Computer graphics processing and selective visual display system – Computer graphics processing – Graph generating
Reexamination Certificate
1998-01-08
2001-02-06
Zimmerman, Mark (Department: 2772)
Computer graphics processing and selective visual display system
Computer graphics processing
Graph generating
C345S428000, C345S475000
Reexamination Certificate
active
06184893
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to the field of computer controlled graphics display systems. More specifically, the present invention relates to computer controlled graphics display systems utilizing texture mapping.
BACKGROUND OF THE INVENTION
Computer controlled graphics systems are used for displaying graphics objects on a display. These graphics objects are composed of graphics primitive elements (“graphics primitives”) that include points, lines, polygons, etc. The graphics primitives can be used to render a 2 dimensional (2D) image of a three dimensional (3D) object on a display screen. In the process of rendering 3D graphics, many techniques are used to create realistic 3D effects. Some of these techniques involve Gouruad shading, texture mapping, bilinear filtering, specular lighting and fogging effects. Texture mapping refers to techniques for adding surface detail to areas or surfaces of these 3D graphics objects displayed on a 2D display. Often texture map information is added to displayed polygons.
Generally, texture mapping occurs by accessing encoded surface detail points or “texels” from a texel map memory space (“texel map”) which stores the surface detail and transferring the surface detail texels to predetermined points of the graphics primitive to be texture mapped. The process of determining the texels that correspond to pixels is called sampling the texture map. The texture image within a texture map is represented in computer memory as a bitmap or other raster-based encoded format. Texels reside in a (u, v) texture coordinate space. However, the display screen includes point elements (pixels) which reside in an (x, y) display coordinate space. Therefore, texture mapping applies color or visual attributes of texels of the (u, v) texture map to corresponding pixels of the graphics object (primitive) on the display screen. Color values for pixels in (x, y) display coordinate space are determined based on sampled texture map values.
After texture mapping, a version of the texture image is visible on surfaces of the graphics primitive. Because the original graphics object is 3D, texture mapping often involves maintaining certain perspective attributes with respect to the surface detail added to the object. Therefore, the manner in which a texture map is sampled, e.g., by du and dv values, is different depending on the perspective and size of the polygon.
FIG. 1A
illustrates a rendering situation in which a relatively large area texel map memory space
10
is used to provide surface detail to a relatively small area polygon
20
of a display screen. In this example, the texel map
10
is in a raster format and contains 512 texels by 512 texels while polygon
20
may contain as few as 10 pixels total. During rendering, because the polygon
20
is relatively small, the texels of the texel map
10
are sampled over large distances to provide the color for the pixels of the displayed polygon
20
. This rendering situation typically causes “flickering,” also called “sparkling,” within the polygon which is shown as lines
25
. Sparkling is caused because the rendering process samples different sets of texels from the texel map
10
for successive update frames of the display screen as the polygon
20
may slightly change its perspective between frames. Because the texels are sampled at large distances from each other, any small change in the sample process can cause very differently colored texels to be obtained from frame to frame; this is especially true for texel maps having high color frequency. Displaying the different texel sets causes an animated effect of sparkling within the polygon
20
. It is desirable to reduce sparkling within graphics systems to provide more realistic 3D effects. As described further below, mipmapping has been used to reduce sparkling.
FIG. 1B
illustrates a rendering situation in which a relatively small area texel map memory space
30
is used to provide surface detail to a larger area polygon
40
of a display screen. In this example, the texel map
30
is in a raster format and contains
10
texels by
10
texels while polygon
40
may have 10 times the area of the texel map
30
. Although this situation does not lead to sparkling, as described above, it does lead to “blocky” image quality within polygon
40
reducing its realism. During rendering, because the polygon
40
is so much larger than the texel map
30
, the texels of the texel map
30
are subsampled and the result is blocky surface detail
42
displayed within the large polygon
40
. As described further below, bilinear filtering has been used to reduce the blocky image quality described above.
In the prior art, the technique of mipmapping has been used to address the problem of sparkling when rendering a polygon in the rendering situation of FIG.
1
A. An example of mipmapping is shown in FIG.
2
. In effect, mipmapping requires that several different texel maps (called “mipmaps”)
51
-
54
be associated for each texel map
50
. The mipmap used depends on the size of the polygon to be rendered. The #0 or original mipmap
50
is the original texel map. Each successive mipmap is then ¼ the size of its predecessor mipmap. Therefore, #1 mipmap
51
is ¼ the area of mipmap
50
; #2 mipmap
52
is ¼ the area of mipmap
51
; #3 mipmap
53
is ¼ the area of mipmap
52
; and #4 mipmap
54
is ¼ the area of mipmap
53
. As such, mipmap
52
is {fraction (1/16)} the size of original mipmap
50
, mipmap
53
is {fraction (1/64)} the size of original mipmap
50
, and mipmap
54
is {fraction (1/256)} the size of original mipmap
50
. In mipmapping, depending on the relative size of the polygon to be rendered, a different mipmap of
50
-
54
is selected to perform texel mapping for the polygon. By using a smaller sized mipmap for a small polygon, the distance over which texels are sampled during rendering is reduced thereby reducing sparkling.
Although effective for reducing sparkling in the rendering situation of
FIG. 1A
, mipmapping has disadvantages. Each mipmap of
50
-
54
needs to be separately generated by a designer and this generation process is typically manual and/or involves complex and time consuming generation procedures. The mipmaps have to be generated for each original texel map. Further, because multiple mipmaps are required for each original texel map, additional computer memory is consumed (e.g., 33% more memory is required for each texel map) when implementing mipmapping techniques. Therefore, mipmapping restricts the number of unique texel maps that can reside within a graphics computer system. It would be advantageous to provide a technique for reducing or eliminating sparkling without the increased memory consumed by mipmapping techniques.
Bilinear filtering has been used in the past to reduce the blocky appearances of surface detail as described with reference to the rendering situation of FIG.
1
B. Bilinear filtering is used as a step in the process of applying texture to polygons. As described below, bilinear filtering involves an average of the nearest neighbor texels of the sampled texels before rendering the polygon. By performing color averaging, the sharp boundaries with blocky appearances, as shown in
FIG. 1
B, are reduced or eliminated. Bilinear filtering has not been applied to reduce sparkling in the rendering situation of FIG.
1
A.
FIG. 3
illustrates bilinear filtering with an exemplary texel map
60
having a red color portion
60
a
, a green color portion
60
b
and a blue color portion
60
c
. Within texel map
60
, three texels are sampled, texel
61
, texel
64
and texel
66
, to be used in the rendering of respective pixels
73
a
,
73
b
and
73
c
of displayed polygon
71
. In bilinear filtering, the color of texel
61
and its nearest neighbors (e.g., adjacent texels)
62
a
-
62
d
are averaged together to obtain the texel color for respective pixel
73
a
. Next, the color of texel
66
and its nearest neighbors
65
a
-
65
d
are averaged to obtain the
Devic Goran
Shaw Christopher W.
Cirrus Logic Inc.
Lin Steven
Murabito Anthony C.
Sealey Lance W.
Violette J. P.
LandOfFree
Method and system for filtering texture map data for... 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 system for filtering texture map data for..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and system for filtering texture map data for... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2615056