, v) (selection 2223). That is,
Select (1.00,3.00, -0.84,12.21). This is shown in FIG.
It is the incremental calculation of 805c shown in FIG.

(Step4) Furthermore, the first edge generator 70
2a and 2nd edge generation part 702b perform processing 2207 and 2208, respectively. Selector 703
, The previously selected data (1.00,3.00, -0.84,12.
To 21), add the increment value as before.

At the same time, in the selecting unit 703, a process 2414 of accumulatively adding E given from the arithmetic unit 701.
Perform After the processes 2407, 2408, and 2414 are completed, the selection unit 703 selects the cumulative error E (sum (E) in the figure).
Data is selected from the value of. Here, since it is 1 or more, it is output from the second edge generation unit 702b.
Select (x, y, u, v) (selection 2224). That is,
Select (0.00,4.00,0.00,16.00). This is the incremental calculation of 805d shown in FIG.

By executing the processing from (step 0) to (step 4) as described above, it is possible to obtain the corrected screen coordinate value and texture coordinate value for the edge drawing point of the polygon edge. That is, it is possible to calculate drawing points and corresponding points regarding ideal polygon edges.

The first edge generator 702a and the second edge generator 702a
A configuration example of the edge generation unit 702b and the selection unit 703 will be described.

FIG. 22 is a block diagram of the first edge generator 702a. Here, it is output from the calculation unit 701.
(x, y, u, v) is an integer. That is, it is assumed that the vertex coordinates of the target polygon and texture are integers. The selectors 2301a to 2301d are the data (x, y, u, v) output from the calculation unit 701 or the adder 2302a.
To 2302d output data (x, y, u, v) is selected. Selectors 2301a, 2301b, 2301c,
2301d is a selector that selects x, y, u, and v, respectively. In these selectors, when calculating the first edge point, the output (x, y, u,
v) is selected, and in other cases, the data (x, y, u, v) output from the adders 2302a to 2302d is selected.

Further, the adder 2302a performs an increment process for x, the incremented value is x output from the selection unit 703, and the incremented value is dx / dy (L) (see (Equation 51)). Similarly, the adder 2302b performs an increment process for y, and the incremented value is output from the selection unit 703.
y, increment value is 1. The adder 2302c performs an increment process for u, the incremented value is u output from the selection unit 703, and the increment value is du / dy (L) (see (Formula 51)).
And The adder 2302d performs an increment process on v, the incremented value is v output from the selection unit 703, and the increment value is dv / dy (L) (see (Equation 51)). These increment values do not change until the processing for one edge is completed.

As a result, the processing 2201 described above,
The first edge generation unit 702a that executes 2203, 2205, and 2207 can be realized. The second edge generation unit 702b has the same processing content as the first edge generation unit 702a. Therefore, they can be realized by the same configuration, and the description thereof will be omitted here.

FIG. 23 is a part of the configuration diagram of the selection unit 703. In FIG. 23, reference numeral 2401 denotes an adder, and the initial value is the difference (E) between the drawing point calculated by the incremental algorithm and the sample point of the screen pixel, and the error E (equation 51) for x in each scan line is calculated. It has a mechanism of cumulative addition (cumulative error is sum (E)).
This error E does not change until the processing for one edge is completed. Selectors 2402a to 24
02d receives the result of the cumulative error calculated by the adder 2401 and selects the output value of the first edge generation unit 702a or the output value of the second edge generation unit 702b.
The selector 2402a is a selector 2402c related to x.
Is for u, and the selector 2402d is for v. Regarding the value of y, the first edge generation unit 702a
, And the output value from the second edge generator 702b are the same, there is no selector for y.

In the adder 2401, when the adder is an adder that performs cumulative addition up to a value less than 1, by detecting whether or not the addition result overflows,
(Equation 47) to (Equation 50) and the above-mentioned processing contents ((ste
It is possible to implement (p0) to (step4)) and the processes 2221 to 2224 shown in FIG. That is, if the addition result of the adder 2401 overflows, the second
The output value (x, y, u, v) of the edge generation unit 702b is selected. If no overflow occurs, the output value (x, y, u, v) of the first edge generator 702a is selected.

FIG. 24 shows the processing outline of these operations. FIG. 25 shows a target triangular polygon to be mapped by the processing of FIG. 24 (target polygon). In FIG. 25, it is assumed that (ix, iy, iu, iv) of the vertices are represented by integers, and the respective vertex coordinates and corresponding point coordinates satisfy the condition of (Equation 52).

[0257]

[Expression 52] iy0 <iy1 = iy2 ix1 <ix0 <ix2 FIG. 24 shows an outline of processing, and each processing will be described with reference to the constituent elements in the fourth embodiment.
The process 1301 is a process in the calculation unit 701. 1302
Is the drawing point (x, y) in the span generation unit 704, the corresponding point
This is the output processing of (u, v). 1303 is the span generation unit 7
This is a process of calculating a drawing point in the span direction and a corresponding point in 04 (see (Equation 7)). A processing end condition 1304 is a processing end condition in the span generation unit 704. 1305 denotes a first edge generator 702a and a second edge generator 702a.
The processing common to the edge generation unit 702b is the calculation processing of the drawing point of the edge (see (Equation 4)) and the calculation processing of the numerical value serving as the selection criterion in 1306. The numerical value as the selection criterion is E shown in (Equation 46). Reference numeral 1306 denotes a selection process in the selection unit 703, which allows du /
dy (L) and du / dy (R) are selected. Reference numeral 1307 denotes a corresponding point calculation process in the first edge generation unit 702a. Reference numeral 1308 is a corresponding point calculation process in the second edge generation unit 702b. Reference numeral 1309 denotes a termination condition for the entire processing, which terminates the processing of this device.

As described above, by performing the processing shown in FIG. 24 to calculate the corresponding points and drawing points for the polygon shown in FIG. 25, it is possible to realize accurate corresponding point calculation.

That is, it is possible to select du / dy (L) in 805a and 805c and select du / dy (R) in 805b and 805d as parameters used for calculation of corresponding points. The selection unit 703 selects the increment for the corresponding point u as described above. Although the corresponding point u has been described here, the same applies to the corresponding point v, illuminance, opacity, and the like.

That is, the corresponding point generated by the first edge generating section 702a or the second edge generating section 702b is selected by the selecting section 703 ((Equation 4
7) to (Equation 50)), it is possible to calculate an ideal corresponding point in a pixel where a polygon edge exists. Furthermore, by ideally calculating the corresponding points of the pixels in which the polygon edge that is the first pixel of the span exists, the corresponding points of the pixels included in this span can also be ideally calculated (see (Equation 7)). ).

As described above, according to this embodiment, the first
Edge generator 702a and second edge generator 702b
703 for selecting parameters generated in
By providing, it becomes possible to obtain an ideal corresponding point for the drawing pixel. Thereby, the first embodiment,
Similar to 2 and 3, the ideal corresponding points can be calculated for all the pixels onto which the polygon is projected, and the image quality of the mapping image can be improved.

Further, each edge generation unit only has the same processing load as the conventional incremental algorithm. The calculation of the parameter (incremental value) used for each edge raw correction can be performed in advance for the edge of the polygon, and does not significantly increase the load on these calculations.

In this embodiment, it is necessary to provide two edge generators. However, as compared with the conventional method, only a delay for selecting the parameters generated by each occurs, and the high speed of the entire processing is not impaired.

In this embodiment, the first edge generator 702a and the second edge generator 702b are operated simultaneously. However, by executing the cumulative error calculation of the selection unit 703 in advance, it is possible to stop the operation of either one of the edge generation units.
This will be described with reference to the process flow diagram of FIG. After the processing of the cumulative error meter 2211 is completed, it is possible to recognize which edge generation unit data is to be selected. Since the result of the process 2201 is selected here, the process 220
There is no problem in stopping the processing (processing circuit) of the second edge generation unit 702b that performs the step 2. That is, it may be possible to execute the cumulative error calculation of the selection unit 703 in advance and stop the operation of either one of the edge generation units. As a result, the power consumption of the entire device can be suppressed.

In the present embodiment, the sample point in the pixel is the point closest to the origin in the pixel, but the position of this sample point is arbitrary.

In the present embodiment, the relationship between the polygon and the texture is set to (Equation 1) and (Equation 2), but the relationship is not limited to (Equation 1) and (Equation 2). For example, even if this relationship is set to (Equation 17) and (Equation 18), accurate corresponding point calculation can be performed. In other words, it is possible to deal with all corresponding point expressions.

In this embodiment, the calculation of the corresponding points of the polygon has been described, but the illuminance calculation, the opacity calculation,
Similar processing can be realized for bump mapping calculation and the like. Also, regarding these, similarly to the corresponding points, it is possible to correspond to all relational expressions. FIG.
0, FIG. 11 shows an example of the overall configuration in the case of realizing texture mapping and illuminance mapping.

In the figure, 1701 and 1801 can realize texture mapping and illuminance mapping by using those described in the present embodiment. Opacity mapping, bump mapping, etc. can be realized with the same configuration. Since these configuration diagrams have been described in the first embodiment, they are omitted here. In this embodiment, the method in which the anti-aliasing process is not performed has been described, but the same process can be performed including the anti-aliasing process.

Even if sub-pixel positioning is performed, the same correction processing as in this embodiment can be performed.

(Embodiment 5) Hereinafter, a fifth embodiment of the present invention will be described with reference to the drawings.

FIG. 26 is a block diagram of the address generator of the mapping apparatus according to the fifth embodiment of the present invention. In FIG. 26, the same reference numerals as those in FIG.
The same as 8 is shown. The edge generation unit 2502 calculates the screen coordinate value at the polygon edge and the corresponding texture coordinate value by performing the incremental calculation using the specified increment value using the data generated by the calculation unit 701. The increment switching unit 2503 includes the calculation unit 7
The increment value in the edge generation unit 2502 is designated using the data generated in 01. Further, the polygon to be mapped here is the same as in the fourth embodiment (FIG. 19), and the corresponding points are (Equation 1) and (Equation 2).
It is assumed that the relationship of

At this time, the texture mapping processing in this embodiment will be described. The calculation unit 701 calculates necessary parameters as in the fourth embodiment ((Equation 42), (Equation 43), (Equation 44), (Equation 45)).
reference). Specifically, the value is shown in (Equation 51).

The edge generation unit 2502 receives dx / dy (L), du / dy (L), d according to the signal from the increment switching unit 2503 shown as next_inc_ident in FIG.
v / dy (L), dx / dy (R), du / dy (R),
The increment value of dv / dy (R) is switched, and the increment process (DDA process) is performed according to the selected increment value.

Incremental switching unit 2503 cumulatively adds E ((Equation 46), (Equation 51)) described in the fourth embodiment, and controls the switching using the result. That is, the increment value used in the edge generation unit 2502 is switched. The switching process performed by the increment switching unit 2503 is similar to the selection process performed by the selection unit 703 described in the fourth embodiment. Specifically, when the cumulative error (sum (E)) is less than 1, the increment value used by the edge generation unit 2502 is dx / dy.
(L), du / dy (L), dv / dy (L).
If the accumulated error is 1 or more, the edge generation unit 2502
The increment values used in dx / dy (R), du / dy (R),
dv / dy (R). Further, when the accumulated error is 1 or more, a process of subtracting 1 from the accumulated error (sum (E)) is also performed.

Here, the processing flow of the edge generation section 2502 and the increment switching section 2503 in this embodiment will be described (the target polygon is shown in FIG. 19). The flow of these processes is shown in FIG. 27, and this process will be described below. In FIG. 27, processing 2601, 2602, 26
03 and 2604 are processes performed by the edge generation unit 2502. Processing 2611, 2612, 2613, 2614
Is a process performed by the increment switching unit 2503. The shaded portion in FIG. 27 is an increment value (screen coordinate, texture coordinate) controlled by the increment switching unit 2503 and used by the edge generating unit 2502. Also, underlined data is output data (x, y, u, v) from the edge generation unit 2502.

(Step0) The point P0 (6,0,0,
0) is the initial value in the subsequent processing.

The initial value of the cumulative error sum (E) is 0.
It is. This (6,0,0,0) is the output value from the edge generation unit 2502.

(Step 1) In the increment switching unit 2503, the error E
Is cumulatively added (process 2611). Since the cumulative error here is less than 1, the edge generation unit 2502 is requested to use the increment value of (L) via next_inc_ident shown in FIG.

Upon receiving this, the edge generation unit 2502
(L) increment value (-2.00, tep0) result (initial value)
1.00, -0.84, 4.21) is added and the addition result is output (process 2601). Therefore, (4.00,1.00, -0.84,4.21) is output. This is the incremental calculation of 805a shown in FIG.

(Step 2) Next, in the increment switching unit 2503, the cumulative error calculation result exceeds 1 (process 2612).
Therefore, the edge generation unit 2502 is requested to use the increment value of (R) via next_inc_ident shown in FIG.

Upon receiving this, the edge generation unit 2502
(R) increment value (-1.00, 1.00, 0.84,
3.79) is added and the addition result is output (process 260).
2). Therefore, (3.00,2.00,0.00,8.00) is output. This is the incremental calculation of 805b shown in FIG.

Further, since the cumulative error exceeds 1, the increment switching section 2503 carries out a process of subtracting 1 from the cumulative error.

(Step3) Further, in the increment switching section 2503,
The error E is cumulatively added (process 2613). Since the cumulative error here is less than 1, the edge generation unit 2502 is requested to use the increment value of (L) via next_inc_ident shown in FIG.

The edge generation unit 2502 which receives this receives (s
(L) increment value (-2.00, 1.00, -0.8) for the result of tep2)
4, 4.21) is added and the addition result is output (Processing 260
3). Therefore, (1.00,3.00, -0.84,12.21) is output. This is the incremental calculation of 805c shown in FIG.

(Step 4) Further, in the increment switching unit 2503, the cumulative error calculation result exceeds 1 (process 261).
4). Therefore, the edge generator 2502 can
Request to use the increment value of (R) via next_inc_ident shown in 6.

The edge generation unit 2502 which receives this receives (s
(r) increment value (-1.00, 1.00, 0.84,
3.79) is added and the addition result is output (process 261).
4). Therefore, (0.00,4.00,0.00,16.00) is output. This is the incremental calculation of 805d shown in FIG.

In addition, since the cumulative error exceeds 1, the increment switching unit 2503 performs a process of subtracting 1 from the cumulative error.

By executing the processing from (step 0) to (step 4) as described above, ideal polygon coordinates and texture coordinates can be obtained for polygon edges.

Next, a configuration example of the edge generating section 2502 and the increment switching section 2503 will be described.

FIG. 28 is a block diagram of the edge generator 2502. Selectors 2701a, 2721c and 270
1d is controlled by next_inc_ident shown in FIG. With these selectors, the edge generation unit 250
Incremental value used in 2 (incremental value of (L) or (R)
(Increment value of).

Adders 2702a, 2702b, 2702
Incremental processing is performed by using c and 2702d. At this time, the increment value selected by the selectors 2701a, 2721c, 2701d is used. As the incremented value, the value output from the edge generation unit 2502 is used. In FIG. 28, the value output from the edge generator is fed back and input to these adders 2702a to 2702d. Thereby, the processing 2601, 2602, 2
603 and 2604 are realized.

Selectors 2703a, 2703b, 270
In 3c and 2703d, when the first edge point is calculated, the output from the calculation unit 701 is (x, y, u, v)
Otherwise, select the data (x, y, u, v) output from the adders 2702a to 2702d. This makes it possible to sequentially calculate (x, y, u, v) from the initial value of the target polygon edge.

FIG. 29 is a block diagram of the increment switching unit 2503. In FIG. 29, the adder 2801 accumulates the error E (equation 51) for x in each scan line, with the initial value being the difference (E) between the drawing point calculated by the incremental algorithm and the sample point of the screen pixel. It is a mechanism for adding (cumulative error sum (E)).

The index for controlling the increment value in the edge generation unit 2502 and the index for switching the increment value in the increment switching unit 2503 depends on whether the accumulated error (sum (E)) exceeds 1, and when it exceeds 1. After that, a process of subtracting 1 is performed. That is, the increment switching unit 2503 is configured by a cumulative adder up to a value less than 1, and determines whether or not the cumulative addition result overflows. As a result, the processes 2611, 2612, 2613 and 2614 are realized.

Therefore, the edge generator 2 shown here
502 and the configuration of the increment switching unit 2503 (see (FIG. 28) and (FIG. 29)), the above-described processing (from (step 0) to (step)
4)) can be realized.

As described above, according to this embodiment, the increment switching unit 2503 calculates the difference (error) between the value x of the screen coordinate value and the sample point of the screen pixel by the calculation unit 701. It is realized by cumulatively adding (sum (E)) the calculated values. Further, specifically, it is only necessary to determine whether or not the cumulative addition result overflows from the value of this cumulative error, and the screen coordinate value and the texture coordinate value of the polygon edge are calculated depending on whether the cumulative addition result overflows or not. Increment value ((L) increment value,
The increment value of (R) is controlled so as to be used in the increment calculation in the edge generation unit 2502.

Further, the edge generator 2502 performs the increment calculation using the increment value designated by the increment switcher 2503. Here, the screen coordinate value and the texture coordinate value of the point representing the desired polygon are ideal values for the screen pixels.

That is, by providing the edge generating section 2502 and the increment switching section 2503, it is possible to obtain the ideal screen coordinate value and texture coordinate value for the drawing pixel. This makes it possible to calculate the ideal corresponding points for all the pixels included in the polygon as in the previous embodiment, and improve the image quality of the image to be mapped.

The circuit scale of the edge generating section 2502 is increased only by the number of selectors corresponding to the number of increment values, and the circuit does not increase dramatically. Further, the processing speed is in a negligible range. Therefore, it has a processing capacity similar to that of the conventional incremental algorithm.

As for the circuit scale of the increment switching unit 2503, as can be seen from the configuration diagram of FIG. 29, the configuration is simple and the circuit scale is small.

Therefore, with regard to the configuration of the present embodiment of the present invention, the processing speed and the circuit scale are the same as those of the conventional incremental algorithm, and moreover, the ideal corresponding points are calculated. Is something that can be done.

In this embodiment, the sample point in the pixel is the point closest to the origin in the pixel, but the position of this sample point is arbitrary.

In this embodiment, the relationship between the polygon and the texture is set to (Equation 1) and (Equation 2), but the relationship is not limited to (Equation 1) and (Equation 2). For example, even if this relationship is set to (Equation 17) and (Equation 18), accurate corresponding point calculation can be performed. In other words, it is possible to deal with all corresponding point expressions.

In the present embodiment, the calculation of the corresponding points of the polygon has been described, but as described in the first embodiment, similar processing is realized for the illuminance calculation, the opacity calculation, the bump mapping calculation and the like. It is possible (see FIGS. 10 and 11). Further, opacity mapping, bump mapping, and the like can be realized with the same configuration.

In the present embodiment, the method in which the anti-aliasing process is not performed has been described, but the same process can be performed including the anti-aliasing process.

Even if sub-pixel positioning is performed, the same correction processing as in the present embodiment can be performed.

(Sixth Embodiment) A sixth embodiment of the present invention will be described below with reference to the drawings.

FIG. 30 is a block diagram of the address generator of the mapping apparatus according to the sixth embodiment of the present invention. 30, reference numerals that are the same as those in FIG.
The same as 8 is shown. The calculation unit 2901 calculates the drawing points inside the polygon and the parameters used for calculating the corresponding points from the positions of the polygon vertices and the corresponding points given to the polygon vertices. The edge generation unit 2902 calculates the screen coordinate value at the polygon edge and the corresponding texture coordinate value by performing an incremental calculation using the parameters generated by the calculation unit 2901 and the data corrected by the correction unit 2903. To do. The correction unit 2903 performs correction processing on the corresponding points generated by the edge generation unit 2902 using the parameters generated by the calculation unit 2901 and the drawing points generated by the edge generation unit 2902. It is also assumed that the polygon to be mapped is the same as in the fourth embodiment (FIG. 19), and the corresponding points have the relationships of (Equation 1) and (Equation 2).

At this time, the texture mapping processing in this embodiment will be described. Unlike the fourth and fifth embodiments, the arithmetic unit 2901 calculates only the increment value of (L) ((Equation 42), (Equation 44), (Equation 4)
5)). Specifically, dx / d shown in (Equation 51)
Calculate y (L), du / dy (L), and dv / dy (L). Further, the difference (E) (Equation 46) between the drawing point and the sample point, which is necessary for determining whether or not the correction processing in the correction unit 2903 is performed, is calculated.

The edge generation unit 2902 has a calculation unit 2901.
From (x, y, u, v) given as the initial value, and thereafter output from the correction unit 2903 following the edge generation unit 2902 (x, y, u, v)
y, u, v), dx / dy calculated by the arithmetic unit 2901
Incremental processing (DDA processing) is performed with (L), du / dy (L), and dv / dy (L) as incremental values.

The correction unit 2903 has an edge generation unit 2902.
Correct the (x, y, u, v) data output from. The correction is required when the cumulative error (sum (E)), which is the cumulative addition result of the difference (E) between the drawing point and the sample point calculated by the calculation unit 2901, is 1 or more. By performing the correction process each time when correction is necessary, the difference between the x value before correction and the x coordinate value of the sample point is always one pixel. Further, after the correction processing is performed, 1 is subtracted from this accumulated error. The correction process here is performed as follows.
That is, if the correction processing of (Equation 53) is performed on the calculated (x, y, u, v), ideal screen coordinate values and texture coordinate values can be calculated. The u, v) value is output from the correction unit 2903.

[0312]

[Expression 53] x ← x + 1 y ← yu u ← u + ∂u / ∂xv ← v + ∂v / ∂x These processes will be described by taking the polygon shown in FIG. 19 as an example. FIG. 31 shows the target polygon shown in FIG.
The transition of the incremental processing and the correction processing are shown. 19 that are the same as those in FIG. 19 are denoted by the same reference symbols. 805
The arrows indicated by a, 3005b, 805c, and 3005d are
This is a transition of sample points by the incremental processing performed by the edge generation unit 2902. The arrows indicated by 3001 and 3002 are transitions of sample points by the correction processing performed by the correction unit 2903.

Regarding the calculation of the screen coordinate value and the texture coordinate value, the target polygon is shown in FIG.
When set to 1, the edge generation unit 2902 and the correction unit 29
The specific processing in 03 is shown in FIG. In FIG. 32, processes 3101, 3102, 3103 and 3104 are shown.
Represents the incremental processing performed by the edge generator 2902. Processes 3111, 3112, 3113, and 3114 are processes for determining whether or not the correction process in the correction unit 2903 is performed, and perform the above-described cumulative error calculation. Processing 31
Reference numerals 22 and 3124 represent correction processing performed by the correction unit 2903.

This processing will be described in order along the polygon edge 801.

(Step0) Point P0 (6,0,0,
0) is the initial value in the subsequent processing. The initial value of the cumulative error sum (E) is 0. This (6,0,0,0) becomes the output value from the edge generation unit 2902.

(Step1) The edge generator 2902 makes (step0)
Incremental value (-2.00,1.00, -0.84,
The result of adding (4.21) (4.00, 1.00, -0.84, 4.21) is output to the correction unit 2903 (process 3101).

The correction unit 2903 cumulatively adds the error E (process 3111). Here, since the result of the accumulated error is less than 1, the correction process is not performed.

That is, the output from the correction unit 2903 is
(4.00,1.00, -0.84,4.21).

(Step2) Next, the edge generator 2902
It is the result of adding the increment value (-2.00,1.00, -0.84,4.21) to the result (4.00,1.00, -0.84,4.21) of (step1) (2.0
0,2.00, -1.68,8.42) is output to the correction unit 2903 (process 3102).

The correction unit 2903 cumulatively adds the error E (process 3112). Here, since the result of the accumulated error is 1 or more, the correction process is performed. The correction process is (1.00,0,00, ∂u / ∂x, ∂) to the input (2.00,2.00, -1.68,8.42).
It is realized by adding v / ∂x) = (1.00, 0.00, 1.68, -0.42) (process 3122). As a result of the correction (3.00, 2.
00,0.00,8.00), which is the output value. After this processing, the value obtained by subtracting 1 from the cumulative error in the correction unit 2903 is set as the cumulative error.

(Step3) The edge generator 2902 makes (step2)
For the result (3.00,2.00,0.00,8.00), the increment value (-2.00,
1.00, -0.84,4.21) is the result (1.00,3.00, -0.
84, 12.21) is output to the correction unit 2903 (process 310).
3).

The correction unit 2903 cumulatively adds the error E (process 3113). Here, since the result of the accumulated error is less than 1, the correction process is not performed.

That is, the output from the correction unit 2903 is
(1.00,3.00, -0.84,12.21).

(Step4) Further, the edge generation unit 2902
Is the result of (step3) (1.00,3.00, -0.84,12.21),
It is the result of adding the increment value (-2.00, 1.00, -0.84, 4.21)
(-1.00, 4.00, -1.68, 16.42) is output to the correction unit 2903 (process 3104).

The correction unit 2903 cumulatively adds the error E (process 3114). Here, since the result of the accumulated error is 1 or more, the correction process is performed. The correction process is as follows: (1.00,4.00, -1.68,16.42)
This is realized by adding (0.00, 1.68, -0.42) (process 3124). As the correction result, (0.00,4.00,0.00,16.00) can be obtained, and this is the output value. After this processing, the value obtained by subtracting 1 from the cumulative error in the correction unit 2903 is set as the cumulative error.

By executing the processing from (step 0) to (step 4) as described above, it is possible to obtain ideal screen coordinate values and texture coordinate values for polygon edges.

Next, the edge generator 2902 and the corrector 29
A configuration example of No. 03 will be described.

FIG. 33 is a block diagram of the edge generator 2902. In FIG. 33, 3201a, 3201b, 3
201c and 3201d are adders that perform an increment process. Each adder sequentially performs addition regarding x, y, u, and v. Thereby, the process 310
Incremental calculations of 1 to process 3104 can be realized.
Also, 3202a, 3202b, 3202c and 32
02d is a selector. When calculating the first edge point of the polygon edge, (x, y, u, v) output from the arithmetic unit 2901 is selected, and in other cases, the correction unit 2
Incremental processing is performed on (x, y, u, v) from 903 (x, y, u, v)
Select y, u, v). This makes it possible to sequentially calculate (x, y, u, v) from the initial value of the target polygon edge.

FIG. 34 is a block diagram of the correction unit 2903. In FIG. 34, 3301a, 3301c and 3
An adder 301d realizes the correction process shown in (Equation 53). Reference numeral 3302 denotes an adder that performs cumulative addition of the difference (E) between the drawing point calculated by the increment algorithm calculated by the calculation unit 2901 and the sample point of the screen pixel. Based on the result of the adder 3302, it is determined whether to output the data of the correction process or the data that is not corrected. Here, the adder 3302
Instead of determining by the value of the addition result in, the range of the adder is set to less than 1 and the data of the correction process is output or the data that is not corrected is output depending on whether the addition result overflows or not. You can also decide what to do. Also, 3303a, 3303
c and 3303d are selectors, and if the adder 3302 overflows, the correction data, that is, the adder 330
The selector outputs the output values from 1a, 3301c, and 3301d, and outputs the data that is not corrected if the adder 3302 does not overflow.

Therefore, the edge generator 2 shown here
902, the configuration of the correction unit 2903 ((FIG. 28), (FIG. 2
9)), the above-mentioned processing (from (step 0) to (step
4)) can be realized.

As described above, according to the present embodiment, by providing the edge generation unit 2902 and the correction unit 2903, it is possible to obtain ideal screen coordinate values and texture coordinate values for the pixel to be rendered. Becomes This makes it possible to calculate the ideal corresponding points for all the pixels included in the polygon as in the previous embodiment, and improve the image quality of the texture mapping image.

The edge generator 2902 has the same circuit scale as the conventional incremental calculation, and the processing speed is
It has the same processing power as the conventional incremental algorithm.

As for the circuit scale of the correction unit 2903, as can be seen from the configuration diagram of FIG. 34, the configuration is extremely simple and does not increase the circuit scale.

Therefore, the present invention can perform ideal corresponding point calculation without increasing the circuit scale and processing power. In this embodiment, the sample point in the pixel is the point closest to the origin in the pixel, but the position of this sample point is arbitrary.

In this embodiment, the relationship between the polygon and the texture is set to (Equation 1) and (Equation 2), but the relationship is not limited to (Equation 1) and (Equation 2). For example, even if this relationship is set to (Equation 17) and (Equation 18), accurate corresponding point calculation can be performed. In other words, it is possible to deal with all corresponding point expressions.

In this embodiment, the calculation of the corresponding points of the polygon has been described, but the illuminance calculation, the opacity calculation,
Similar processing can be realized for bump mapping calculation and the like (see FIGS. 10 and 11). Further, opacity mapping, bump mapping, and the like can be realized with the same configuration.

In the present embodiment, the method in which the anti-aliasing processing is not performed has been described, but the same processing can be performed including the anti-aliasing processing.

Even if sub-pixel positioning is performed, the same correction processing as in this embodiment can be performed.

(Embodiment 7) A seventh embodiment of the present invention will be described below with reference to the drawings.

FIG. 35 is a block diagram of a texture mapping device according to the seventh embodiment of the present invention. 35, the same reference numerals as those in FIG. 30 denote the same parts as those in FIG.

3404 is calculated by the arithmetic unit 2901,
E, ∂u / ∂x, ∂v / ∂x are input, correction processing of corresponding points at the drawing points calculated by the edge generation unit 2902, and span generation for calculating screen coordinates and texture coordinates of span points inside the polygon It is a department.

At this time, the texture mapping processing in this embodiment will be described. In this embodiment, by providing the span generation unit 3404 that simultaneously realizes the processing in the correction processing unit 2902 and the processing in the span generation unit 704 in the sixth embodiment, ideal screen coordinates and textures inside the polygon are obtained. Calculate the coordinates. The outline of the operation is similar to that of the sixth embodiment, and the increment processing is performed by the edge generation unit 2902.
The correction process shown in (Equation 53) is performed. The feature of the seventh embodiment is that the span generation unit 3404 performs this correction process. The span generation unit 3404 can perform the correction process and the span process because the correction process of (Equation 53) is the same as the process (Equation 7) in the span generation.

The span generation unit 34 for realizing this
A configuration diagram of 04 is shown in FIG. In FIG. 35, the process performed by the correction unit 2903 (Equation 53) in FIG. 30 is performed by the span generation unit 3403. In this case, this can be realized by adding the cumulative addition process of E for determining whether or not the correction process is executed to the span generation unit. FIG. 36 shows an example of its realized configuration.

In FIG. 36, the correction unit 2 shown in FIG.
The same components as those in the configuration diagram of 903 are designated by the same reference numerals. At this time, the adders 3301a and 3301
As described above, c, 3301d is used for the correction process.
Here, by inputting the operation trigger signal for correction processing and the span processing signal to this adder, the correction processing and the span processing can be realized by one circuit. That is, the span generation unit 3404 can perform correction processing and span processing. Also, the buffer 350
1a, 3501b, 3501c and 3501d are output buffers with enable.

Since the data before the correction processing (the output from the adder 3301 before the correction processing trigger is input) is not the span value, the span generation unit 34 uses this buffer.
It is for not outputting from 04.

The circuit scale can be reduced by implementing the correction process and the span process in a common circuit.

[0347]

As described above, according to the present invention, the corresponding points on the mapping image are corrected by using the difference between the drawing point obtained by the incremental algorithm and the sample point of each pixel, A high quality image can be generated. In addition, a pixel value determined by using the drawing point obtained by the incremental algorithm is distributed to a plurality of pixel values based on the relative positional relationship between the drawing point and the pixel to generate a high-quality image. You can Further, the present invention can be applied not only to the mapping device but also to a general rendering process for generating a computer image, and the same effect can be obtained.

[Brief description of drawings]

FIG. 1 is a configuration diagram of a mapping device of the present invention.

FIG. 2 is a block diagram of a rendering processor of the mapping apparatus of the present invention.

FIG. 3 is a configuration diagram of an edge generation unit of the mapping device of the present invention.

FIG. 4 is a configuration diagram of a correction unit of the mapping device of the present invention.

FIG. 5 is a configuration diagram of a span generation unit of the mapping device of the present invention.

FIG. 6 is a diagram showing displacement of a texture due to a quantization error at a value x.

FIG. 7 is a configuration diagram of an address generation unit of the mapping device according to the first embodiment of the present invention.

FIG. 8 is an explanatory diagram of a parameter correction process according to the first embodiment of the present invention.

FIG. 9 is a configuration diagram of a correction unit 103a according to the first embodiment.

FIG. 10 is a configuration diagram of a texture mapping device of the present invention.

FIG. 11 is a configuration diagram of an illuminance mapping device of the present invention.

FIG. 12 is an explanatory diagram of a parameter correction process according to the second embodiment of the present invention.

FIG. 13 is a configuration diagram of a correction unit 103a in the second embodiment.

FIG. 14 is a configuration diagram of a mapping device according to a third embodiment of the present invention.

FIG. 15 is an explanatory diagram of pixel correction processing according to the third embodiment of the present invention.

FIG. 16 is an enlarged view of a shared area according to the third embodiment.

FIG. 17 is a generated image storage unit 5 according to the third embodiment.
It is a block diagram of 05.

FIG. 18 is a configuration diagram of an address generating unit of a mapping device according to a fourth embodiment of the present invention.

FIG. 19 is an explanatory diagram of parameter correction processing according to the fourth embodiment of the present invention.

FIG. 20 is an explanatory diagram of parameter correction processing according to the fourth embodiment of the present invention.

FIG. 21 is an explanatory diagram of a processing outline in a first edge generation unit 702a, a second edge generation unit 702b, and a selection unit 703.

FIG. 22 is a configuration diagram of a first edge generation unit 702a.

FIG. 23 is a configuration diagram of a selection unit 703.

FIG. 24 is a flowchart showing an outline of processing according to the fourth embodiment of the present invention.

FIG. 25 is a diagram of target polygons according to the fourth embodiment of the present invention.

FIG. 26 is a configuration diagram of an address generation unit of a mapping device according to a fifth embodiment of the present invention.

FIG. 27 is an explanatory diagram illustrating a flow of processing according to the fifth embodiment of the present invention.

28 is a configuration diagram of an edge generation unit 2502. FIG.

FIG. 29 is a configuration diagram of an increment switching unit 2503.

FIG. 30 is a configuration diagram of an address generator of a mapping device according to a sixth embodiment of the present invention.

[Fig. 31] Fig. 31 is a diagram showing a polygon targeted in the sixth embodiment of the present invention and transitions of increment processing and correction processing.

FIG. 32 is a diagram illustrating a processing flow in the sixth embodiment of the present invention.

FIG. 33 is a configuration diagram of an edge generation unit 2502.

FIG. 34 is a configuration diagram of a correction unit 2503.

FIG. 35 is a configuration diagram of an address generator of a mapping device according to a seventh embodiment of the present invention.

FIG. 36 is a configuration diagram of a span generation unit 3404.

[Fig. 37] Fig. 37 is a diagram illustrating polygons targeted in the description of the mapping process by the incremental algorithm.

FIG. 38 is an explanatory diagram of corresponding point calculation processing of texture mapping by the incremental algorithm.

FIG. 39 is an explanatory diagram showing pixel drawing points and ideal pixel sample points by the incremental algorithm.

FIG. 40 is a diagram showing an example of image quality deterioration in mapping processing and an ideal image.

[Explanation of symbols]

101 arithmetic unit 102a edge generation unit 103a correction unit 104 span generation unit 201 pixels on screen 202 polygon edge 203 edge drawing point generated by the edge generation unit 102a 204 sample points (X, Y) 205 of the pixel 201 correction unit 103a (X ', y') 401 for explaining the correction processing of the pixel 401 pixel on the screen 402 polygon edge 403 edge drawing point generated by the edge generation unit 102a 404 sample point (X, Y) 501 calculation unit of the pixel 201 502 edge generation unit 503 span generation unit 504 original image storage unit 505 generated image storage unit 506 image storage processing unit 601a screen pixel a 601b screen pixel b 601c screen pixel c 601d screen pixel d 602 polygon edge 603 drawing point 604a screen Sample point 605 texture pixels 606a texture pixel 605 sample points 604d screen pixel 601d of the sample point 604c screen pixel 601c of sample points 604b screen pixel 601b of over emission pixels 601a and screen pixel 60
1a shared portion Sa 606b texture pixel 605 and screen pixel 60
1b shared portion Sb 606c texture pixel 605 and screen pixel 60
1c Shared portion Sc 606d Texture pixel 605 and screen pixel 60
1d shared portion Sd 701 operation portion 702a first edge generation portion 702b first edge generation portion 703 selection portion 704 span generation portion 801 polygon edge 802a drawing point P0 802b drawing point P1 802c drawing point P2 802d drawing point P3 802e Drawing point P4 803a Screen pixel 803b Screen pixel 803c Screen pixel 803d Screen pixel 803e Screen pixel 804a Pixel 803a sample point 804b Pixel 803b sample point 804c Pixel 803c sample point 804d Pixel 803d sample point 804e Pixel 803a sample point Pixel 803b from the sample point of 803a
From the sample point of the pixel 803b to the pixel 803c
From the sample point of pixel 803c to pixel 803d
From the sample point of the pixel 803d to the pixel 803e
Changes to sample points 901 sample points to the left 902 sample points to the right 903 change from sample point 804a to sample point 901 change 904 from sample point 804a to sample point 902 1001 edge of polygon to draw 1002 polygon edge point 1003 Drawing pixel of polygon edge point 1002 1004 Polygon edge point 1005 Drawing pixel of polygon edge point 1004 1006 Polygon span point 1007 Drawing pixel of polygon span point 1006 1108 Difference between polygon span point 1002 and ideal sample point (L1) 1109 Difference between polygon span point 1004 and ideal sample point (L2) 1110 Difference between polygon edge point 1006 and ideal sample point (L3) 1301 Calculation of differential value of y and partial differential value of x Processing by unit 701) 1302 Output processing of drawing points (x, y) and corresponding points (u, v) (processing by span generation unit 704) 1303 Processing of drawing points in the span direction and corresponding points (span generation unit) 704) 1304 Termination condition for processing 1303 (span generation unit 70
4 processing end condition) 1305 Edge drawing point calculation processing (edge generation unit 7)
02a and edge generation unit 702b common processing) 1306 selection reference numerical value calculation processing (processing by the selection unit 703) 1306 selection of du / dy (L) and du / dy (R) (processing by the selection unit 703) 1307 Corresponding point calculation process (edge generation unit 702a) 1308 Corresponding point calculation process (edge generation unit 702b) 1309 End condition of the entire process 1401 Vertex 0 (8,0,0,0) 1402 Vertex 1 (0,6, 0,16) 1403 Vertex 2 (11,9,16,16) 1404 Side 01 1405 Side 12 1406 Side 20 1601 First multiplier 1602 First adder 1603 Second multiplier 1604 Second adder 1701 Address generation unit 1702 Texture memory 1703 Frame memory 1801 Illuminance calculation unit 1803 Frame memory 1901 Multiplier 1902 Adder 2101 Multiplier 2102 Multiplier 103 Multiplier 2104 Multiplier 2105 Multiplier 2106 Multiplier 2107 Multiplier 2108 Multiplier 2201 Process performed by first edge generation unit 702a 2202 Process performed by second edge generation unit 702b 2203 Performed by first edge generation unit 702a Process 2204 Process performed by second edge generation unit 702b 2205 Process performed by first edge generation unit 702a 2206 Process performed by second edge generation unit 702b 2207 Process performed by first edge generation unit 702a 2208 Second edge Processing performed by the generation unit 702b 2211 Processing performed by the selection unit 703 2212 Processing performed by the selection unit 703 2213 Processing performed by the selection unit 703 2214 Processing performed by the selection unit 703 2221 Selection processing performed by the selection unit 703 2222 Performed by the selection unit 703 Selection processing 2223 Selection processing performed by the selection unit 703 2224 Selection processing performed by the selection unit 703 2301a x value selector 2301b y value selector 2301c u value selector 2301d v value selector 2302a x value adder 2302b y value addition 2302c u-value adder 2302d v-value adder 2401 Cumulative adder 2402a x-value selector 2402c u-value selector 2402d v-value selector 2502 Edge generation unit 2503 Incremental switching unit 2601 Edge generation unit 2502 processing 2602 Edge Process performed by generation unit 2502 2603 Process performed by edge generation unit 2502 2604 Process performed by edge generation unit 2502 2611 Process performed by increment switching unit 2503 2612 Process performed by increment switching unit 2503 261 Incremental switching unit 2503 processing 2614 Incremental switching unit 2503 processing 2701a Incremental value (dx / dy) selector 2701c Incremental value (du / dy) selector 2701d Incremental value (dv / dy) selector 2702a x Incremental processing Adder 2702by performing increment processing adder 2702c u performing increment processing adder 2702d v performing increment processing adder 2703a x value selector 2703by y value selector 2703cu u value selector 2703d v value Selector 2801 Cumulative adder 2901 Operation unit 2902 Edge generation unit 2903 Correction unit 3005b Transition of increment processing in the edge generation unit 2902 3005d Transition of increment processing in the edge generation unit 2902 3001 Transition of correction performed in the correction processing 2903 3002 Transition of correction performed in correction processing 2903 3101 Processing in edge generation unit 2902 3102 Processing in edge generation unit 2902 3103 Processing in edge generation unit 2902 3104 Processing in edge generation unit 2902 3111 Processing performed in correction unit 2903 3112 Correction Processing performed by the unit 2903 3113 processing performed by the correction unit 2903 3114 processing performed by the correction unit 2903 3122 correction processing performed by the correction unit 2903 3124 correction processing performed by the correction unit 2903 3201a x increment processing of the adder 3201by Adder 3201c u Adder to perform incremental processing 3201d v Adder to perform incremental processing 3202a x value selector 3202b y value selector 3202c u value selector 3202d v value selector 3301a Adder for correction processing for x 3301c Adder for correction processing for u 3301d v Adder for correction processing for v 3303a Non-correction / correction selector for x value 3303c u Non-correction / correction selector for value 3303d v Value Non-correction / correction selector 3404 Span generation unit 3501a Output buffer for x value 3501b Output buffer for y value 3501c Output buffer for u value 3501d Output buffer for v value

Claims (27)

[Claims] 1. A rendering calculation is performed based on a means for determining a pixel including a drawing point on a polygon projected on a plane having a plurality of pixels, and a sample point having a fixed position with respect to the pixel. A rendering device having means for performing and outputting a calculation result. 2. A means for performing rendering calculation based on a drawing point of a polygon projected on a plane having a plurality of pixels and outputting a calculation result, and a means for determining a plurality of pixels corresponding to the drawing point. And means for determining a weighting coefficient for each of the corresponding plurality of pixels based on the position of the drawing point, and based on the weighting coefficient determined for each of the corresponding plurality of pixels And a means for allocating the calculation result to each of the corresponding plurality of pixels. 3. An arithmetic unit for generating a parameter corresponding to a polygon based on the positions of the vertices of the polygon projected on a plane having a plurality of pixels and the corresponding points given to the polygon vertices. An edge generation unit that determines a position of a first drawing point on the edge of the polygon and a position of a first corresponding point corresponding to the first drawing point based on a parameter; and the first drawing A correction unit that determines a drawing pixel corresponding to the point, further determines a sample point corresponding to the drawing pixel, and corrects the position of the first corresponding point based on the sample point, the parameter, and the sample point And a span generation unit that determines the position of the second drawing point inside the polygon based on the above, and the position of the second corresponding point corresponding to the second drawing point. 4. The mapping device according to claim 3, wherein the correction unit performs one-dimensional correction on the first drawing point. 5. The edge generation unit first determines a point where an edge of the polygon intersects with an upper edge of the drawing pixel.
4. The mapping device according to claim 3, wherein the drawing point is the drawing point, and the correction unit sets the point closest to the origin in the drawing pixel as the sample point. 6. The mapping device according to claim 5, wherein the bit precision expressing the parameter is set to a predetermined value, and the correction unit includes a shifter and an adder. 7. An arithmetic unit that generates a parameter corresponding to a polygon based on the positions of the vertices of the polygon projected on a plane having a plurality of pixels and the corresponding points given to the vertices of the polygon, An edge generation unit that determines a position of a first drawing point on the edge of the polygon and a first corresponding point corresponding to the first drawing point based on the parameter, the parameter, and the first drawing The position of the second drawing point, which is a point inside the polygon, is determined based on the point and
Span generating section for determining the position of the second corresponding point corresponding to the drawing point, the generated image storing section for storing the values corresponding to each of the plurality of pixels on the plane, and the first drawing point A pixel storage processing unit that stores a position or a pixel value generated based on the position of the second drawing point as a value corresponding to one pixel of the generated image storage unit or a plurality of adjacent pixels. Equipped mapping device. 8. The pixel storage processing unit includes one pixel on the plane or an adjacent pixel of an area of a region on the plane defined by a position of the first drawing point or the second drawing point. An area ratio calculation unit that distributes the generated pixel value to one pixel or a plurality of adjacent pixels on the plane based on a ratio of the area occupied by a plurality of pixels to each of the distributed values. The mapping device according to claim 7, further comprising: a pixel value distribution unit that stores one pixel of the generated image storage unit or a value corresponding to a plurality of adjacent pixels. 9. The area ratio calculation unit defines (1-x) and (1-y) when the position of the first drawing point or the second drawing point is (x, y). A first multiplier for multiplying, a second multiplier for multiplying x and (1-y), a third multiplier for multiplying (1-x) and y, and x and y A fourth multiplier for multiplying, wherein the pixel value distribution unit has a fifth multiplier for multiplying an output from the first multiplier by the generated pixel value, and a second multiplier A sixth multiplier that multiplies the output from the multiplier and the generated pixel value, a seventh multiplier that multiplies the output from the third multiplier and the generated pixel value, The mapping device according to claim 8, further comprising an eighth multiplier for multiplying an output from the multiplier of No. 4 and the generated pixel value. 10. The bit precision expressing the position (x, y) of the first drawing point or the second drawing point or the generated pixel value is set to a predetermined value, and the first to the first The mapping device of claim 9, wherein each of the eight multipliers includes a shifter and an adder. 11. The edge generation unit defines, as the first drawing point, a point where an edge of the polygon intersects with an upper edge of the corresponding pixel, and thereby the second multiplier and the fourth drawing point. The mapping device according to claim 9, wherein a multiplier, the sixth multiplier, and the eighth multiplier are omitted. 12. An arithmetic unit that generates a parameter corresponding to a polygon based on the positions of the vertices of the polygon projected on a plane having a plurality of pixels and the corresponding points given to the polygon vertices. The first on the edge of the polygon based on the parameters
A first edge generation unit that determines a position of a first candidate point corresponding to the drawing point of the first polygon and a position of a first corresponding point corresponding to the first candidate point, and the polygon based on the parameter. First on the edge of
Of the first edge generation unit, and a second edge generation unit that determines the position of the second candidate point corresponding to the drawing point of the second position and the position of the second corresponding point corresponding to the second candidate point. A selection unit that selects one of the output and the output of the second edge generation unit, and the position of the second drawing point inside the polygon based on the parameter and the output selected by the selection unit. And a span generation unit that determines the position of a third corresponding point corresponding to the second drawing point. 13. The mapping device according to claim 12, wherein while either one of the first edge generator and the second edge generator is operating, the other edge generator is not operating. 14. The calculating unit calculates the inclination of the edge of the polygon on the plane, the selecting unit cumulatively adds the decimal parts of the inclination, and determines whether the result of the cumulative addition exceeds a predetermined value. The mapping device according to claim 12, wherein one of the output of the first edge generation unit and the output of the second edge generation unit is selected according to whether or not. 15. An operation for generating at least two sets of parameters corresponding to a polygon based on the positions of the vertices of the polygon projected on a plane having a plurality of pixels and the corresponding points given to the vertices of the polygon. A unit, an increment switching unit that selects one set of the at least two sets of parameters, and a first drawing point on the edge of the polygon based on the selected one set of parameters. On the basis of the position of the sample point, the edge generation unit that determines the position of the first corresponding point corresponding to the first drawing point, the selected set of parameters and the position of the sample point, A mapping device, comprising: a span generation unit that determines a position of a second drawing point inside a polygon and a position of a second corresponding point corresponding to the second drawing point. 16. The calculation unit calculates a slope of an edge of the polygon on the plane, and the increment switching unit cumulatively adds the fractional parts of the slope, and whether the cumulative addition result exceeds a predetermined value. 16. The mapping device according to claim 15, wherein one set of parameters is selected from the at least two sets of parameters according to whether or not the mapping is performed. 17. An arithmetic unit that generates a parameter corresponding to a polygon based on the positions of the vertices of the polygon projected on a plane having a plurality of pixels and the corresponding points given to the polygon vertices, An edge generation unit that determines a position of a first corresponding point corresponding to a first drawing point on the edge of the polygon and a position of a sample point corresponding to the first drawing point based on the parameter; A correction unit that corrects the position of the sample point when the distance between the first drawing point and the sample point exceeds a predetermined value, and based on the parameter and the sample point, A mapping device, comprising: a span generation unit that determines positions of two drawing points and positions of second corresponding points corresponding to the second drawing points. 18. The correction unit is included in the span generation unit, and the span generation unit corrects the position of the sample point, the position of the second drawing point, and the second drawing point. Perform both the process of determining the position of the corresponding points,
The mapping device according to claim 17. 19. The calculation unit calculates the inclination of an edge of the polygon on the plane, the span generation unit cumulatively adds the fractional parts of the inclination, and whether the result of the cumulative addition exceeds a predetermined value. The mapping device according to claim 17, wherein it is determined whether or not to correct the position of the sample point according to whether or not to correct. 20. The polygon vertex has an attribute value for expressing a material of a polygon, and values corresponding to the first corresponding point and the second corresponding point are set based on the attribute value. 20. The mapping device according to claim 3, further comprising a generating unit. 21. The mapping device according to claim 3, wherein the polygon vertex has a coordinate value of bump or displacement, and further has means for calculating a bump or displacement inside the polygon. . 22. The mapping device according to claim 3, further comprising means for performing an anti-aliasing process. 23. Each of the plurality of pixels on the plane has a plurality of corresponding sub-pixels, and the first drawing point is determined corresponding to sub-pixel positioning. The mapping device according to claim 6 or any one of claims 12 to 19. 24. A step of determining a pixel including a drawing point on a polygon projected on a plane having a plurality of pixels, and a rendering calculation based on a sample point having a fixed position with respect to the pixel. A rendering method including performing and outputting a calculation result. 25. A step of performing a rendering calculation based on a drawing point of a polygon projected on a plane having a plurality of pixels and outputting the calculation result; and a step of determining a plurality of pixels corresponding to the drawing point. And a step of determining a weighting coefficient for each of the plurality of corresponding pixels based on the position of the drawing point, and a step of determining a weighting coefficient for each of the plurality of corresponding pixels And distributing the calculation result to each of the corresponding plurality of pixels. 26. A step of generating a parameter corresponding to the polygon based on the positions of the vertices of the polygon projected onto a plane having a plurality of pixels and the corresponding points given to the polygon vertices; Determining the position of the first drawing point on the edge of the polygon and the position of the first corresponding point corresponding to the first drawing point based on Determining a drawing pixel to be performed, further determining a sample point corresponding to the drawing pixel, and correcting the position of the first corresponding point based on the sample point, and based on the parameter and the sample point Determining a position of a second drawing point inside the polygon and a position of a second corresponding point corresponding to the second drawing point. 27. A step of generating a parameter corresponding to the polygon based on a position of a vertex of the polygon projected on a plane having a plurality of pixels and a corresponding point given to the polygon vertex, and the parameter. Determining the position of the first drawing point on the edge of the polygon and the first corresponding point corresponding to the first drawing point, based on the parameter, and the parameter and the first drawing point. The position of the second drawing point, which is a point inside the polygon, is determined based on the
Determining the position of the second corresponding point corresponding to the drawing point, and the value generated based on the position of the first drawing point or the position of the second drawing point on the plane. One of the plurality of pixels
And a value corresponding to one pixel or a plurality of adjacent pixels.