JPH11144082A - Image generating device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To delete coordinate transformation and luminance calculation by frame by saving polygon information and image by polygon unit and separating image generation into a polygon for which new luminance calculation is performed and a polygon for deforming the saved polygon image at new time. SOLUTION: A coordinate transformation part 11 performs transformation to a view point coordinate system, a three-dimensional image coordinate system by perspective transformation, and finally a screen coordinate system to calculate a polygonal region, which is saved in a polygon information saving part 12. Then, a luminance calculation part 13 calculates, for example, the luminance of each vertex of the polygon based on the distance from a light source and finds the pixels in the polygon by interpolating the luminance of three vertexes. Then, a movement quantity decision part 15 decides whether or not the movement quantity of the polygonal region is within the range of specified movement quantities. Then the movement quantities of all calculated polygons are decided at specific time to update the screen buffer 18 by luminance calculation or image deformation accompanying the update of the polygonal information.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an image generating apparatus for generating a high-speed and high-quality image in the field of three-dimensional computer graphics (three-dimensional CG).

[0002]

2. Description of the Related Art In recent years, image generation apparatuses have been developed for use in the movie and commercial industries due to improvements in the performance of CPUs (central processing units) and development of dedicated graphics chips. It is also being generalized.

An example of the above-described conventional image generating apparatus will be described below with reference to the drawings.

FIG. 8 is a block diagram showing a configuration of a conventional image generating apparatus. In FIG. 8, 81 is a coordinate conversion unit, 82 is a luminance calculation unit, and 83 is a screen buffer unit.
The operation of the image generating apparatus configured as described above will be described below.

First, the coordinate conversion section 81 performs a series of coordinate conversions described below. That is, a virtual three-dimensional object, which is a set of polygonal shapes defined in a unique object coordinate system, is arranged in the world coordinate system and subjected to modeling transformation. Next, when a virtual viewpoint is set, the viewpoint is converted into a viewpoint coordinate system having the viewpoint as the origin, and after the perspective transformation, a portion outside the field of view is removed by clipping and converted into a screen coordinate system. The brightness calculation unit 82 obtains the brightness of each pixel in the screen from the brightness value at the vertex of the polygon by linear interpolation or the like, and when attaching a texture, interpolates the texture coordinates of each pixel from the texture coordinates of the vertex. Then, the texture value at that coordinate is referenced from the texture memory. The screen buffer unit 83 selects a polygon closest to the viewpoint position on a pixel-by-pixel basis based on the depth value from the viewpoint, and writes the luminance value into a monitor output buffer.

[0006]

However, in the above configuration, drawing accompanying movement of a virtual viewpoint or a three-dimensional polygon involves coordinate conversion from the world coordinate system to the screen coordinate system and pixel calculation for obtaining luminance. Is calculated every frame for all polygons, so that the calculation cost is high.

SUMMARY OF THE INVENTION In view of the above problems, an object of the present invention is to provide an image generating apparatus which uses polygon information and polygon image data as intermediate data and reduces the operation cost in consideration of coherency of a moving image. I do.

[0008]

In order to solve the above-mentioned problems, an image generating apparatus according to the present invention uses a position and a velocity vector of a virtual viewpoint in a virtual space at a certain time to calculate a polygonal motion vector and a polygon. A coordinate conversion unit that calculates polygon information including a region in a screen coordinate system; a polygon information storage unit that stores polygon information; a brightness calculation unit that calculates the brightness of each pixel of the polygon region; and a brightness calculation unit Image storage unit that stores the image calculated for each polygon, and whether the movement amount of the polygon area calculated from the polygon information and the elapsed time α is within a specified range or not. And, if the movement amount is out of the specified range, the updated polygon region and the elapsed time α are updated from the polygon information to the polygon information storage unit as the update time, and the brightness of the updated polygon region is calculated. Store the result in the polygon image storage A polygon information updating unit to be stored, and an image deforming unit that deforms the polygon image in the polygon image storage unit to calculate an image after time α.

Thus, the polygon information and the image in units of polygons are stored, and the image generation at the new time is separated into a polygon for performing a new luminance calculation and a polygon for deforming the stored polygon image. By doing so, it is possible to reduce coordinate conversion and luminance calculation for each frame.

[0010]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention relates to a coordinate system for calculating polygon information comprising a polygon-shaped motion vector and a polygon region from a position and a velocity vector of a virtual viewpoint in a virtual space at time t in a screen coordinate system. A conversion unit, a polygon information storage unit that stores the polygon information, a luminance calculation unit that calculates the luminance of each pixel in the polygon area, and an image calculated by the luminance calculation unit that is stored for each polygon. A polygon image storage unit, and a movement amount discrimination unit that determines whether or not the movement amount of the polygon area calculated from the polygon information and the elapsed time α at time t + α is within a specified range. If the movement amount is out of the specified range, the updated polygon area and the elapsed time α are updated from the polygon information as the update time in the polygon information storage unit, and the brightness calculation result of the updated polygon area is calculated. For polygon image storage And polygon information updating unit to exist, time by modifying the polygon image of the polygonal image storage unit
An image transformation unit that calculates an image at t + α, and a screen buffer unit that performs hidden surface elimination by comparing depth values on a pixel basis from the transformed image by the image transformation unit and stores each pixel of the screen.

In the present invention, with the above-described configuration, the polygon motion information and the polygon information including the polygon area in the screen coordinate system at time t are stored in the polygon information storage unit. Further, the brightness calculation unit calculates a brightness value of each pixel in the polygon by performing shading processing such as texture mapping or illuminance mapping based on the polygon area which is one of the polygon information. Each polygon is stored in the polygon image storage unit. At this time t, the polygon image whose luminance has been calculated is subjected to hidden surface elimination by comparison with a depth value that is a distance from the viewpoint, stored in a screen buffer, and displayed. On the other hand, at time t + α, the movement amount, which is the degree of change of the polygon, is calculated from the motion vector, which is the polygon information at time t, and the polygon area, and the movement amount determination unit determines whether the movement amount is within the specified range. Is done. If it is within the range, the image deforming unit maps the polygon image at time t to the updated polygon area after the change and writes it to the screen buffer. If it is out of the range, the updated polygon area and the elapsed time α are updated as polygon information, and the brightness calculation is performed on the updated polygon area to update the polygon image in the polygon image storage unit. I do. At this time, the updated polygon image is written to the screen buffer without performing the image deformation. Thus, by storing the polygon information and the image in units of polygons, and separating the image generation at the new time into a polygon for performing a new luminance calculation and a polygon for transforming the stored polygon image. Thus, coordinate transformation and luminance calculation for each frame can be reduced.

Hereinafter, embodiments of the present invention will be described with reference to the drawings. (Embodiment 1) FIG. 1 is a block diagram showing a configuration of an image generating apparatus according to Embodiment 1 of the present invention. In FIG. 1, reference numeral 11 denotes a coordinate conversion unit for converting a polygon shape in an object coordinate system into a screen coordinate system and calculating a polygon motion vector by moving a viewpoint or the like; and 12, a polygon position in the screen coordinate system. A polygon information storage unit for storing polygon information including a certain polygon region and a motion vector at the position of the polygon, a luminance calculation unit 13 for calculating the luminance of the pixels in the polygon region, and a luminance calculation unit 14 for calculating the luminance. A polygon image storage unit for storing the obtained polygon image, a movement amount for comparing a movement amount from the polygon region before movement obtained from the polygon information to the polygon region after movement with a predetermined value; A discriminating unit 16 is a polygon information updating unit that updates the polygon region after movement as a new polygon region for a polygon that requires luminance calculation again, and 17 is a polygon image before moving to the polygon region after movement. The image transformation unit that maps It is a surface buffer portion.

The operation of the image generating apparatus configured as described above will be described below with reference to FIGS.

Here, FIG. 3 shows a coordinate system in three-dimensional CG, and FIG. 4 shows an example of a polygon at time t in the screen and movement of the polygon at time t + α.

In the coordinate transformation unit 11 shown in FIG. 1, an object constituted by a polygon (a triangle in this example) is arranged in the world coordinate system as shown in FIG. The polygon area is calculated by converting the image into a three-dimensional screen coordinate system and finally to a screen (screen) coordinate system. Furthermore, a polygon motion vector is also calculated from the virtual viewpoint velocity vector v at time t using coordinate transformation, and the polygon area and the motion vector are stored in the polygon information storage unit 12 as polygon information.
Here, in this example, only the movement of the virtual viewpoint is treated,
The movement of the polygon itself is relative to the movement of the virtual viewpoint,
There is no problem with unification.

In FIG. 4, the position vector P of each vertex of the triangle is shown.
t, Qt, Rt and motion vectors Mp, Mq, Mr are shown.
Here, the position vector and the motion vector of each vertex for expressing the polygon area are three-dimensional vectors having elements in the depth direction. The brightness calculator 13 calculates, for example, the brightness of each vertex of the polygon from the distance from the light source, and obtains the pixels in the polygon by interpolating the brightness of the three vertices. When texture mapping is performed, the luminance values of the pixels inside the polygon are sampled and calculated from the texture data based on the correspondence between each vertex and the coordinate system unique to the texture data. The polygon image thus obtained is stored in the polygon image storage unit 14. Here, at time t, all the polygonal images have been calculated by the luminance calculation unit 13;
The image transforming unit 17 maps all polygonal images to the respective polygonal areas, that is, writes them in the screen buffer unit 18 without any modification.

Next, the operation at time t + α will be described. First, in FIG. 4, it is assumed that there are no motion vectors Mp, Mq, and Mr acceleration components. α time, Pt,
A triangle with three vertices Qt and Rt is Pt + α = Pt + α · Mp, Qt
+ Α = Qt + α · Mq, Rt + α = Rt + α · Mr Move to a triangle having vertices. The movement amount determination unit 15 determines whether the movement amount of the polygonal area is within the range of the specified movement amount.

FIG. 5 shows an example of a method for determining the amount of movement, which will be described with reference to FIG. In FIG. 5, 51
Is a region area calculation unit, 52 is an area comparison unit, 53 is a normal line calculation unit, 54 is a normal line comparison unit, and 55 is a logical sum unit that calculates the logical sum of the area comparison result and the normal line comparison result.

A triangle at time t with three vertices Pt, Qt, and Rt and a time at three vertices with Pt + α, Qt + α, and Rt + α
The area of the triangle at t + α is determined by the area calculation unit 51. The area comparison unit 52 calculates the rate of change of the area from the two areas, and outputs a logically positive determination result if smaller than the specified amount. At the same time, normal vectors of these triangles are obtained by a normal calculation unit 53, and an angle between the vectors is calculated by a normal comparison unit 54. If the angle is smaller than the specified amount, a logically positive determination result is output. I do. The logical sum unit 55 calculates the logical sum of these two determination results, and outputs that the result is within the range of the designated movement amount only when both are positive.

FIG. 6 is an example showing a remarkable difference between the normal vectors. Although both triangles in FIG. 6 have the same shape and the same area, the texture-mapped stripes differ by the perspective transformation. That is, the R point of the left triangle is
This is because the point R is closest to the right triangle in the right triangle, although it is farther from the viewpoint than the vertex. This can be determined by determining the angle between the normal vectors.

As described above, when the result of the movement amount discriminating section 15 is out of the specified movement amount range, the Pt + α, Qt of FIG.
+ Α, Rt + α as updated polygon area, and elapsed time α
Also, as the update time, the polygon information updating unit 16 writes in the polygon information storage unit 12. When the polygon information is changed, the brightness is calculated for the polygon area, and the polygon storage unit 14 is also updated. As in the case of the time t, the image deforming unit 17 writes the updated polygon image in the screen buffer unit 18 without performing the deformation.

On the other hand, when the result of the movement amount discriminating section 15 is within the specified movement amount range, the polygon information is not updated and the brightness calculation is not executed. Then, the image transformation unit 17 transforms the polygon image stored as the three vertices of Pt, Qt, and Rt in the polygon image storage unit 14 into a polygon area of Pt + α, Qt + α, and Rt + α, and writes the polygon area into the screen buffer 18.

As described above, for all the polygons calculated at time t, the movement amount is determined at time t + α.
The screen buffer 18 is updated by brightness calculation or image deformation accompanying the update of the polygon information. Here, the screen buffer 18 stores a polygon area of the polygon information, for example, Pt, Qt, Rt.
Further, the hidden value is removed by calculating the depth value of each pixel using Pt + α, Qt + α, and Rt + α.

At the next time t + β, in addition to the case where the polygonal image at the time t is image-deformed and the case where the luminance calculation is performed, the polygonal image updated after the luminance calculation at the time t + α is further processed. May be deformed. in this case,
The update time α of the polygon information is used. For example, Pt + β =
Pt + β · Mp = Pt + α + (β−α) · Mp, and Pt + α and α will be used.

The motion vectors Mp, Mq, Mr originally change over time, and the elapsed times α, β
Is a sufficiently small value that the above linear approximation holds. That is, when the elapsed time from the time t increases, the coordinate conversion unit 11 newly calculates polygon information.

As described above, according to the present embodiment, the time
a coordinate conversion unit 11 for calculating a polygonal motion vector and polygonal information comprising a polygonal area in a screen coordinate system from the position and the velocity vector of the virtual viewpoint in the virtual space at t;
A polygon information storage unit 12 for storing polygon information, a luminance calculation unit 13 for calculating the luminance of each pixel in the polygon area, and a polygon image for storing the image calculated by the luminance calculation unit 13 for each polygon A storage unit 14, a movement amount determination unit 15 that determines whether the movement amount of the polygon area calculated from the polygon information and the elapsed time α at time t + α is within a specified range, and a movement amount Is outside the specified range, the updated polygon area and the elapsed time α are updated from the polygon information to the polygon information storage unit 12 as the update time, and the brightness calculation result of the updated polygon area is stored in the polygon image storage unit 14. A polygon information updating unit 16 to be stored, and an image deforming unit 17 for deforming the polygon image in the polygon image storing unit 14 to calculate an image at time t + α
Is provided, the coordinate transformation associated with the movement of the virtual viewpoint is reduced by using the motion vector of the virtual viewpoint, and the polygon image is reused to reduce complicated luminance calculation, thereby shortening the calculation time. be able to.

(Embodiment 2) Hereinafter, Embodiment 2 of the present invention will be described with reference to the drawings.

FIG. 2 is a block diagram showing a configuration of the image generating apparatus according to the present embodiment. In FIG.
Reference numeral 17 is the same as the configuration in FIG. 1 in that a convex object division unit 21 and a multilayer screen buffer unit 28 are provided.

FIG. 7 is a diagram showing the internal configuration of the multilayer screen buffer unit 28. In FIG. 7, reference numeral 71 denotes a multilayer buffer unit that is written to a layer for each convex object, 72 denotes a depth table in which a depth comparison result between layers is written, 73 denotes a depth determination unit that performs depth determination between layers, and 74 denotes a depth comparison result. And a multi-layer combining unit that performs the combining process and outputs the result.

The operation of the image synthesizing apparatus configured as described above will be described below. The convex object dividing unit 21 re-divides an object having irregularities into convex objects. The coordinate conversion unit 11 converts the polygon inside the convex object into a screen coordinate system. Hereinafter, a polygon image is calculated in the same manner as in the first embodiment, and is written into one layer in the multilayer buffer unit 71 in the multilayer screen buffer unit 28 for each convex object. When the writing to the multilayer buffer unit 71 is completed, the depth determination unit 73 refers to the depth table 72 and determines the depth of the pixel to be written. To determine the depth, first scan the pixels in the multilayer buffer,
The pixel on which the convex object overlaps and the layer number are detected. If the detected context is described in the depth table 72, the scanning is continued. If there is no description, which polygon is overlapped is detected, and polygon areas of the polygon information, for example, Pt, Qt, Rt, Pt + α, Qt +
The depth value of the detection pixel is calculated using α and Rt + α, and the depth relation of the overlapping layers is written in the depth table. If the depth table 72 describes the relationship between all layers, the scanning is stopped. Otherwise, the scanning is continued. Also, the scanning is stopped at the time when the scanning of the last pixel is completed.

As described above, the convex object division unit 21 that divides a plurality of virtual three-dimensional objects each having a polygonal shape into convex objects and saves the positions and speeds of the virtual viewpoint in the virtual space at time t. A coordinate conversion unit 11 for calculating, in a screen coordinate system, polygon information composed of a polygon-shaped motion vector and a polygon area constituting a convex object from a vector, and a polygon information storage unit 12 for storing polygon information A brightness calculation unit 13 for calculating the brightness of each pixel in the polygon area, a polygon image storage unit 14 for storing the image calculated by the brightness calculation unit 13 for each polygon, and polygon information at time t + α. A movement amount determining unit 15 for determining whether the movement amount of the polygonal area calculated from the elapsed time α is within a specified range or not; Update polygon area and meridian The polygon information storage unit 12 updates the polygon information storage unit 12 with the time α as an update time, and stores the brightness calculation result of the updated polygon area in the polygon image storage unit 14. An image transformation unit 17 for transforming the rectangular image to calculate an image at time t + α,
The first embodiment is provided by storing the image calculated by the brightness calculation unit 13 in one layer for each convex object, comparing the depth of each layer, synthesizing the image, and outputting the combined image. In addition to the effect of the above, it is not necessary to perform the depth comparison in pixel units for all pixels.

[0032]

As described above, according to the present invention, from the position and the velocity vector of the virtual viewpoint in the virtual space at the time t, the polygon information composed of the polygon-shaped motion vector and the polygon area is calculated in the screen coordinate system. Coordinate conversion unit, a polygon information storage unit for storing polygon information, a brightness calculation unit for calculating the brightness of each pixel in the polygon area, and an image calculated by the brightness calculation unit is stored for each polygon. Polygon image storage and time
At t + α, a movement amount determination unit that determines whether the movement amount of the polygonal area calculated from the polygon information and the elapsed time α is within a specified range, and if the movement amount is outside the specified range. A polygon information update unit that updates the updated polygon area and the elapsed time α from the polygon information as the update time in the polygon information storage unit, and stores the brightness calculation result of the updated polygon area in the polygon image storage unit; By providing an image transformation unit that computes an image at time t + α by transforming a polygon image in the polygon image storage unit, the polygon information and the image in units of polygons are saved, and an image is generated at a new time. Can be separated into a polygon for performing a new luminance calculation and a polygon for deforming the stored polygon image, so that coordinate transformation and luminance calculation for each frame can be reduced.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of an image generation device according to a first embodiment of the present invention.

FIG. 2 is a block diagram illustrating a configuration of an image generation device according to a second embodiment of the present invention.

FIG. 3 is an explanatory diagram relating to a coordinate system in image generation.

FIG. 4 is an explanatory diagram showing a temporal change of a polygon in a screen coordinate system.

FIG. 5 is a block diagram showing a method of determining a moving amount according to the present invention.

FIG. 6 is a diagram showing an example of a remarkable difference between normal vectors in moving amount discrimination.

FIG. 7 is a block diagram illustrating an internal configuration of a multilayer screen buffer unit according to a second embodiment of the present invention.

FIG. 8 is a block diagram showing a schematic configuration of a conventional image generating apparatus.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 11 Coordinate conversion part 12 Polygon information storage part 13 Brightness calculation part 14 Polygon image storage part 15 Moving amount discrimination part 16 Polygon information update part 17 Image transformation part 18 Screen buffer part 21 Convex object division part 28 Multilayer screen buffer part

Claims (6)

[Claims] 1. A coordinate conversion unit for calculating, from a position and a velocity vector of a virtual viewpoint in a virtual space at time t, polygon information including a polygon-shaped motion vector and a polygon area in a screen coordinate system, A polygon information storage unit for storing polygon information, a brightness calculation unit for calculating the brightness of each pixel in the polygon area, and a polygon image storage unit for storing an image calculated by the brightness calculation unit for each polygon At a time t + α, a movement amount discrimination unit that determines whether or not the movement amount of the polygonal area calculated from the polygon information and the elapsed time α is within a specified range; If it is out of the range, the updated polygon area and the elapsed time α are updated from the polygon information as the update time in the polygon information storage unit, and the brightness calculation result of the updated polygon area is stored in the polygon image storage unit. Polygon to save An information updating unit, the time to deform the polygonal image of the polygonal image storage unit t +
An image generation unit comprising: an image deformation unit that calculates an image at α; and a screen buffer unit that performs hidden surface removal by comparing depth values in units of pixels from the deformed image by the image deformation unit and saves each pixel of a screen. apparatus. 2. A method according to claim 1, wherein a plurality of virtual three-dimensional objects each having a polygonal shape are divided into convex objects, and the convex object dividing unit divides and stores the convex three-dimensional objects. A coordinate conversion unit that calculates polygon information composed of a polygon-shaped motion vector and a polygon area constituting a convex object in a screen coordinate system; a polygon information storage unit that stores the polygon information; A brightness calculation unit that calculates the brightness of each pixel in the polygonal area; a polygon image storage unit that stores the image calculated by the brightness calculation unit for each polygon; and at time t + α, the polygon information and the elapsed time α. A moving amount determining unit that determines whether or not the calculated moving amount of the polygonal area is within a specified range; and, if the moving amount is outside the specified range, updating polygons from the polygon information. Area and elapsed time α A polygon information updating unit that updates the polygon information storage unit as a new time, and stores the brightness calculation result of the updated polygon area in the polygon image storage unit, and a polygon image of the polygon image storage unit. Deformed and time t +
an image transformation unit for calculating an image with α, a multi-layer screen for storing the image calculated by the luminance calculation unit in one layer for each of the convex objects, comparing depths on a layer-by-layer basis, synthesizing and displaying and outputting the images An image generation device comprising a buffer unit. 3. The moving amount determining unit includes a region area calculating unit that calculates the area of the polygon region at time t and the updated polygon region at time t + α, and an area comparing unit that calculates the area ratio. The image generation device according to claim 1 or 2, wherein 4. A moving amount determining unit for calculating a normal vector between a polygon region at a time t and an updated polygon region at a time t + α, and a method for calculating an angle between the normal vectors. 2. The apparatus according to claim 1, further comprising a line comparison unit.
Or the image generation device according to 2. 5. A movement amount determination unit, comprising: a region area calculation unit that calculates an area of a polygonal region at time t and an updated polygonal region at time t + α; an area comparison unit that calculates an area ratio; A normal line calculation unit that calculates a normal vector between the polygon region and the updated polygon region at time t + α, and a normal line comparison unit that calculates an angle between the normal vectors. 3. The image generation device according to 1 or 2. 6. The multi-layer screen buffer unit detects a depth table indicating an inter-layer relation between layers, detects a pixel on which a convex object between layers overlaps during a display scan, and describes the inter-layer relation between the target layers in the depth table. 3. A depth judging section for judging a front-back relationship between layers from polygon information and writing the depth table in the depth table when there is no polygon information.
An image generation device according to claim 1.