MXPA00006830A - Image generation method and image generation device - Google Patents

Abstract

It is an object of the present invention to generate more realistic pattern by mapping a texture to a three-dimensional polygon. An image generation device (1) has image memory (7), in which are stored basic textures to be mapped to generate overall patterns on polygons by mapping and modulation textures with which amplitude modulation is applied to the patterns generated by mapping of the basic textures, and a pixel engine (6) which, by amplitude modulation mapping of modulation textures, performs amplitude modulation processing on the patterns generated by mapping of basic textures.

Description

METHOD FOR GENERATION OF IMAGE AND DEVICE FOR GENERATION OF IMAGE FIELD OF THE INVENTION This invention relates to a method for image generation and a device for image generation. Specifically, the present invention relates to a method for generating an image and a device for generating images that generate two-dimensional images starting from three-dimensional polygons and textures.

BACKGROUND OF THE INVENTION The surfaces of the objects that surround us often have repeated patterns of complex appearance, and the more complex and fine the appearance or pattern, the more difficult it is to model it with triangles. One technique to solve it is to make a texture map. Making texture maps produces very real images with a small number of vertices overlaying image data read by a browser on the surface of objects. On the other hand, in a graphics system, a complete three-dimensional image is drawn by dividing the three-dimensional image into triangles or other polygons (unit figures) and drawing these polygons.

In a graphics system that generates two-dimensional images from such polygons and three-dimensional textures, the drawing from the point of view to a distant point in the distance is obtained by making texture maps of land surfaces, water surfaces and floors. When it is intended to provide an image by making texture maps, the texture to which a map will be made in a manner close to the point of view is greatly enlarged, which tends to blur the image and greatly diminishes the effect of realism. One way to avoid this is to use a high density texture. But this method has the disadvantage that it wastes a large amount of texture memory. Also, there is a large decrease in the processing speed due to page breaks in the texture memory that is constituted with the usual type of dynamic RAM, because it is required to have access to a large address space when performing the texture maps. Another way to avoid this is to express textures with a combination of recurrent affine transformations using fraction compression technology, controlling the depth of the recurrent drawing in accordance with the precision required at the time of execution. However, this method has the drawback of needing a large amount of computing resources for recurring drawings, without having a great possible compression for the images that are not so recurrent. And with respect to the reduction or enlargement of an original image at different reduction rates, the MIPMAP technique, in which the image patterns are provided with reduced or enlarged textures, is superior in that it presents a low distortion of the image, and the Making texture maps can be done at high speed. But a problem that occurs with this MIPMAP technique is that the forms near the point of view are blurred.

BRIEF DESCRIPTION OF THE INVENTION One objective of the present invention, which was devised keeping in mind the situation described above, is to provide a method for image generation and a device for image generation that can generate more real patterns in two-dimensional images by means of mapping of texture to three-dimensional polygons. To solve the aforementioned problems, the method for the generation of image of this invention generates general patterns on polygons making maps of the basic textures, and performs an amplitude modulation processing on patterns generated by the realization of maps of basic textures, by means of the realization of amplitude modulation modulation modulation maps. That is, the method for generating image of the present invention by means of making modulation modulation amplitude maps of modulation textures, performs amplitude modulation processing in patterns generated by making basic texture maps. This method for generating images adequately generates the details of the patterns in the region near the point of view, ie foreground. Also to solve the aforementioned problems, the image generation device of the present invention has memory means that store basic textures to which maps are made to generate the general pattern on a polygon, and the modulation textures used to modulate by amplitude. the patterns generated by making maps of the basic textures, and an image processing means that, by making modulation maps by amplitude of the modulation textures, performs amplitude modulation processing in the patterns generated by the performance of maps of the basic textures. By making modulation maps by amplitude of the textures by modulation, an image generation device having such a composition performs, through its image processing means, the amplitude modulation processing on patterns generated by making maps of basic textures In this way, the image generating device suitably generates the details of the pattern in the region near the point of view.

BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a block diagram showing the composition of an image generation device in accordance with an embodiment of the present invention; Figure 2 is a block diagram showing the configuration of the image generation device together with the data flow; Figure 3 is a diagram showing the relationship between the modulation textures and the dynamic scale; Figure 4 is a diagram showing the case where the modulation texture is not a repeated texture; Figure 5 is a diagram showing the relationship of the MIP values between a basic texture and a modulation texture; Figure 6 is a diagram showing the basic texture of MY PO; Figure 7 is a diagram showing the modulation texture of MIPO; Figure 8 is a diagram showing the basic texture of MIP1; Figure 9 is a diagram showing the modulation texture of MIPI; Figure 10 is a diagram showing the basic texture of MIP2; Figure 11 is a diagram showing the modulation texture of MIP2; Figure 12 is a diagram showing the basic texture of MIP3; Figure 13 is a diagram showing the modulation texture of MIP3; Fig. 14 is a flow diagram showing the series of processing steps by which an image is generated using basic textures, showing the processing by which amplitude modulation by modulation textures is applied to the pattern of the generated image using basic textures; Fig. 15 is a flow diagram showing the series of processing steps by which amplitude modulation by modulation textures is applied to an image, showing the processing by which amplitude modulation by modulation texture is applied to the image pattern generated using basic textures; Figures 16A and 16B are diagrams showing an image generated by making texture maps to which the present invention is applied, and an image generated by making conventional texture maps; Figures 17A and 17B are diagrams showing another image generated by making texture maps to which this invention is applied, and an image generated by making conventional texture maps; and Figures 18A and 18B are diagrams showing the process of making texture maps.

DETAILED DESCRIPTION OF THE MODALITIES THAT ARE PREFER Next, an embodiment of the present invention will be described in detail with reference to the drawings. In this embodiment, the method for image generation and the image generating device of the present invention are applied to a device for image generation that generates two-dimensional images by making texture maps to three-dimensional polygons. For example, the device for image generation which is a mode made in such a way that it can be applied to a three-dimensional television game device, a three-dimensional moving image screen device and a three-dimensional moving image transfer device. As shown in Figure 1, a device for image generation 1 has a unit for calculating brightness and coordinate transformation 2, a LOD calculation unit (level of detail) 3, a texture coordinate calculation unit 4, a DDA (digital differential analysis) unit 5, a pixel tool 6, and an image memory In the image generation device 1, the image memory 7 is made of a memory medium in which the memories are stored. basic textures made in maps to generate general patterns on polygons, and modulation textures with which the amplitude modulation is applied to the patterns generated by making basic texture maps, and the pixel tool 6 is made from a medium of image processing that, by realization of modulation maps by amplitude of modulation textures, performs the amplitude modulation processing on patterns generated by making basic texture maps. Fig. 2 shows the configuration, including the data flow, within the image generation device 1 shown in Fig. 1. That is, the image generation device 1 shown in Fig. 2 which is arranged in FIG. such that it includes a mapping unit 21, an anti-aliasing unit 22, a hazy processing unit 23, a memory interface 24, and a PCRTC (programmable CRT controller) 25. It is also shown in FIG. 2 shows the preprocessing unit 2, 3, 4 which includes brightness calculation and coordinate transformation unit 2, a LOD calculation unit 3, and a texture coordinate calculation unit 4 of the device for image generation 1 shown in figure 1.

The parts constituting this device for generating image 1 are described in detail below. Diverse information is fed to the device for generating image 1 to generate three-dimensional images, which generates images by means of this diverse information. For example, the diverse information that is fed includes three-dimensional polygon vertex information, combination information, texture coordinate information, light source information, and point of view information. The device for generating image 1 obtains all this information from, for example, communication lines or memory devices, etc. For example, a description is given below for feeding polygons to the device for image generation 1, which, in general, consists of polygons but in this mode they are taken as independent triangles, to simplify the description. As the functions of the device system for image generation 1 can be applied to various primitives, including points and straight lines. The information of polygons is information consisting of the coordinates p1 (px1, p1, pz1), p2 (px2, p2, pz2), p3 (p3, p3, pz3) of the vertices of polygons that are independent triangles. And the information about the normals of these vertices is fed as n1 (nx1, nyl, nyl), n2 (nx2, ny2, nz2), n3 (nx3, ny3, nz3). The texture coordinate information is information that consists of the texture coordinates a1 (s1, t1, q1), a2 (s2, t2, q2), a3 (s3, t3, q3) of the vertices of polygons that are independent triangles . The combination information is a combination coefficient (combination coefficient a) that shows the combination location of the image when the textures are made on maps. Specifically, the combination coefficient a is set to a1, a2, a3 corresponding to RGB. This combination information together with the RGB values constitute the RGBA values. The information of the vision point (ojox, ojoy, ojoz) and the information of the light source (Ix, ly, Iz) is information used to perform the brightness calculations and the coordinate transformations with respect to the polygons. There may also be multiple points of light source information. Various additional information such as color information for cloudiness values can also be associated with each polygon vertex. Here, the value of nebulosity is a value that is used to perform image processing in such a way that objects in the distance assume a hazy color; this is a value that is used to generate images with an effect known as nebulous. The polygons of multiple triangles independent to which such information is added are fed into the device for image generation 1. The texture information, in which the pixels having values of RGBA are aligned in a group, is fed by coordinates of texture of the vertices of the polygon. Such polygon information is first fed to the brightness calculation and coordinate 2 transformation unit in the image generation device 1. The brightness calculation and coordinate transformation unit 2 takes the polygon information that has been fed and the combines with the point of view information to make a coordinate transformation to a coordinate system for drawing. Then the brightness calculation and coordinate transformation unit 2 calculates the brightness of each vertex of each polygon from the point of view information and the light source information. Together with the execution of the calculations described in previous paragraphs, the brightness calculation and coordinate transformation unit 2 performs a perspective transformation, etc. For example, by a coordinate transformation of a polygon of independent triangles, the coordinates of the vertices become q1 (qx1, qy1, iz1), q2 (qx2, qy2, iz2), q3 (qx3, qy3, iz3). Here, the coordinates qx, qy are the coordinates in the drawing screen and the coordinates iz are coordinates in the depth direction of the plane transformed into integers for the buffer z. The brightness of each vertex is determined from the normal vertex vector, the information of the editing point and the light source information.

The values calculated in this way by the brilliance calculation unit and coordinate transformation 2 are fed from the preprocessing unit 2, 3, 4 to the DDA unit as the gradient of the value XYZ, the gradient of the value of RGBA, and the gradient of the value of F consisting of the value of nebulosity, as shown in figure 2. Here, the value XYZ consists of the values of the X, Y, Z coordinates of each of the three vertices of the three-dimensional polygon; that is, it is established in points concerning the information, lines and the shape of the polygon, etc. The calculation unit LOD 3 calculates the value of LOD from the transformed coordinates z. Based on this LOD value, the pixel tool 6 selects a basic texture stored in the basic texture buffer 9. Here, the basic textures are textures that are mapped for three-dimensional polygons with the use of the MIPMAP method. The MIPMAP method is a method of making maps where the textures of different sizes 1/2, 1/4, 1/8, ... (length ratio of the sides) is prepared as a texture to be attached to the three-dimensional polygons , these prepared textures are chosen in accordance with the reduction rate and the realization of maps takes place in the three-dimensional polygons; for example, generation is performed for various textures using a low pass filter. In the MIPMAP method, adopting a mapping method like this prevents an overlap from occurring when the making of texture maps is done, because the mapping of polygons on the screen is done with the original texture with a size reduced. For example, the texture of each level (MIP level) can be obtained by applying a low pass filter to images that each have a low MIP value, and reducing to 1/2. In the following explanation for each type of texture, the same texture as the original image is called MIPO texture, the texture of 1/2 of the original image is called texture MIP1 and the texture of 1/4 of the original image is called MIP 2, and so on progressively. The numeric value that follows MIP represents the MIP level, and this corresponds to the LOD value. The LOD value is a value that is calculated from the reduction rate of each polygon, and a reduction rate is presented, for example, by the logarithm of the distance from the point of view to the polygon. The basic texture corresponding to the LOD value calculated by the calculation unit LOD 3 is read from the basic texture buffer 9 of the image memory 7. The texture coordinate calculation unit 4 calculates from the coordinate value of texture for the basic texture the texture coordinate value that is used for reading the modulation texture. As shown in Figure 2, the value calculated by the calculation in the texture coordinate calculation unit 4 is fed to the DDA unit 5 as the gradient of the UV value and the gradient of the STQ value of the preprocessing unit 2. , 3,4. Here, the UV value is the coordinate value of the texture, and the STQ value is the value of the texture coordinate in each of the three vertices of the polygon; that is, they consist of texture coordinate values of the same order (for perspective correction). The DDA unit converts bi-directional polygon vertex information, z information, and brightness information, etc. obtained from the texture coordinate calculation unit 4 in information in pixels. Specifically, the DDA unit 5 satisfactorily determines the coordinates in pixels (apx, apy), z-value (aiz), brightness and texture coordinate (as, at, aq) by linear interpolation. As shown in FIG. 2, the DDA unit 5 outputs processed XYZ value, F value, Rf value, Gf value, Bf value, Af value, STQ value and UV value as processed transformation values. Here, the XYZ value is fed into the pixel tool 6, the value F is fed into the haze unit 23 and the value Rf, value Gf, value Bf, value Af, value STQ and value UV are fed into the texture mapping unit 21. The anti-aliasing unit 22 , by processing the image in such a way that it is uniform by blurring the trimmed edges of the lines and images, performs the combination processing by using the value a from the texture mapping unit 21. The haze processing unit 23 is the part that performs the processing by a nebulosity effect using the value of nebulosity. Specifically, by using the value of F from the DDA unit 5, it carries out the processing on values in pixels emitted from the texture mapping unit 21. The texture mapping unit 21 is the part that develops the control of making texture maps based on diverse information. The PCRTC 25 is the part that emits the image signal that has passed through the desired processing to the monitor as analog RGB and digital RGB. The pixel tool 6 develops, as pixel operations on pixels, processes known as cut, test a, target test a, depth test, combination a, vibrational action and color fixation. Here, the cut is a processing by which the data that comes out of the screen is removed; The test a is a processing in which the drawing to be made is controlled by the value a of a pixel; The destination test a is a processing in which if the drawing is to be made it is controlled by the combination coefficient a of a pixel of the frame buffer that will be recorded in it; The depth test is a test performed by the buffer Z; The combination a is a processing in which the pixel value of the frame buffer and the value in pixels to be recorded are linearly interpolated by the combination coefficient a; The vibratory action is a processing in which the colors intermingle to produce many colors with a low number of colors; and Color fixation is a processing where, when calculating the colors, the value is restricted so that it does not exceed 255 or is not less than 0. Specifically, the pixel tool 6 performs the following processing: pixel tool 6 calculates the pixel information between the pixel information obtained from the basic texture buffer 9 and records it in the frame buffer 8. And when it registers in the frame buffer 8, as necessary, the pixel tool 6 controls the record by referring to buffer z. In addition, the pixel tool 6 has a bilinear interpolation function, whereby the values in pixels are determined by linear interpolation of four pixels in a lattice reading from the basic texture buffer 9, as well as a function by which the pixel values of the frames buffer 8 are obtained and the multiplication (modulation calculation) is performed between this value and the pixels of the modulation texture that are recorded. This is, for example, the pixel tool 6 has the function of making texture maps, z comparison, calculation in pixels, and the reading and recording in pixels of and towards the frames buffer 8, as well as performing the modulation, etc. The memory interface 24 shown in Figure 2 is an interface for transmitting data between the pixel tool 6 and the image memory 7. Specifically, this memory interface 24 makes it possible to send and receive XYZ values, A values ( coefficients a of combination) and values of RGB between the pixel tool 6 and the image memory 7. The image memory 7 has memory regions, i.e. intermediate memory of frames 8, where the frames are stored, the basic texture buffer 9, where the basic textures are stored, and the modulation texture buffer 10, wherein the modulation textures are stored. For example, as shown in Figure 2, the image memory 7 consists of a main memory 7a and a texture memory 7b here, the high-speed texture buffer 7b is constituted as a high-speed buffer used for have access to texture information at high speed. As stated above, the basic textures are textures that are generally used by MIPMAP. These basic textures are used to generate the general pattern of a texture with maps. For example, in the basic texture buffer 9 basic textures are stored in compressed form. Modulation textures are textures that are used to add high frequency components to basic textures. For example, a modulation texture is set to a texture consisting of higher frequency components by misaligning its repetition period of the basic pattern repetition period. As shown in Figure 3, the modulation textures are such that the dynamic scale of the modulation decreases when the MIP level increases. This is because it is desirable to work in the part where the scale of elongation of a basic texture is greater than 1. The value in pixels of a modulation texture represents the intensity to further develop the multiplication and apply the modulation to the values in pixels of an image drawn using a basic texture. For example, if a texture is expressed with 8 bits, the texture value 0 is associated with the multiplication coefficient 0.0, the texture value 128 with the multiplication coefficient 1.0, and the texture value 256 with the multiplication coefficient 2.0. . That is, for the images of each MIP level, not only a low pass filter reduction but also dynamic scale reduction processing is performed. As stated above, the repetition period of a modulation texture is misaligned with the repetition period of the basic texture, specifically, if the image size of the modulation texture is the same as that of the basic texture, the period of repetition is misaligned using for texture reading the modulation texture coordinates (s2i, t2i) which are the result of the misalignment of the texture coordinates (si.ti) of the basic texture multiplied several times. This relationship can be shown by formulas (1) and (2). s2i =? x if + ß ... (1) t2i =? x ti + ß ... (2) Here, (si, ti) are the original texture coordinates of the basic texture, and (s2i, t2i) are the texture coordinates used for reading the modulation texture. As well ? is an integer where? > 1, and ß is a constant. If the modulation texture is not a repeating texture, it will be necessary to change the way the values advance if the texture coordinate value exceeds 0 and 1. Because of the above, if someone denoted by (s3i, t3i) the coordinates Finally, a transformation is made from the texture coordinates (s2i, t2i) to the texture coordinates (s3i, t3¡) in accordance with a graph as shown in figure 4. And because to which the modulation texture is reduced to 1 /? with respect to the basic texture, it is necessary to give a polygon LOD value changed by log2 (?). For example, yes ? = 2, then the MIP levels of the basic textures and modulation textures used with respect to the LOD value of the polygon will be as shown in Figure 5. The basic textures mentioned in the previous paragraphs and the modulation textures are stored in the basic texture buffer 9 and the modulation texture buffer 10, respectively. Also, basic textures and modulation textures can be stored within the same buffer. For example, the basic buffer may be provided in the RGB part of the RGBA buffer, and the modulation textures may be in its part A. Specific examples of basic textures and modulation textures are shown in FIGS. The figures 6, 8, 10 and 12 show the basic textures of MIPO, MIP1, MIP2, and MIP3, respectively, while figures 7, 9, 11, and 13 show the modulation textures of MIPO, MIP1, MIP2 , And MIP3, respectively. Part of the MIPO texture in figures 6 and 7 is omitted from the diagram. As noted above, the dynamic scale of modulation is reduced when the MIP level increases; when the MIP level increases, a "blurry" feeling is obtained, as shown in figures 7, 9, 11, and 13. In fact, if the image size of the modulation texture is the same as the basic texture , then, as indicated above, mapping will be carried out at a higher frequency, in such a way that the modulation texture can be read based on the texture coordinate values where the texture coordinate values of The basic texture changes several times. The parts of the device for generating image 1 are constituted as described in previous paragraphs. And in the image memory 7 are stored, in addition to the basic textures, the modulation textures. Next, the processing procedure by which the modulation textures are used to perform the amplitude modulation in the patterns of images generated by basic textures will be described. Figure 14 shows the series of steps in the processing by which an image is generated using basic textures, and Figure 15 shows the series of processing steps by which amplitude modulation is applied to an image with modulation textures. As shown in Fig. 14, in step S1, the image generation device 1 reads a basic texture per pixel tool 6. In step S2, the pixel tool 6 confirms whether the drawing of all the objects has been completed. polygons to draw. If the pixel tool 6 confirms that the drawing of all the polygons to be drawn is completed, proceed with step S21, of figure 15, and if it confirms that the drawing of all the polygons to be drawn has not been completed, it continues to step S3. In step S3, the pixel tool 6 calculates the brightness of each vertex of the polygon to be drawn. Subsequently, in step S4 the device for generating image 1, by means of the DDA unit 5, determines the texture coordinates, the value of LOD, brightness and A (value a) of each image of the polygon to be drawn. In step S5, the pixel tool 6 selects the base direction of the base texture that will be used in accordance with the LOD value and reads the values in pixels of the basic texture corresponding to the texture coordinates. Also, here, a bilinear or trilinear filter is applied as necessary. If there is a need, a bilinear filter or a trilinear filter is developed. Here, the so-called trilinear processing, where a trilinear filter is applied, is a processing in which, if the LOD value has an intermediate value, the linear interpolation of the values in pixels of different levels of MIP is developed. Even if such trilinear processing is performed, the effectiveness of the technique is the same. Then, in step S6, the pixel tool 6 calculates the values in final pixels of the values in pixels of the basic texture, brightness of the polygons, the alpha value of the basic texture and the alpha value of the polygons. In step S7, the pixel tool 6 draws the values in pixels in the frame buffer 8. Here, the buffer processing Z is done as needed. Following the processing of this step S7, the pixel tool 6 again decides in step S2 whether the drawing of all the polygons to be drawn has been completed. In step S21, which is shown in figure 15 which is followed if in step 2 it is confirmed that the drawing of all the polygons to be drawn is completed, the device for image generation 1 by means of the pixel tool 6, reads in a modulation texture from the modulation texture buffer 10. If it is confirmed that the drawing of all the polygons to be drawn is completed, the pixel tool 6 finishes this processing and if it has been confirmed that the drawing of all the polygons to draw has not been completed, continue with step S23. In step S23, the texture coordinate values of the vertices of the polygon to be drawn are transformed to fit the modulation texture. Then, in step S24, the image generation device 1, by means of the DDA unit 5, determines the texture coordinates in the LOD value of each pixel of the polygon to be drawn. In step S25, the pixel tool 6 decides by the LOD value the base direction of the modulation texture that will be used and read in the pixel values of the texture corresponding to the texture coordinates. Also, a bilinear filter or a trilinear filter is applied as necessary. In step S26, the pixel tool 6 reads the corresponding pixel value from the frames buffer 8 and modulates this value in pixels (multiplies it) using the modulation texture. In step S27, the pixel tool 6 draws the resulting pixel values towards the frames buffer 8. Here, the buffer processing Z is done as needed. After the processing of step S27, once again in step S22, it is decided whether the drawing of all the polygons to be drawn has been completed. For the aforementioned series of processing steps, the device for generating image 1, using a modulation texture, applies the amplitude modulation to the patterns of the image generated using the basic textures. The image generation device 1 has the above composition, and when generating images, it can generate images that have the appropriate detail in a region near the point of view, where a decrease in the image resolution is observed in the processing operations of ordinary texture maps. Figures 16A and 16B and 17A and 17B show specific examples of images generated by making texture maps.

Here, the making of texture maps is done by making texture maps to a base consisting of a group of polygons, as shown in Figure 18B to that shown in Figure 18A. Specifically, in Figure 16A an image generated by applying the present invention is shown, while in Figure 16B there is an image generated by making ordinary texture maps; clearly, the image shown in figure 16A, which is an image generated by applying this invention, is drawn as an image having the appropriate detail. In this way, with respect to regions different from those near the point of view, what is obtained is an image that is almost as with the basic texture, making it easier to create an image as the designer imagined it. And by combining this with the alpha destination test, it is possible to create the resolution only for specified regions. Also, because the device for image generation 1 can use the existing MIPMAP processing mechanism, it is easy to accelerate image generation. By increasing the dynamic scale of modulation textures when the MIP value becomes smaller, the resolution of nearby parts can be emphasized effectively. In a region where the MIP value > 0, where the basic texture tends to be reduced, a natural decrease in resolution is observed by the processing that reduces the dynamic scale of the modulation, thus avoiding the presence of overlap. Furthermore, in a device for generating image 1, the modulation textures can be constituted in such a way that they are different from the basic textures. By making the textures by modulation in textures that are not related to the basic textures, for example when elaborating them in textures that fulfill the sensation of a material like the surface of a fabric, it becomes possible to naturally provide the details of an image. Specifically, in Figure 17A is an image that is generated by modulation textures made from a patch, and in Figure 17B is an image made with original ordinary textures; clearly, the image shown in figure 17A is drawn as an image in greater detail. Also, a device for image generation 1 can limit the region where the texture is modulated. By limiting the region that is modified using modulation textures it is possible to increase the resolution or apply a pattern to only a part of the generated image. To limit the region, it is enough, when drawing a basic texture, to draw with a specified alpha value only the part that will be modified, and to perform the destination alpha test when modulation textures are drawn. Also, the image generation device 1 can compress the amount of data by unifying basic textures and modulation textures. In some cases, basic textures and modulation textures are used in fixed combinations (pairs). For example, in the case of an image that has a self-similarity, as an image of nature, natural resolution can be created using as modulation texture an image where the color components have been removed from the basic texture, and in that case it can be handled as an ordinary RGBA texture if the basic texture is stored in the RGB region and the modulation texture is stored in the texture region A. If the RGBA texture is expressed using a color lookup table (CLUT), it is possible to express the basic textures and modulation textures with 8-bit textures by separating and using the CLUT as 8 bits, for example, ranging from 0 to 127 from it to the MIPO texture, 128 to 191 of this to the texture MIP1, 192 to 223 to the texture MIP2, and so on. In general, if the MIP value is greater, the values in pixels that have an average value of the values in pixels just produced, and the CLUT can not be shared between the textures for different MIPMAP, which makes the division of the CLUT as described in previous paragraphs adequate enough. Such data compression makes it possible, for example, to express a texture 256 x 256 x 4 = 262 KB with 128 x 128 x 1 = 16 KB. The method for generating the image of the present invention makes it possible to generate general patterns on polygons by making basic texture maps, and amplitude modulation processing of the modulation textures, on patterns generated by making basic texture maps. Therefore, the details of the patterns in the region near the point of view can be generated properly. The image generating device of the present invention has a memory medium that stores basic textures on maps to generate the general pattern on a polygon and modulation textures used to amplitude-scale the patterns generated by making basic texture maps, and an image processing means that, by performing modulation maps by amplitude of the modulation textures, performs the amplitude modulation processing on the patterns generated by making basic texture maps, and thus is capable, by performing of modulation modulation amplitude maps of modulation textures, to perform amplitude modulation processing in patterns generated by making maps of basic textures. In this way, the device for generating images can adequately generate the details of the patterns in the region near the point of view.

Claims (10)

NOVELTY OF THE INVENTION CLAIMS

1. - A me for generating an image to generate a two-dimensional image by making texture maps to three-dimensional polygons, comprising the steps of: generating a general pattern on said polygons by making basic texture maps and executing by making modulation maps of amplitude of modulation textures, the amplitude modulation processing on patterns generated based on the realization of maps of said basic textures.

2. The me for generating image according to claim 1, further characterized in that in said processing for amplitude modulation, an amplitude becomes smaller with a distance that increases from the vicinity of the point of view.

3. The me for generating image according to claim 1, further characterized in that a period of repetition of said basic textures and a repetition period of said modulation textures are misaligned with respect to each other.

4. The me for generating image according to claim 1, further characterized in that said modulation textures are set at spatial frequencies greater than e of said basic textures, with color information removed from said basic textures.

5. The me for generating images in accordance with that described in claim 1, further characterized in that said modulation textures consist of different patterns of said basic textures.

6. A device for generating an image to generate a two-dimensional image by making texture maps to three-dimensional polygons, comprising: a memory medium that stores basic textures that will be made on maps to generate the general pattern on a polygon, and modulation textures used to modulate by amplitude the patterns generated by making maps of the basic textures; and an image processing means which, by performing amplitude modulation maps of the modulation textures, executes amplitude modulation processing on the patterns generated based on the realization of maps of the basic textures.

7. The device for generating image according to claim 6, further characterized in that in said amplitude modulation processing, the amplitude becomes smaller with the distance increasing from the vicinity of the point of view.

8. The device for generating image according to claim 6, further characterized in that a period of repetition of said basic textures and a repetition period of said modulation textures are misaligned with respect to each other.

9. The device for generating image according to claim 6, further characterized in that said modulation textures are set in spatial frequency greater than e of said basic textures, with color information removed from said basic textures.

10. The device for generating image according to claim 6, further characterized in that said modulation textures consist of different patterns of said basic textures.