Computer graphics processing and selective visual display system – Computer graphics processing – Three-dimension
Reexamination Certificate
2000-04-03
2004-05-04
Zimmerman, Mark (Department: 2671)
Computer graphics processing and selective visual display system
Computer graphics processing
Three-dimension
C345S426000, C345S582000, C345S589000, C345S609000
Reexamination Certificate
active
06731281
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to a method for the two-dimensional pixel representation of objects on a display device based on selectable objects whose respective surfaces are divided into a plurality of polygonal surface elements.
RELATED TECHNOLOGY
The most common arrangements for the two-dimensional representation of objects, in particular for the representation of moving objects and/or representations from different viewing angles or on a variable scale, are those that include a data processing system and a display device controlled by the data processing system. The display device is typically a color monitor with a defined pixel raster.
To control the monitor by pixels, the data processing system (DPS) ordinarily includes a main processor, frequently referred to as the central processing unit (CPU) or simply the processor of the data processing system, as well as an image generating unit, which is designed specifically for generating pixel values and controlling the monitor and is connected to the CPU by an internal data line of the data processing system, in particular an internal bus system. An image generating unit of this type is typically designed as a graphics card that is inserted into a slot of the bus system.
Graphics cards of this type, in turn, contain memory and computing arrangements, for example another processor having multiple arithmetic units, which, however, unlike the general-purpose central processing unit (CPU), can be oriented toward and optimized for the specific image generation tasks. For example, graphics cards are commonly used which automatically generate a two dimensional representation of a three dimensional object that can be defined as a polyhedron with a specific spatial orientation and color information for the individual surface elements according to a lighting model, which, if necessary, is selected from a plurality of lighting models, and can be varied within specific limits via parameters, by means of a processing program provided in the graphics card, with the individual surfaces or surface elements being shaded in accordance with the relative orientation of the model lighting and surface elements of the object. Ordinarily, the reason for transferring processing steps from the CPU to specialized hardware components is to relieve the load on the CPU.
It has been demonstrated, particularly when varying the object representation, that computation-intensive processing operations on the graphics card as well as the amount of data to be transferred from the CPU to the graphics card over the bus system prolong the time needed to set up a new image, even in the case of limiting this data to the object data describing the polyhedron framework, making it difficult to display rapid variations or causing them to appear as sequences of jerky movements.
SUMMARY OF THE INVENTION
An object of the present invention is to provide a method for the two-dimensional pixel representation of objects on a display device which makes it possible to shorten the necessary image setup time and thus shorten the refresh time between modified image representations.
The present invention provides a method for the two-dimensional representation of objects on a display device, the display device being controlled by an image generating unit which, in turn, receives image data from a central processing unit over a data line, and, based on selectable objects whose respective surfaces are divided into a plurality of polygonal surface elements and based on a selectable lighting model, pixel values for controlling the display device being generated by performing arithmetic operations in the central processing unit and in the image generating unit. The vertices of the surface elements and pixel values for the vertices are determined in the central processing unit on the basis of the surface elements of the object and the corresponding lighting model. Coordinates and pixel values of the vertices are transferred to the image generating unit over the data line, and the pixel values of the object pixels reaching the display are generated in the image generating unit by interpolation between the pixel values of the vertices.
The present invention specifically selects a certain distribution of processing steps between the central processing unit (referred to below as CPU for short) and the image generating unit (referred to below as graphics card) and thus certain image data for transmission from the CPU to the graphics card.
The present invention speeds up the image setup time, thus shortening the image refresh times, by reducing the volume of data transferred over the internal bus system as well as by considerably reducing the computing power needed in the graphics card. The latter is achieved, in particular, by transferring to the CPU the measurement of the backscattered luminosity of the individual surface elements according to a selected lighting model (including their immediate periphery without using the bus system). The functions on the graphics card are then largely limited to transforming the three-dimensional coordinates of the surface element vertices to two-dimensional coordinates of the pixel display and interpolating the pixel values within the individual surface elements. These operations require only a small amount of computing power, making it possible to process a large volume of data per time unit in these processing steps. If different interpolation methods are available on the graphics card, an additional parameter to select an interpolation method may also be necessary.
An increase in processing speed can also be achieved by disabling the processing stage provided in the graphics card for calculating energy values based on a lighting model, which can preferably be accomplished with a control command, such as “glDISABLE (GL_LIGHTNING)” in the OpenGL programming language commonly used for hardware pipelines. According to one particularly advantageous embodiment of the present invention, the processing devices (memory, arithmetic units) provided in the graphics card for calculating energy values are also used for interpolation between the vertex pixel values along with the devices normally provided for interpolation within a surface element. An energy value is a quantitative measure of the intensity or luminosity of a pixel. The energy values are preferably measured in the form of value triplets for the three basic colors (red, green, and blue) and output to the display device for pixel by pixel control. The pixel values of the vertices are also advantageously transmitted as energy values for the three basic colors during transmission from the CPU to the graphics card.
Although data processing load is transferred from the graphics card to the CPU according to the present invention, this places very little extra load on the CPU when displaying an object composed of a plurality of surface elements. This is due, in particular, to the fact that a lighting model for representing an object is typically selected in the beginning, and the pixel values are determined at the vertices, as well as the fact that subsequent variations in the representation involve only varying the object viewing direction in the case of the object-related static lighting model and/or changing the scale during representation. In these operations, which appear in the monitor display in the form of a rotation or the object being zoomed in or out, respectively, and which can also be visually perceived as a variation in camera position with the object remaining stationary, the energy values advantageously remain unchanged at the vertices, while only the vertex coordinates vary. Fast algorithms, for example those that include a matrix multiplication, are available for coordinate transformations of this type, ensuring fast conversion with very little additional load on the CPU. On the whole, it is even possible to greatly reduce processing load, since sufficient storage space can be provided in the CPU memory periphery for all vertices of all surface elements and the energy
Avanion GmbH
Kenyon & Kenyon
Nguyen Kimbinh T.
Zimmerman Mark
LandOfFree
Method for two-dimensional pixel representation of 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 for two-dimensional pixel representation of objects..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for two-dimensional pixel representation of objects... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3193154