Computer graphics processing and selective visual display system – Computer graphics processing – Three-dimension
Reexamination Certificate
1998-05-21
2004-08-10
Luu, Matthew (Department: 2672)
Computer graphics processing and selective visual display system
Computer graphics processing
Three-dimension
C345S419000, C345S582000, C345S589000, C345S592000
Reexamination Certificate
active
06774896
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a method for generating an image for display on a graphic computer, entertainment device, video device, etc., and more particularly to a method and apparatus for generating, by shading, an image having a stereoscopic or three-dimensional vision.
2. Description of Related Art
For generating a game displayed on a graphic computer, computer-game device, video device or the like, a method called “texture mapping” is known by which an image is generated by deforming a texture and pasting it on a three-dimensionally defined object.
FIG. 1
shows the texture mapping method. As shown, a rectangular texture
101
is pasted on a spherical region defined as a three-dimensional object to generate an image of a sphere
102
.
Although the object image thus generated is defined three-dimensionally, its stereoscopic vision is poor since the image is only a monotonous figure displayed two-dimensionally. For easier recognition of the shape and spatial position of a object when displayed, it has been proposed to shade an object by irradiating to the object a light from a virtual source, thereby allowing the object to have a three-dimensional vision.
For such a shading, a variety of models different in type of virtual light source and reflection at object surface from one another has been proposed. Typical ones of the shading models are a perfect scattered reflection model and a specular reflection model which will be described herebelow.
FIG. 2
shows a perfect scattered reflection model for shading a sphere
102
.
It is assumed in the perfect scattered reflection model that a virtual light source emits parallel rays L having only directions and colors and the surface of an object is rough and has only colors, not any certain reflecting direction depending upon the position of the light source.
Assume here that the parallel rays L have colors Lr, Lg and Lb and directions Lx, Ly and Lz and that a point P on the surface of an object have colors Rt, Gt and Bt (texture information) and normals N to the point P are Nx, Ny and Nz. Then, colors of reflected rays of light from the point P on the object surface are given by the following equations (1). Note that a shift and clamp for a calculation using a fixed point are omitted.
R=Rt*Lr
*(
Lx*Nx+Ly*Ny+Lz*Nz
)
G=Gt*Lg
*(
Lx*Nx+Ly*Ny+Lz*Nz
) (1)
B=Bt*Lb
*(
Lx*Nx+Ly*Ny+Lz*Nz
)
Namely, in the perfect scattered reflection model, the reflected rays of light from a point on the surface of an object depend upon the direction and colors of rays from a light source and normals to and colors of a point on the object surface, independently of the position of a viewer.
Next, the specular reflection model will be described below.
FIG. 3
shows a specular reflection model for shading the sphere
102
.
It is assumed in the specular reflection model that a virtual light source is a point source H having a position and colors and rays of light are diverged in a same manner in all directions from the source H. Also, the surface of an object is smooth like a mirror and reflected rays of light travel in directions symmetrical with respect to a normal N to a point P on the object.
Assume here that a straight line PS connecting the view point S of a viewer and the point P on the object forms an angle &thgr; with reflected rays of light OR. Then, colors of reflected rays of light from the point P on the object surface are given by the following equations (2).
R=Hr
*(cos &thgr;)
n
G=Hg
*(cos &thgr;)
n
(2)
B=Hb
*(cos &thgr;)
n
where n is a value depending upon a material of the object. Note that the larger the value n, the smoother the object surface is and the larger the light reflecting area of the object is.
Namely, in the specular reflection model, the reflection at the object surface varies depending upon a viewer's position and is independent of colors (texture information) of a point on the object.
A combination of the aforementioned perfect scattered reflection model and specular reflection model will be described below.
FIG. 4
shows a combination of the aforementioned two types of reflection models to shade a sphere
102
with rays of light from parallel rays L and point source H.
The combination of the two types of reflection model provides an object shading more approximate to a real shading of an actual object than use of one model of reflection model, and a stereoscopic or three-dimensional vision of the object. A calculation of reflected rays of light for shading of an object using the combination of two types of reflection model is done with a following set of expressions:
R=Rt*Lr
*(
Lx*Nx+Ly*Ny+Lz*Nz
)+
Hr
*(cos &thgr;)
n
G=Gt*Lg
*(
Lx*Nx+Ly*Ny+Lz*Nz
)+
Hg
*(cos &thgr;)
n
(3)
B=Bt*Lb
*(
Lx*Nx+Ly*Ny+Lz*Nz
)+
Hb
*(cos &thgr;)
n
The first term of each relation in (3) indicates a reflection by the perfect scattered reflection model, and the second term indicates a reflection by the specular reflection model.
However, a vast volume of calculation will be required for obtaining reflected rays of light for all points on an object by using the set of expressions (3). Therefore, it is impossible as a matter of fact to effect the calculation at a high speed. For solution of this problem, it has been proposed to divide the surface of an object into a plurality of small polygonal areas and effect the above-mentioned calculation with only their vertexes of the polygonal areas, thereby reducing the volume of calculation.
FIG. 5
shows how to divide a spherical surface into small triangles for the above-mentioned calculation.
As seen, colors of all pixels included in a triangle thus defined through the division of the object surface are calculated with a linear interpolation using colors calculated with three vertexes. This method is called “glow shading”.
FIG. 6
shows how to calculate all colors of pixels included in each triangle on the object surface. As shown, a triangle
3
a
of which a color (texture information) of a point on the surface is given, and another triangle
3
b
of which a color change information is given, are used in a linear interpolation to provide an triangle
3
c
of which colors of all pixels are calculated.
Generally, such an interpolation is done by a graphic processing unit (GPU) (will be referred to as “graphic processor” hereafter), and the results of the interpolative calculation are written into an image memory.
The graphic processor is a hardware which uses three vertexes of a triangle, color information of the three vertexes and texture information to effect a linear interpolation of the color information, interpolation of the texture, and a multiplication of them.
Each of the color information and texture information is generally composed of four color components R, G, B and A. The color components R, G and B are color information indicative of red, green and blue, respectively. The component A is an information used for a translucent plotting and generally called “alpha channel”. Generally, the graphic processor comprises a color information interpolation circuit and a texture mapping circuit provided for each of the four color components R, G, B and A.
FIG. 7
shows the configuration of a conventional graphic processor.
As shown, the graphic processor comprises a texture mapping circuit
11
, and a color information interpolation circuit
12
. The texture mapping circuit
11
is provided to use supplied texture information Rt, Gt, Bt and At and coordinates of three vertexes of a triangle to effect a texture mapping. The color information interpolation circuit
12
uses the three-vertex color information of the triangle and the coordinates of the three vertexes to effect an interpolation of color information.
An output from the texture mapping circuit
11
and a one from the color information interpolation circuit
12
are supplied to a modulation circuit
13
and
Chung Daniel J
Fulwider Patton Lee & Utecht LLP
Luu Matthew
Sony Computer Entertainment Inc.
LandOfFree
Method and apparatus for generating image data by modulating... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for generating image data by modulating..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for generating image data by modulating... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3312581