Computer graphics processing and selective visual display system – Computer graphics processing – Three-dimension
Reexamination Certificate
2000-07-14
2002-12-03
Vo, Cliff N. (Department: 2671)
Computer graphics processing and selective visual display system
Computer graphics processing
Three-dimension
Reexamination Certificate
active
06489958
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to computer systems and in particular to graphic processing systems which graphically represent or render on a display surface objects defined in an object coordinate system.
BACKGROUND OF THE INVENTION
Methods for rendering object coordinate system data into screen coordinate system data (screen space data) have been known for some time. A comprehensive description of a known method can be found in the book entitled “Computer Graphics, (Pinciples and Practice”, by Foley, van Dam, Feiner, Hughes, published by Addison-Wesley Publishing Company, ISBN 0-201-84840-6. Normally objects are defined by triangles in an object coordinate system or object space. A triangle is defined by its three vertices P
1
, P
2
, P
3
. In the case of a three-dimensional graphic application, each point in the three-dimensional coordinate system requires three coordinate values. The origin of the three-dimensional rectangular coordinate system in the object space or object coordinate system can be chosen at random. If the object to be represented is a sphere, for example, the origin of the object coordinate system is normally located at the centre of the sphere. The representation in the object coordinate system is thus defined from the outset and is independent of an observer standpoint.
An observer should now be able to move through space so as to be able to observe the surfaces defined in the object coordinate system, and thus objects, from his particular standpoint. A simple example should make this clear. Consider a house which has been specified in object coordinates. An observer should now be able to approach this house from e.g. the garden gate, in which case it is clear that he can then only see the facade but not any objects inside the house. Furthermore the observer can only see one side of the house wall, the visible side of the house call being of course the outer side if he is outside the house and the inner wall of a room if he is inside the house. The object coordinate system origin could e.g. be located on the ground floor, so that the cellar of the house will have negative y-coordinates in the object coordinate system while the first floor will have positive y-coordinates. It is therefore necessary to process the polygons, and thus triangles, specified “absolutely” in the object coordinate system in order to represent only what the observer sees on a screen. In addition it is necessary to light or shade the objects in the house in three dimensions. While the light source, e.g. the sun or a lamp in a room, may here be fixed absolutely in object coordinates, the observer will see different lit objects depending on his standpoint.
DESCRIPTION OF PRIOR ART
FIG. 4
shows a flowchart of a known method for rendering object coordinate system data so as to be able to display them on a monitor. Such a method is shown in J. D. Foley, A. Van Dam et al., “Computer Graphics”, Addison-Wesley, 1996, FIG. 18.3.3.
In a first step
100
the object coordinate system data are transformed into an eye coordinate system or eye space, in which necessary lighting calculations
200
are performed. After the lighting
200
in the eye coordinate system a perspective projection is preferably performed so as to obtain clip coordinates or clip-space coordinates from the original object coordinates. After clipping
300
in the clip coordinate system, the homogeneous clip coordinates are dehomogenized.
In a first step
100
the object coordinate system data are transformed into an eye coordinate system or eye space, in which necessary lighting calculations
200
are performed. After the lighting
200
in the eye coordinate system a perspective projection is preferably performed so as to obtain clip coordinates or clip-space coordinates from the original object coordinates. After clipping
300
in the clip coordinate system, the homogeneous clip coordinates are dehomogenized.
The position of a triangle in the object coordinate system is normally represented redundantly in that the three coordinates of each vertex of the triangle in the coordinate system each has a respective scaling factor assigned to it. In accordance with the technical terminology, an unambiguous representation of points is termed an inhomogeneous representation, whereas an overdetermined or redundant representation of points is termed a homogeneous representation of these points. If a point on a two-dimensional surface is represented by means of two coordinates, this constitutes an inhomogeneous representation of the point. If a point on a two-dimensional surface is represented by means of three coordinates, this representation is overdetermined and is termed a homogeneous representation. By analogy thereto, a point in the three-dimensional coordinate system can be represented in an inhomogeneous way by its x-, y- and z-coordinates if the three-dimensional space is defined by a Cartesian coordinate system. A homogeneous representation of the point in the three-dimensional space is achieved by specifying
4
“coordinates” for the point in the three-dimensional space. The fourth coordinate w is by definition a scaling factor. A homogeneous x-, y-, z-coordinate is converted into an inhomogeneous x-, y- or z-coordinate by dividing by the scaling factor w. If a point in the space has inhomogeneous coordinates (1, 2, 3), its homogeneous representation could be (1, 2, 3, 1) or (2, 4, 6, 2), etc.
The advantage of the homogeneous representation of a point in the space is that all transformations which normally occur in a three-dimensional graphic processing or graphic pipeline (e.g. a translation, a rotation, a scaling and in particular a perspective projection) can be represented as a multiplication of the position of the point with a 4×4 transformation matrix. The coordinates used in the steps
100
,
200
and
300
are homogeneous coordinates, which are converted into inhomogeneous coordinates in the dehomogenization step
400
. This dehomogenization is effected in a simple way by dividing the homogeneous x-, y-, z-values by the corresponding scaling factor w of this point. After dehomogenization, the talk is of coordinates in the screen coordinate system or screen space. The step
500
is then performed in the screen coordinate system. In this step triangles which do not have the desired orientation are culled. If two-sided lighting is employed, triangles with undesired orientation are not culled but are simply lit differently. The step
500
thus determines the type of lighting to be used for a triangle. In the case of a house where the outer wall is also the inner wall of a room, the outer wall is lit e.g. in yellow, if the house is painted yellow, while the inner wall of the room is lit up in white. Thus if an observer is outside the house, the graphic rendering method determines that only triangular surfaces which represent outer walls, i.e. which have a particular orientation, are to be displayed, while inner walls, i.e. triangular surfaces with opposite orientation, can be ignored. If there is only one type of lighting, the triangular surfaces whose orientation corresponds by definition to an interior orientation, are culled or, in the case of two-sided lighting, are provided with a type of lighting which discloses that what is involved here is a triangular surface which should describe the outer wall of a house.
It should be pointed out here that in the step
200
in the known method all the triangles are lit or both types of lighting are calculated for all triangles in the case of two-sided lighting, although half the triangles are culled again in the step
500
or one of the two types of lighting of a triangle is ignored when the triangles are subsequently displayed on a screen in a step
600
.
As an alternative to the prior art described in
FIG. 4
, triangles can also be lit directly in the object coordinate system, whereupon the lit object coordinate surfaces are transformed directly into the clip coordinate system (clip space) without involving the eye coordinate system. This approa
Patton & Boggs LLP
SP3D Chip Design GmbH
Vo Cliff N.
LandOfFree
Method and device for graphic representation of an object... 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 device for graphic representation of an object..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and device for graphic representation of an object... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2988096