Computer graphics processing and selective visual display system – Computer graphics processing – Attributes
Reexamination Certificate
1998-03-11
2002-08-06
Zimmerman, Mark (Department: 2671)
Computer graphics processing and selective visual display system
Computer graphics processing
Attributes
C345S426000
Reexamination Certificate
active
06429872
ABSTRACT:
BACKGROUND OF THE INVENTION
The invention relates to a method for representing a computer-modeled object, as defined in the preamble to claim
6
, and an apparatus for executing the method, as defined in the preamble to claim
6
.
In computer-graphics systems, bodies are usually simulated with grid models that are described by the spatial coordinates of the node points. To represent the polygon surfaces lying between the node points on a screen, the spatial coordinates of the node points are then converted from the three-dimensional coordinate system, for example through central projection, into a two-dimensional screen-coordinate system.
The image impression of the individual polygon surfaces is calculated in perspective, with the position and orientation of the polygon surface relative to the light sources and the predetermined observer's location being considered in order to achieve the most natural image impression possible.
It is also known that the image impression should not be calculated all at once for the entire polygon surface, but the individual polygon surfaces should be divided into a plurality of pixels, and an image impression should be calculated individually for each pixel, so even a relatively crude grid model having large polygon surfaces can be used to attain a natural image impression. To this end, so-called shading methods are used to calculate the shading—that is, the color or brightness gradation—within the polygon surface.
One shading method of this type is referred to as Phong shading. In this method, a data set is predetermined for each polygon surface; the local surface normal at the corner points of the polygon surface ensues from this set, so a curved polygon surface can also be realized, which further improves the degree of realism in the representation of an image. For each pixel within the polygon surface, the local surface normal resulting from the curvature of the polygon surface is interpolated from the predetermined corner-point normals. The image impression of the individual pixels is then calculated with the consideration of the local surface normal, according to a local illumination model.
While the above-described Phong shading advantageously permits a very natural image impression by taking into consideration the curvature of the individual polygon surfaces, the calculation is very complicated because the local illumination model for each pixel must be calculated.
An advantage of Phong shading, however, is the possibility of simulating a surface structure of the individual polygon surfaces without it being necessary to refine the grid model accordingly. For example, in representing a wood surface, it is possible to simulate the wood grain. For this purpose, a two-dimensional storage matrix (bump map) is provided, which defines the desired surface structure, with each storage location containing an inclination value that indicates how the local surface normal is inclined due to the surface structure; each pixel of the polygon surface is associated with a storage location of the storage matrix. In the calculation of the image impression of the individual pixels, first the associated storage location of the storage matrix is specified, and the inclination value representing the surface structure is read out. In the calculation of the image impression corresponding to the local illumination model, the interpolated local surface normal of the smooth polygon surface is not considered, but rather the local surface normal that is changed corresponding to the inclination value. This process utilizes the fact that the position of the pixel is not decisive for the image impression of a pixel of a relief-structured surface, but rather the change in the local surface normal caused by the surface structure is decisive.
However, the above-described simulation of a surface structure basically presupposes a shading method that interpolates the local surface normal individually for each pixel and calculates the image impression according to a local illumination model for each pixel, because the inclination of the local surface normal can only be considered in the calculation of the local illumination model. Up to now, therefore, so-called bump mapping has only been possible in connection with relatively complicated shading methods.
A significantly simpler shading method is known as Gouraud shading. In this instance, too, a data set is predetermined for each polygon surface, from which the spatial position of the local surface normals at the corner points of the polygon surface ensues. In contrast to the above-described Phong shading, however, the local illumination model is only calculated for the corner points of the polygon surface; in contrast, the image impression of the pixels within the polygon surface is interpolated from the image impression at the corner points of the polygon surface as a function of the position of the respective pixel within the polygon surface.
On the one hand, this saves calculation time in comparison to Phong shading, because the complicated calculation of the local illumination model is only effected at the corner points of the polygon surface, while the interpolation of the image impression for the individual pixels is considerably simpler in terms of the calculation effort.
On the other hand, the image impression in Gouraud shading is less natural than in the more complicated Phong shading, and it is not possible to simulate a surface structure (bump mapping) in the above-described manner.
SUMMARY OF THE INVENTION
It is therefore an object of the invention to provide a method of representing computer-modeled objects that also permits the simulation of a surface structure, with the least possible calculation effort, and without necessitating a corresponding refinement of the grid model. It is a further object of the invention to provide a corresponding apparatus for executing a method of this type.
These objects are accomplished by a method including the steps of: calculating at least one surface normal of a polygon surface from a first parameter set that predetermines the spatial position of individual polygon surfaces; calculating at least one global color value for the polygon surface as a function of the respective calculated surface normal; dividing the polygon surface into a plurality of pixels and calculating a respective coordinate set (X, Y, Z) which represents the spatial position of a respective pixel from the first parameter set; interpolating a first local color value for each individual pixel and the at least one calculated global color value of the polygon surface; individually addressing a first storage matrix with a plurality of storage locations for each individual pixel as a function of the respective coordinate set (X, Y, Z), each storage location of the first storage matrix containing an inclination value for simulating a surface structure of the polygon surface where the inclination value determines the inclination of the local surface normal due to the surface structure; reading out the inclination value for each individual pixel from the addressed storage location of the first storage matrix; calculating a second local color value that takes into consideration the influence of the surface structure on an image impression as a function of the read-out inclination value; and calculating a third local color value from the first local color value and the second local color value where the third local color value is the ultimate color value and serves in image representation for the individual pixels.
According to the invention, the apparatus for executing the method of representing a computer-modeled object that is simulated by a plurality of a adjacent polygon surfaces including an input for receiving the first parameter set, a second parameter set that determines a viewer's perspective, and an illumination-data set which determines spatial position and irradiation properties of at least one light source; a first assembly for calculating at least one surface normal of the polygon surface
GMD-Forschungszentrum Informationstechnik GmbH
Spencer George H.
Stevenson Philip H.
Venable
Voorhees Catherine M.
LandOfFree
Method and apparatus for representing computer-modeled objects 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 representing computer-modeled objects, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for representing computer-modeled objects will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2904698