Method for drawing object having rough model and detailed model

0001The present disclosure relates to subject matter contained in <patcit id="pcit0001" dnum="JP2003418635A"><text>Japanese Patent Application No. 2003-418635</text></patcit>, filed on December 16, 2003.

0002The present invention relates to an image processing technique for drawing an object having rough model data of multiple polygons and detailed model data of more polygons than the rough model data.

0003Conventionally, in a video game, a virtual three-dimensional space is perspective-transformed onto a virtual screen from a predetermined viewpoint to draw an image of an object existing in the three-dimensional space. In order to display the progress of the game as a moving image, time in which one image must be drawn is limited to, for example, one frame period. For this reason, an amount of processing for drawing the image needs to be reduced. A Level Of Detail (LOD) technique is widely used to reduce the processing amount.

0004In the LOD technique, multiple model data, each having a different degree of detail for the same object (for example, a different number of polygons), is prepared. When a distance between a viewpoint and an object to be displayed is smaller than a predetermined threshold, an image of the object is drawn using detailed model data (for example, model data with a large number of polygons). When a distance between a viewpoint and an object to be displayed is larger than the predetermined threshold, an image of the object is drawn using rough model data (for example, model data with a small number of polygons).

0005According to the application of the LOD technique, an image of an object which is placed a short distance from the viewpoint can be displayed in detail using the detailed model data. An object which is placed a long distance from the viewpoint does not have to be displayed in detail, and an amount of processing for drawing the image can be limited by using the rough model data. For example, as shown in Unexamined <patcit id="pcit0002" dnum="JP2003115055A"><text>Japanese Patent Publication No. 2003-115055</text></patcit>, Unexamined <patcit id="pcit0003" dnum="JP2003228725A"><text>Japanese Patent Publication No. 2003-228725</text></patcit>, Unexamined <patcit id="pcit0004" dnum="JP2001250128A"><text>Japanese Patent Publication No. 2001-250128</text></patcit>, and Unexamined <patcit id="pcit0005" dnum="JP2001076180A"><text>Japanese Patent Publication No. 2001-76180</text></patcit>, a distance between a viewpoint and an object to be displayed is determined by a distance between a typical point, which is set at a central position of the object, and the viewpoint. ,

0006A general discussion of LOD is provided in the textbook "Level of Detail for 3D Graphics" Morgan Kaufman, 2002, by D. Luebke et al. A plurality LOD selection factors are discussed, including distance, perceptual factors, and others.

0007An image synthesizing system for game playing apparatus is described in <patcit id="pcit0006" dnum="US5577960A"><text>US patent 5,577,960</text></patcit>. Game objects such as a racing car are rendered employing different LODs. The LOD is determined based on the distance from the object from a virtual camera viewpoint.

0008However, in the conventional LOD technique, when the distance between the typical point of the object and the viewpoint is larger than the threshold, an image of a part placed at a position much closer to the viewpoint is drawn using rough model data. When the distance between the typical point of the object and the viewpoint is smaller than the threshold, an image of a part placed at a position much farther from the viewpoint is drawn using detailed model data.

0009The Article "<nplcit id="ncit0001" npl-type="s"><text>Multiresolution modelling of arbitrary polygonal surfaces: a characterization", Computers and Graphics vol. 26, no. 3, pages 449-462, by J. Ribelles</text></nplcit> et al. describes multiresolution modelling wherein an object is represented by means of a set of approximations, each with a different number of polygons. In particular, different approximations (i.e. different LODs) may be employed within a single object. Plural criteria for determining a particular one of several discrete LODs to be applied to a particular portion of an image, including distance, semantic importance and perceptual criteria are defined. Further details for displaying an object with variable resolution LODs in a single object are discussed in the Article "Variable Resolution Level-of Detail of Mulitresolution Ordered Mashes", 9^th International Conference in Central Europe on Computer Graphics, Visualization and Computer Vision, 2001 by J. Ribelles, et al.

0010Patent document <patcit id="pcit0007" dnum="WO03081567A1"><text>WO 03/081567 A1</text></patcit> describes a terrain rendering for a flight simulator employing 3D computer graphics for rendering geospecific terrains. The terrain is organized into a finite number of layers, which are centered about the viewer. The terrain is rendered with different levels of detail, wherein the greatest level of detail is provided in the central layer. The layers include overlapping regions. In the overlapping regions, Z-buffer technology is applied in order to employ the LOD of the smaller layer.

0011In the conventional LOD technique, when two or more such objects exist, there is a possibility that a part with a short distance from the viewpoint of one object will be drawn using rough model data, and that a part with a long distance from the viewpoint of the other object will be drawn using detailed model data. In this case, the position closer to the viewpoint is displayed roughly and the position farther from the viewpoint is displayed in detail. This gives an uncomfortable feeling to a player who watches the displayed image. Particularly, when multiple such objects exist close to each other, the player's discomfort increases.

0012An object of the present invention is to provide an image drawing apparatus that is capable of appropriately drawing an image of multiple objects to be displayed while restraining an amount of processing required for drawing the image.

0013This is achieved by the features of the independent claims. Further features and advantages of the present invention are the subject matter of dependent claims.

0014In the above image drawing apparatus, the image-drawing field, which is the image drawing range in the virtual three-dimensional space, is divided into the first field and the second field. When even a part of the object is included in the first field, the image obtained by perspective-transforming the polygons of the rough model data is drawn. When even a part of the object is included in the second field, the image obtained by perspective-transforming the polygons of the detailed model data included in the second field is drawn.

0015The image drawn using detailed model data, which includes more polygons than the rough model data, is drawn in more detail than the image drawn using rough model data. The image of the part included in the first field in the object is relatively roughly drawn using rough model data. The image of the part included in the second field in the object is drawn in detail using detailed model data. In this way, since the degree of detail of the image to be drawn differs for each part of the object depending on the field in which the corresponding part is included, an appropriate image can be displayed on the display screen.

0016The image of the object whose entirety is included in the first field is drawn using only rough model data. The image of the object whose entirety is included in the second field is drawn using only detailed model data. In such a case, there is no need to determine in which field each polygon is included, so that the image can be drawn with a small amount of processing. <ul id="ul0001" list-style="none" compact="compact"><li><figref idref="f0001">FIG 1</figref> is a block diagram illustrating a configuration of a video game apparatus according to an embodiment of the present invention;</li><li><figref idref="f0002">FIG. 2</figref> is a view illustrating an example of an object table prepared for each object to be displayed;</li><li><figref idref="f0003">FIG. 3A</figref> is a view illustrating a configuration of a detailed model data table (or rough model data table);</li><li><figref idref="f0003">FIG. 3B</figref> is a view illustrating an example of a detailed model;</li><li><figref idref="f0003">FIG 3C</figref> is a view illustrating an example of a rough model;</li><li><figref idref="f0004">FIG. 4A</figref> is a view illustrating a configuration of a detailed model bounding box table (or rough model bounding box table);</li><li><figref idref="f0004">FIG. 4B</figref> is a view illustrating an example of a detailed model bounding box;</li><li><figref idref="f0004">FIG. 4C</figref> is a view illustrating an example of a rough model bounding box;</li><li><figref idref="f0005">FIG. 5</figref> is a view illustrating an example of a vertex table in which the respective vertexes of the detailed model data table (or rough model data table) and the respective vertexes of the detailed model bounding box table (or rough model bounding box table) are registered;</li><li><figref idref="f0006">FIGS. 6A and 6B</figref> are views each explaining perspective-transformation in a video game according to an embodiment of the present invention;</li><li><figref idref="f0007">FIG. 7</figref> is a view illustrating an example of a perspective-transformation matrix;</li><li><figref idref="f0007">FIG. 8</figref> is an explanatory view of a far image-drawing field and a near image-drawing field;</li><li><figref idref="f0008">FIG. 9</figref> is a flowchart illustrating processing executed for each frame period in order to draw and display an image of an object in a video game according to an embodiment of the present invention;</li><li><figref idref="f0009">FIG. 10</figref> is a flowchart specifically illustrating image generation processing of <figref idref="f0008">FIG. 9</figref>;</li><li><figref idref="f0010">FIGS. 11A to 11D</figref> are views each illustrating an example of a display form of one object that moves in a virtual three-dimensional space; and</li><li><figref idref="f0010">FIG. 12</figref> is a view illustrating an example of a display form of two objects existing close to each other in a virtual three-dimensional space.</li></ul>

0017An embodiment of the present invention will be specifically described with reference to the drawings. The following explains an example in which the present invention is applied when an object having rough model data and detailed model data is drawn in a video game.

0018<figref idref="f0001">FIG. 1</figref> is a block diagram illustrating a configuration of a video game apparatus 100 for executing the video game according to an embodiment of the present invention. As illustrated in the figure, the video game apparatus 100 is mainly constructed to include a video game main body 101. The video game main body 101 includes a control section 103, a RAM (Random Access Memory) 105, a hard disk drive (HDD) 107, a sound processor 109, a graphics processor 111, a DVD/CD-ROM drive 113, a communications interface 115, and an interface section 117, each of which is connected to an internal bus 119.

0019The sound processor 109 is connected to a sound output device 125, which is a speaker. The graphics processor 111 is connected to a display device 121 having a display screen 122. A storage medium (DVD-ROM or CD-ROM in this embodiment) 131 can be attached to the DVD/CD-ROM drive 113. The communications interface 115 is connected to a network 151. An input section (controller) 161 and a memory card 162 are connected to the interface section 117.

0020The control section 103 includes a CPU (Central Processing Unit), a ROM (Read Only Memory), etc., and executes a program stored on the HDD 107 or the storage medium 131 to control the video game main body 101. The control section 103 has an internal timer. The RAM 105 is a work area for the control section 103. The HDD 107 is a storage area for storing a program and data. In the case where a program executed by the control section 103 instructs the sound processor 109 to output a sound, the sound processor 109 interprets the instruction and outputs a sound signal to the sound output device 125.

0021The graphics processor 111 develops an image onto the frame memory (frame buffer) 112 and outputs a video signal, which displays the image on the display screen 122 of the display device 121 according to a drawing command output from the control section 103. The frame memory 112 includes two units of frame memory. It is assumed that one frame period of the image included in outputting the video signal is, for example, 1/30 sec. The graphics processor 111 draws one image in one frame period (namely, 1/30 sec.). The DVD/CD-ROM drive 113 reads the program and data from the storage medium 131. The communications interface 115 is connected to the network (Internet 3) to perform communications with other computers.

0022The interface section 117 outputs input data sent from the input section 161 to the RAM 105. The control section 103 interprets the input data from the input section 161 to carry out arithmetic processing. The input section 161 includes a directional key and multiple operation buttons. The directional key is used to move a player character and a cursor in the game. Each operation button is used to instruct an operation of the player character and a decision of an item indicated by the cursor. The interface section 117 forwards data, indicative of the progress of the game stored in the RAM 105, to the memory card 162 based on the instruction from the control section 103. The interface section 117 reads data of the game at the time of interruption from the memory card 162 and transfers the read data to the RAM 105 based on the instruction from the control section 103.

0023The program and data for performing the game by the video game apparatus 100 are first stored on, for example, the storage medium 131. The program and data are read by the DVD/CD-ROM drive 113 and loaded onto the RAM 105 at the time of execution. The control section 103 processes the program and data loaded onto the RAM 105, outputs a drawing command to the graphics processor 111, and outputs an instruction of a sound output to the sound processor 109. Intermediate data is stored in the RAM 105 while the control section 103 performs processing.

0024In the video game according to this embodiment, an explanation is given of data prepared to draw and display an image of an object. Data prepared in advance includes an object table, a detailed model data table, a rough model data table, a detailed model bounding box table, a rough model bounding box table, and a vertex table. These tables are hereinafter referred to as a table group in some cases. Data of these tables are stored in the storage medium 131 or HDD 107, and retrieved to the RAM 105 as required.

0025In each object to be displayed in this video game, model data including multiple polygons is prepared. Each object has two kinds of model data, namely, rough model data and detailed model data. Though there is an object having only one kind of model data, such an object is not a target of the present invention and the explanation is omitted.

0026<figref idref="f0002">FIG. 2</figref> is a view illustrating an object table 200 prepared in connection with each object to be displayed. As illustrated in the figure, in the object table 200, an object ID 201, a detailed model data pointer 202, a detailed model bounding box pointer 203, a rough model data pointer 204, and a rough model bounding box pointer 205 are registered.

0027The object ID 201 is used to uniquely identify the object. The detailed model data pointer 202 is a pointer that refers to the detailed model data table in which detailed model data is registered. The detailed model bounding box pointer 203 is a pointer that refers to the detailed model bounding box table in which the detailed model bounding box is registered. The rough model data pointer 204 is a pointer that refers to the rough model table in which rough model data is registered. The rough model bounding box pointer 205 is a pointer that refers to the rough model bounding box table in which the rough model bounding box is registered.

0028<figref idref="f0003">FIG. 3A</figref> is a view illustrating the detailed model data table 300 shown by the detailed model data pointer 202. Regarding the rough model data table shown by the rough model data pointer 204, the basic configuration is the same as that of the detailed model data table 300 except that the number of registered polygons is smaller than that of the registered polygons registered in the detailed model data table 300. In the detailed model data table 300, polygon identification numbers 311, 321, ... of the respective polygons that form the detailed model of the object and vertex identification numbers 312, 322, ... of the respective polygons are registered.

0029The polygon identification numbers 311, 321, ...are numbers for uniquely identifying the respective polygons that form the object. The vertex identification numbers 312, 322, ... are numbers for uniquely identifying the vertexes of the respective polygons. The shapes of the polygons are triangles or squares. Regarding the triangle polygon, three vertex identification numbers are registered. Regarding the square polygon, four vertex identification numbers are registered. The coordinate positions of the vertexes of the respective polygons shown by the vertex identification numbers 312, 322 ... are registered in a vertex table to be described later.

0030<figref idref="f0003">FIG. 3B</figref> is a view illustrating a detailed model 301 of a spherical object. The detailed model 301 is one in which the polygons registered in the detailed model table 300 are schematically shown. <figref idref="f0003">FIG 3C</figref> is a view illustrating a rough model 302 of the same spherical object as that of <figref idref="f0003">FIG. 3B</figref>. The rough model 302 of <figref idref="f0003">FIG. 3C</figref> is one in which the polygons registered in the rough model data table are schematically shown. As is obvious from <figref idref="f0003">FIGS. 3B and 3C</figref>, the detailed model 301 includes the larger number of polygons than that of the rough model 302. Even if the detailed model 301 is largely displayed on a display screen 122, no unconformable feeling is given to the player.

0031<figref idref="f0004">FIG. 4A</figref> is a view illustrating a detailed model bounding box table 400 shown by the detailed model bounding box pointer 203. <figref idref="f0004">FIG. 4B</figref> is a detailed model bounding box 401 specified according to the detailed model box table 400. The detailed model bounding box 401 includes a rectangular parallelepiped that encloses the entirety of the detailed model 301 and has eight vertexes. In the detailed model bounding box table 400, vertex identification numbers 411 to 418 are registered to correspond to eight vertexes of the detailed model bounding box 401.

0032The configuration of the rough model bounding box table shown by the rough model bounding box pointer 205 is the same as that of the detailed model bounding box 400. <figref idref="f0004">FIG. 4C</figref> illustrates a rough model bounding box 402 shown by the rough model bounding box table. The rough model bounding box 402 includes a rectangular parallelepiped that encloses the entirety of the rough model 302 and has eight vertexes. In the rough model bounding box table, vertex identification numbers of eight vertexes of the rough model bounding box 402 are also registered.

0033<figref idref="f0005">FIG. 5</figref> is a view illustrating a vertex table in which coordinate positions of the respective vertexes having registered vertex identification numbers are registered in the detailed model data table 300, the detailed model bounding box table 400, the rough model data table, and the rough model bounding box table. In a vertex table 500, vertex identification numbers 511, 521, 531, ... and coordinate positions 512, 522, 532, ... are registered to be associated with one another. The vertex identification numbers 511, 521, 531, ... correspond to vertex identification numbers 312, 322, 411 to 418. The coordinate positions 512, 522, 532, ... are coordinates of a local coordinate system of the corresponding object. The coordinate positions 512, 522, 532, ... are transformed to coordinates of a world coordinate system, those of a viewpoint coordinate system, and further transformed to those of a perspective coordinate system (screen coordinate system) at the time of perspective-transformation.

0034An explanation is given of a method for displaying an object existing in the virtual three-dimensional space on the display screen 122 in the video game according to this embodiment. The virtual three-dimensional space including an object to be displayed is perspective-transformed onto a virtual screen that is set according to a direction of a visual axis from a position of a viewpoint of a virtual camera, and a perspective-transformed image is displayed on the display screen 122. <figref idref="f0006">FIGS. 6A and 6B</figref> are views each schematically illustrating a state in which the virtual three-dimensional space is perspective-transformed.

0035As illustrated in <figref idref="f0006">FIG. 6A</figref>, a viewpoint 601 is set in the virtual three-dimensional space and a visual axis is set from a position of the viewpoint 601 to a predetermined direction (-Z'-axial direction). A virtual screen 603 is set at a position with a fixed short distance from the viewpoint 601 to the direction of the visual axis. A range of a square pole, which is formed by connecting the viewpoint 601 to four comers of the virtual screen 603, becomes a field of view 602. A surface, which is placed at a position with a fixed distance from the viewpoint 601 to the direction of the visual axis and perpendicular to a Z'-axis (whose Z' coordinate is the same) in the range of the field of view 602, is set as a limit surface 604. In the range of the field of view 602, the range between the limit surface 604 and the virtual screen 603 is fixed as an image-drawing field 605, which is a range where an image is drawn by perspective-transformation.

0036A coordinate system for projecting the virtual three-dimensional space on the virtual screen 603 is a viewpoint coordinate system (X', Y', Z') as illustrated in <figref idref="f0006">FIG. 6A</figref>. Coordinates (X, Y, Z) of a world coordinate system, which is a coordinate system set in the virtual three-dimensional space, is transformed to the viewpoint coordinate system (X', Y', Z') by a vector operation using a predetermine matrix. Transformation from the coordinates of the world coordinate system to the coordinates of the viewpoint coordinate system can be executed using the method that is conventionally applied in the field of the three-dimensional image processing. This is not directly related to the present invention and the specific explanation is omitted.

0037The coordinates (X', Y', Z') of the viewpoint coordinate system included in the range of the image-drawing field 605 shown in <figref idref="f0006">FIG. 6A</figref> is transformed to coordinates (x', y', z') of a normalized perspective coordinate system as shown in <figref idref="f0006">FIG. 6B</figref>. In the coordinates of the normalized perspective coordinate system, each of x' coordinate, y' coordinate, and z' coordinate enters in the range of the image-drawing field 605 if any of these coordinates is included in the range from -1 to +1. Though transformation to the coordinates (x', y', z') of the normalized perspective coordinate system from the coordinates (X', Y', Z') of the viewpoint coordinate system is also executed using the vector operation, a perspective-transformation matrix (to be described later) for executing the vector operation is set.

0038Coordinates of the local coordinate system of the respective vertexes of the detailed model, the rough model, the detailed model bounding box and the rough model bounding box registered in the vertex table 500 are transformed to the coordinates of the world coordinate system. The coordinates of the world coordinate system of the virtual three-dimensional space, which include the coordinates of the world coordinate system of the respective vertexes transformed from the coordinates of the local coordinate system, are transformed to the coordinates of the viewpoint coordinate system. The coordinates of the viewpoint coordinate system are further transformed to the coordinates of the normalized perspective coordinate system by the perspective-transformation matrix. It is assumed that the perspective-transformation in this embodiment includes the transformation to the normalized coordinates of the perspective coordinate system.

0039<figref idref="f0007">FIG. 7</figref> is a view illustrating an example of the perspective-transformation matrix. The perspective-transformation matrix is a matrix of 4 rows and 4 columns. In the perspective-transformation matrix, 1 indicates an X' coordinate of a left end of a front clip surface (virtual screen), r indicates an X' coordinate of a right end of a front clip surface (virtual screen), t indicates a Y' coordinate of an upper end of a front clip surface (virtual screen), b indicates a Y' coordinate of a lower end of a front clip surface (virtual screen), n indicates a distance from the viewpoint 601 of a front clip surface (virtual screen), and f indicates a distance from the viewpoint 601 of an inner clip surface (limit surface). In the case where perspective-transformation is performed symmetrically with respect to a plane (Y'- Z' plane) where the X' coordinate through which the visual axis passes is 0, any of element values of the first row and third column, second row and third column, and third row and third column becomes 0.

0040In the video game according to this embodiment, the image-drawing field 605, which is the range where the image is drawn by perspective-transformation, is divided into a far image-drawing field and a near image-drawing field. In a part included in the far image-drawing field of the object to be displayed, an image is drawn using rough model data. In a part included in the near image-drawing field of the object to be displayed, an image is drawn using detailed model data.

0041<figref idref="f0007">FIG. 8</figref> is an explanatory view of a far image-drawing field and a near image-drawing field. In the case where an image is drawn by perspective-transformation, in order to decide a drawing range of the image, the virtual screen 603 is set at a position with a fixed short distance (distance is 1 in this case) from the viewpoint 601, the limit surface 604 is set at a position with a fixed long distance (distance is 100 in this case) from the viewpoint 601, and the image-drawing field 605 is formed therebetween. According to this embodiment, a position of the predetermined distance from the viewpoint 601 (distance is 50 in this case) is set as a boundary position, so that the image-drawing field 605 is divided into a far image-drawing field 605F and a near image-drawing field 605N.

0042In the case where an object existing in the far image-drawing field 605F is perspective-transformed to draw an image, the boundary position is set as a virtual screen 603F in the far image-drawing field 605F and the original limit surface 604 is set as a limit surface 604F in the far image-drawing field 605F. A perspective-transformation matrix for a far image-drawing field is set according to the coordinates of the vertexes of four corners of the virtual screen 603F, a distance from the viewpoint 601 to the virtual screen 603F, and a distance from the viewpoint 601 to the limit surface 604F.

0043In the case where an object existing in the near image-drawing field 605N is perspective-transformed to draw an image, the original virtual screen 603 is set as a virtual screen 603N in the near image-drawing field 605N and the boundary position is set as a limit surface 604N in the near image-drawing field 605N. A perspective-transformation matrix for a near image-drawing field is set according to the coordinates of the vertexes of four comers of the virtual screen 603N, a distance from the viewpoint 601 to the virtual screen 603N, and a distance from the viewpoint 601 to the limit surface 604N.

0044The following explains processing in the video game according to this embodiment. In order to simplify the explanation, explanation of processing except processing relating to the present invention is sometimes omitted. It is assumed that processing of the control section 103 explained below sometimes includes processing executed by the graphics processor 111 based on an instruction from the control section 103.

0045<figref idref="f0008">FIG. 9</figref> is a flowchart illustrating processing executed for each frame period in order to draw and display an image of an object in a video game according to an embodiment of the present invention.

0046When a new frame time is started, the control section 103 outputs a drawing command to the graphics processor 111 (step S101). In response to the drawing command, the graphics processor 111 generates a video signal based on image data developed on one unit of frame memory 112 and outputs it to the display device 121. Accordingly, the image of the object drawn in the previous frame is displayed on the display screen.

0047The control section 103 performs displaying object decision processing for deciding an object to be displayed according to the progress of the game at the present time after outputting the drawing command (step S102). In the displaying object decision processing, the control section 103 stores a new displaying object in this frame period and the corresponding table group in the RAM 105. When the object is not to be used as the displaying object in this frame period, the corresponding table group is erased from the RAM 105. When there is an object whose display position is changed, the control section 103 updates the stored contents of the object to the changed contents.

0048The control section 103 performs viewpoint and visual axis decision processing and decides a position of the viewpoint 601 and a direction of the visual axis, which are used at the time of perspective-transformation, based on the position of the object stored as a displaying object (step S103). When the position of the viewpoint 601 and the direction of the visual axis are decided, the control section 103 performs image generation processing for perspective-transforming the virtual three-dimensional space including the displaying object from the position of the viewpoint 601 to draw the perspective-transformed image on the other unit of frame memory 112 (step S104). The details of the image generation processing are described later. Then, processing for one frame period is ended and the same processing is started for a next frame period.

0049The image generation processing of the step S 104 will be specifically explained. <figref idref="f0009">FIG. 10</figref> is a flowchart specifically illustrating image generation processing of step S104. In the image generation processing, the objects as displaying objects stored in the RAM 105 are sequentially used as processing objects. However, the control section 103 determines whether there is an object that is not yet processed among the objects stored as the displaying objects (step S201).

0050When there is an object that is not yet processed, the control section 103 uses a new object as a processing object and sets a perspective-transformation matrix for a far image-drawing field 605 in a predetermined area of the RAM 105 according to the positions of the virtual screen 603F and the limit surface 604F in the far image-drawing field 605F (step S202). The control section 103 selects a rough model data table and a rough bounding box table and sets them in the RAM 105 in connection with the processing object (step S203).

0051The control section 103 perspective-transforms the rough model bounding box 402 registered in the rough bounding box table of the processing object and transforms coordinates of the respective vertexes to coordinates of the normalized perspective coordinate system (step S204). The control section 103 determines whether even a part of the rough model bounding box 402 is included in the far image-drawing field 605F, depending on whether there is at least one coordinate that is included in the far image-drawing field 605F among coordinates of the normalized perspective coordinate system of the vertexes of the rough model bounding box 402 (step S205). When there are vertexes where all of x'-coordinate, y'-coordinate and z'-coordinate of the perspective coordinate system are placed between -1 and +1, it is determined that the rough bounding box 402 is included in the far image-drawing field 605F. When the rough model bounding box 402 is not included in the far image-drawing field 605F, the processing directly proceeds to step S210.

0052When even a part of the rough model bounding box 402 is included in the far image-drawing field 605F, polygons registered in the rough model data table of the processing object are sequentially used as processing polygons. The control section 103 determines whether there is a polygon that is not yet processed among the polygons registered in the corresponding rough model data table (step S206). When there is no polygon that is not yet processed, the processing proceeds to step S210.

0053When there is a polygon that is not yet processed, the control section 103 perspective-transforms a new processing polygon and transforms coordinates of the respective vertexes to coordinates of the normalized perspective coordinate system (step S207). The control section 103 determines whether even a part of the processing polygon is included in the far image-drawing field 605F, depending on whether there is at least one coordinate that is included in the far image-drawing field 605F among coordinates of the normalized perspective coordinate system of the vertexes of the processing polygon (step S208). When there are vertexes where all of x'-coordinate, y'-coordinate and z'-coordinate of the perspective coordinate system are placed between -1 and +1, it is determined that the processing polygon is included in the far image-drawing field 605F. When the processing polygon is not included in the far image-drawing field 605F, the processing proceeds to step S206.

0054When even a part of the processing polygon is included in the far image-drawing field 605F, the control section 103 performs drawing processing for drawing an image, which is obtained by perspective-transforming the processing polygon, on the frame memory 112. In order to prevent the image of a polygon that is placed before the processing polygon from disappearing, the control section 103 performs hidden surface removal processing such as a Z-buffer method (step 209). Then, the processing returns to step S206.

0055In step S210, the control section 103 sets a perspective-transformation matrix for a near image-drawing field in a predetermined area of the RAM 105 according to the positions of the virtual screen 603N and the limit surface 604N in the near image-drawing field 605N. The control section 103 selects a detailed model data table 300 and a detailed model bounding box table 400 and sets them in the RAM 105 in connection with the processing object (step S211).

0056The control section 103 perspective-transforms the detailed model bounding box 401 registered in the detailed model bounding box table 400 of the processing object and transforms coordinates of the respective vertexes to coordinates of the normalized perspective coordinate system (step S212). The control section 103 determines whether even a part of the detailed model bounding box 401 is included in the near image-drawing field 605N, depending on whether there is at least one coordinate that is included in the far image-drawing field 605F among coordinates of the normalized perspective coordinate system of the vertexes of the detailed model bounding box 401 (step S213). When there are vertexes where all of x'-coordinate, y'-coordinate and z'-coordinate of the perspective coordinate system are placed between -1 and +1, it is determined that the detailed model bounding box 401 is included in the near image-drawing field 605N. When the detailed model bounding box 401 is not included in the near image-drawing field 605N, the processing directly returns to step S201.

0057When even a part of the detailed model bounding box 401 is included in the near image-drawing field 605N, polygons registered in the detailed model data table of the processing object are sequentially used as processing objects. The control section 103 determines whether there is a polygon that is not yet processed among the polygons registered in the corresponding detailed model data table (step S214). When there is no polygon that is not yet processed, the processing returns to step S201.

0058When there is a polygon that is not yet processed, the control section 103 perspective-transforms a new processing polygon and transforms coordinates of the respective vertexes to coordinates of the normalized perspective coordinate system (step S215). The control section 103 determines whether even a part of the processing polygon is included in the near image-drawing field 605N, depending on whether there is at least one coordinate that is included in the near image-drawing field 605N among coordinates of the normalized perspective coordinate system of the vertexes of the processing polygon (step S216). When there are vertexes where all of x'-coordinate, y'-coordinate and z'-coordinate of the perspective coordinate system are placed between -1 and +1, it is determined that the processing polygon is included in the near image-drawing field 605N. When the processing polygon is not included in the near image-drawing field 605N, the processing returns to step S214.

0059When even a part of the processing polygon is included in the near image-drawing field 605N, the control section 103 performs drawing processing for drawing an image, which is obtained by perspective-transforming the processing polygon, on the frame memory 112. In order to prevent the image of a polygon that is placed before the processing polygon from disappearing, the control section 103 performs hidden surface removal processing such as a Z-buffer method (step 217). Then, the processing returns to step S214.

0060When there is no object that is not yet processed among the objects registered as displaying objects in step S201, the control section 103 ends image generation processing and returns to the flowchart of <figref idref="f0008">FIG. 9</figref>.

0061The following explains specific display examples of the object in the video game according to this embodiment. An explanation is given of an example of change in an image that is displayed when one object moves in the virtual three-dimensional space and an example of an image displayed when two objects exist close to each other in the virtual three-dimensional space.

0062<figref idref="f0010">FIGS. 11A to 11D</figref> are views each illustrating an example of a display form of one object that moves in the virtual three-dimensional space. In this example, an object 210 moves in the virtual three-dimensional space to go away from the position of the viewpoint 601. It is assumed that about 10/12 of the diameter of the object 210 is included in the near image-drawing field 605N and the remaining part is included in the far image-drawing field 605F in a state before the object moves. In this state, as illustrated in <figref idref="f0010">FIG. 11A</figref>, the 10/12 part included in the near image-drawing field 605N is drawn using the detailed model data 301 and the remaining part included in the far image-drawing field 605F is drawn using rough model data 302.

0063When the object 210 moves away from the viewpoint 601 by only 1/10 of the diameter, about 9/12 of the diameter of the object 210 is included in the near image-drawing field 605N and the remaining part is included in the far image-drawing field 605F. In this state, as illustrated in <figref idref="f0010">FIG. 11B</figref>, the 9/12 part included in the near image-drawing field 605N is drawn using the detailed model data 301 and the remaining part included in the far image-drawing field 605F is drawn using rough model data 302. As a result, the amount of parts of the object 210 drawn using the detailed model data 301 is smaller than that of <figref idref="f0010">FIG. 11A</figref>.

0064When the object 210 further moves away from the viewpoint 601 by only 1/10 of the diameter, about 8/12 of the diameter of the object 210 is included in the near image-drawing field 605N and the remaining part is included in the far image-drawing field 605F. In this state, as illustrated in <figref idref="f0010">FIG. 11C</figref>, the 8/12 part included in the near image-drawing field 605N is drawn using the detailed model data 301 and the remaining part included in the far image-drawing field 605F is drawn using rough model data 302. As a result, the amount of parts of the object 210 drawn using the detailed model data 301 is smaller than that of <figref idref="f0010">FIG. 11B</figref>.

0065When the object 210 further moves away from the viewpoint 601 by only 1/10 of the diameter, about 7/12 of the diameter of the object 210 is included in the near image-drawing field 605N and the remaining part is included in the far image-drawing field 605F. In this state, as illustrated in <figref idref="f0010">FIG. 11D</figref>, the 7/12 part included in the near image-drawing field 605N is drawn using the detailed model data 301 and the remaining part included in the far image-drawing field 605F is drawn using rough model data 302. As a result, the amount of parts of the object 210 drawn using the detailed model data 301 is smaller than that of <figref idref="f0010">FIG. 11C</figref>.

0066<figref idref="f0010">FIG. 12</figref> is a view illustrating an example of a display form of two objects existing close to each other in a virtual three-dimensional space. Two objects 220 and 230 in this figure have the same shape and size, and are drawn using the same rough model data and detailed model data. The object 220 is placed at a farther position from the viewpoint 601 than the object 230, and the amount of parts of the object 220 included in the near image-drawing field 605N is smaller than that of the object 230 included in the near image-drawing field 605N.

0067As illustrated in <figref idref="f0010">FIG. 12</figref>, since the amount of parts of the object 220 included in the near image-drawing field 605N is relatively small, the amount of parts to be drawn using the detailed model data 301 is small and almost all parts are drawn using the rough model data 302. Since the amount of parts of the object 230 included in the near image-drawing field 605N is larger than that of the object 220, the amount of parts to be drawn using the detailed model data 301 is larger than that of the object 220 and the amount of parts to be drawn using the rough model data 302 is smaller than that of the object 220.

0068As explained above, in the video game according to this embodiment, two kinds of model data, namely, rough model data and detailed model data, each having a different degree of detail, are prepared to the respective objects existing in the virtual three-dimensional space. Detailed model data includes more polygons than rough model data and an image to be drawn using detailed model data is displayed on the display screen 122 in more detail than an image to be drawn using rough model data.

0069The image-drawing field 605, which is the range where the image is drawn by perspective-transformation, is divided into the far image-drawing field 605F and the near image-drawing field 605N according to the distance from the viewpoint 601. When even a part of the displaying object is included in the far image-drawing field 605F, the image of the part included in the far image-drawing field 605F of the object is drawn using the rough model data. When even a part of the displaying object is included in the near image-drawing field 605N, the image of the part included in the near image-drawing field 605N of the object is drawn using the detailed model data.

0070Accordingly, regardless of whether the central position of the object is included in the far image-drawing field 605F or the near image-drawing field 605N, the part included in the far image-drawing field 605F is drawn roughly and the part included in the near image-drawing field 605N is drawn in detail. An appropriate image is drawn for each part of the object. Since the degree of detail of the image to be drawn is different for each part of the object, the player, who sees two objects displayed on the display screen 122, does not feel uncomfortable even when the different two objects exist close to each other in the virtual three-dimensional space.

0071Since it is determined whether even a part of the object is included in the far image-drawing field 605F using the rough model bounding box 402 of the rectangular parallelepiped, the number of vertexes used in the determination is only 8. When 8 vertexes of the rough model bonding box 402 are not included in the far image-drawing field 605F, there is no need to determine whether the respective polygons that form rough model data 302 of the object are included in the far image-drawing field 605F. Also, since it is determined whether even a part of the object is included in the near image-drawing field 605N using the detailed model bounding box 401 of the rectangular parallelepiped, the number of vertexes used in the determination is only 8. When 8 vertexes of the detailed model bounding box 401 are not included in the near image-drawing field 605N, there is no need to determine whether the respective polygons that form detailed model data 301 of the object are included in the near image-drawing field 605N.

0072For this reason, even when the image of the object is drawn for each part according to the distance from the viewpoint 601, there is no case in which the amount of processing for drawing the image becomes excessive. Also, regarding the amount of processing for determining whether even a part of the object is included in each of the far image-drawing field 605F and the near image-drawing field 605N, the use of the bounding boxes 401 and 402 makes the amount of processing small..

0073Regarding even an object determined as being included in the far image-drawing field 605F by the rough model bounding box 402, only polygons of rough model data included in the far image-drawing field 605F are drawn. Also, regarding even an object determined as being included in the near image-drawing field 605N by the detailed model bounding box 401, only polygons of detailed model data included in the near image-drawing field 605N are drawn. Accordingly, there is no need to perform processing for drawing an image of a polygon that it is not necessary to display on the display screen 122, and there is no case in which the amount of processing for drawing the image is too large.

0074Whether each polygon that forms rough model data 302 is included in the far image-drawing field 605F is determined using all vertexes of the polygon. Among the polygons that form rough model data 302, such polygons where at least one of three or four vertexes is included in the far image-drawing field 605F are drawn. Also, whether each polygon that forms detailed model data 301 is included in the near image-drawing field 605N is determined using all vertexes of the polygon. Among the polygons that form detailed model data 301, such polygons where at least one of three or four vertexes is included in the near image-drawing field 605N are drawn. This makes it possible to avoid a disadvantage that the image of the object cannot be drawn at a portion in the vicinity of a boundary position between the far image-drawing field 605F and the near image-drawing field 605N.

0075Regarding the image of the object, after the part included in the far image-drawing field 605F is drawn using rough model data, the part included in the near image-drawing field 605N is drawn using detailed model data. This prevents the image drawn in detail, using detailed model data, from disappearing due to the rough image of rough model data. Accordingly, even if a polygon included in the far image-drawing field 605F and a polygon included in the near image-drawing field 605N overlap each other, a detailed image of the part included in the near image-drawing field 605N of the object can be drawn.

0076The bounding boxes 401 and 402 are prepared to detailed model data 301 and rough model data 302, respectively. The bounding boxes 401 and 402 can be set according to the shape and size in the detailed model data 301 and rough model data 302, respectively. For this reason, it is possible to reduce the number of cases in which the object is determined as being included in the far image-drawing field 605F though rough model data 302 is not included for a polygon included in the far image-drawing field 605F. It is also possible to reduce the number of cases in which the object is determined as being included in the near image-drawing field 605N though detailed model data 301 is not included for a polygon included in the near image-drawing field 605N. This increases the number of cases in which no determination is performed on whether the respective polygons (that form detailed model data 301 and rough model data 302) are included in the far image-drawing field 605F or the near image-drawing field 605N.

0077The present invention is not limited to the aforementioned embodiment, and various modifications and applications may be possible. The following explains modifications of the aforementioned embodiment to which the present invention can be applied.

0078In the aforementioned embodiment, different bounding boxes are prepared to detailed model data 301 and rough model data 302, respectively. In contrast to this, a bounding box common to detailed model data and rough model data may be prepared. In this case, it is possible to reduce a storage capacity necessary for a bounding box table. Regarding the detailed model data and rough model data of the same object, though there is a large difference in a degree of detail therebetween, no difference occurs in an outer side therebetween. Accordingly, even when the bounding box common to detailed model data and rough model data is used, there is little possibility that a disadvantage will occur in terms of the amount of processing.

0079The shape of the bounding box is not limited to the rectangular parallelepiped and any polyhedron can be used as long as the entirety of the corresponding object is included therein. For example, a bounding box applied to a spherical object may be a regular dodecahedron or a regular icosahedron. In accordance with an increase in the number of faces of the bounding box and the number of vertexes, the amount of processing for determining whether the object is included in the far image-drawing field 605F or the near image-drawing field 605N is increased. However, it is possible to reduce the cases in which the object is determined as being included in the far image-drawing field 605F or the near image-drawing field 605N though there are no polygons included in the far image-drawing field 605F or the near image-drawing field 605N. This makes it possible to reduce the amount of processing as a whole.

0080The shape of the bounding box may not be a polyhedron that is formed of planes, and the part or entirety may be formed of curved surfaces. For example, a sphere including the entirety of the corresponding object may be applied for a bounding box. In this case, the shape of the bounding box can be defined by a mathematical expression of a sphere having a predetermined radius where the entire object can be included about a typical point set at a substantially central position of the object. Whether even a part of the object is included in each of the far image-drawing field and the near image-drawing field can be determined by a calculation result using the mathematical, expression of the bounding box and a mathematical expression that defines a boundary surface by which the image-drawing field is divided into the far image-drawing field and the near image-drawing field.

0081In the case where the number of polygons that form rough model data 402 is relatively small, it is possible to determine whether the object is included in the far image-drawing field 605F or the near image-drawing field 605N using rough model data 402. The rough model data 402 can serve as both the rough model bounding box and the detailed model bounding box. In this case, since there is no need to prepare the bounding box table in addition to the detailed model data table and the rough model data table, an amount of storage capacity necessary for storing data may be small.

0082In the aforementioned embodiment, it is determined whether the respective polygons that form detailed model data and rough model data are included in the far image-drawing field 605F or the near image-drawing field 605N by using the coordinate positions of all vertexes. In contrast to this, if the typical point of each polygon is set, it may be determined whether the respective polygons that form detailed model data and rough model data are included in the far image-drawing field 605F or the near image-drawing field 605N by using the coordinate position of the typical point. In this case, it is possible to reduce the amount of processing for determining whether each polygon is included in the far image-drawing field 605F or the near image-drawing field 605N. The coordinate position of the typical point of each polygon can be set to, for example, the position of its center of gravity.

0083In the aforementioned embodiment, the image-drawing field 605 that indicates the range where the perspective-transformation is performed in the virtual three-dimensional space is set symmetrically with respect to the visual axis. The image-drawing field 605 is divided into the far image-drawing field 605F and the near image-drawing field 605N according to the distance (z' coordinate of the perspective coordinate system) from the viewpoint 601. However, the method for dividing the image-drawing field 605 is not limited to this. The image-drawing field 605 may not be set symmetrically with respect to the Y'-Z' plane where the visual axis passes.

0084The boundary position at which the image-drawing field 605 is divided into the far image-drawing field 605F and the near image-drawing field 605N may be changed according to the kind of object. For example, in the case where an image of an object (for example, a player character), which has a high importance in the progress of the game, is drawn, the boundary position at which the image-drawing field 605 is divided into the far image-drawing field 605F and the near Image-drawing field 605N may be set at a farther position from the viewpoint 601 than that set in the case where an image of the other object is drawn. This makes it possible to display the object with a high importance on the display screen 122 in more detail than the other object regardless of the distance from the viewpoint 601.

0085The boundary position at which the image-drawing field 605 is divided into the far image-drawing field 605F and the near image-drawing field 605N may be changed according to the progress of the game. In this case, since the degree of detail of the object on the display changes even if the position of the object does not change, a visual effect is caused on the image displayed on the display screen 122 according to the progress of the game. When the image-drawing field 605 is not set symmetrically with respect to the Y'-Z' plane, the boundary surface between the far image-drawing field 605F and the near image-drawing field 605N may not be perpendicular to the visual axis.

0086When the player character is included as a displaying object, the progress of the game can be determined by a change in a parameter that the player character has. For example, when a value of HP (Hit Point: a remaining value of damage that the player character can receive) of the player character is high at a battle with an enemy character, the boundary surface, which divides the image dividing field into the far image-drawing field 605F and the near image-drawing field 605N, can be set to a position relatively close to the viewpoint 601 and when the value of HP is low, the boundary position can be set to a position relative far from the viewpoint 601.

0087In the aforementioned embodiment, two kinds of model data, namely, detailed model data and rough model data are prepared to one object. However, three or more kinds of model data each having a different degree of detail may be prepared to one object. In this case, the image-drawing field 605 can be divided according to the prepared number of kinds of model data. For example, the following considers a case in which three kinds of model data, which include rough model data formed of a predetermined number of polygons, normal model data whose number of polygons is larger than that of rough model data, and detailed model data whose number of polygons is larger than that of normal model data, are prepared to one object.

0088In this case, the image-drawing field 605 can be divided into three image-drawing fields, namely, a far image-drawing field, an intermediate image-drawing field, and a near image-drawing field in descending order of distance from the viewpoint 601. In the case where even a part of the object is included in the far image-drawing field, polygons of rough model data included in the far image-drawing field are drawn. In the case where even a part of the object is included in the intermediate image-drawing field, polygons of normal model data included in the intermediate image-drawing field are drawn. In the case where even a part of the object is included in the near image-drawing field, polygons of detailed model data included in the near image-drawing field are drawn.

0089In the aforementioned embodiment, in the case where the displaying object is included in both the far image-drawing field 605F and the near image-drawing field 605N, after polygons of rough model data included in the far image-drawing field 605F are drawn, polygons of detailed model data included in the near image-drawing field 605N are drawn.

0090When all polygons of rough model data are drawn, the images developed on the frame memory 112 are rough. However, if there is at least one polygon of detailed model data at the same position, the frame memory 112 is rewritten by the polygon of detailed model data, so that the part included in the near image-drawing field 605N of the object is displayed in detail on the display screen 122. In the case where all polygons of rough model data are drawn, there is no need to determine whether the object is included in the far image-drawing field 605F and the rough model bounding box is not needed.

0091In the aforementioned embodiment, when the number of displaying objects is multiple, image drawing using rough model data and image drawing using detailed model data are sequentially performed to one object, and thereafter image drawing using rough model data and image drawing using detailed model data are sequentially performed to a next object. In contrast to this, if the boundary positions between the far image-drawing field 605F and the near image-drawing field 605N, which are applied to the multiple objects, are the same, image drawing using rough model data is sequentially performed to the multiple displaying objects, and thereafter image drawing using detailed model data may be sequentially performed to the multiple displaying objects.

0092In this case, even when different objects overlap each other complicatedly in connection with the image projected on the virtual screen, the following disadvantage is not caused. Namely, a problem has been known when the image of a polygon of detailed model data existing in the vicinity of the boundary position in connection with one object disappears due to the image of a polygon of rough model data existing in the vicinity of the boundary position in connection with the other object.

0093In the aforementioned embodiment, the video game apparatus 100, which is a special-purpose machine, is used as a platform. In contrast to this, any apparatus such as a general-purpose computer may be used if the apparatus includes the same structural components as those of the video game main body 101 and a function of drawing an image. Moreover, a portable video game apparatus, which contains the display device 121 and the sound output device 125 in the same cabinet as that of the video game main body 101, may also be used.

0094In the aforementioned embodiment, the program and data of the video game apparatus 100 are stored in the storage medium 131 and distributed. A semiconductor memory card can be used as the storage medium 131 in place of a DVD-ROM or CD-ROM. In this case, the computer apparatus applied as a platform must include a card slot for inserting the memory card may be formed in place of the DVD/CD-ROM drive 113. The program and data relating to the present invention may be prestored in the HDD 107. Regarding the storage medium for storing the program and data relating to the present invention, any storage medium may be used according to the physical form of hardware and the distribution thereof.

0095The programs and data relating to the present invention may be stored on a fixed disk apparatus provided in a Web server. The Web server may convert the program and data stored in the fixed disk apparatus to a signal and superimpose the signal on a carrier wave, and distribute it to the video game main body 101 via the network 151. For example, the program and data, which the communications interface 115 receives from the Web server, can be stored in the HDD 107 and loaded to the RAM 105 at an executing time.

0096In the aforementioned embodiment, the explanation is given of the case in which the present invention is applied to the video game in which the object having rough model data and detailed model data is moved in the three-dimensional space to progress the game. The applicable range of the present invention is not limited to the video game. The present invention can be used in the general field of the computer graphics processing for drawing an image of the object having rough model data and detailed model data. A program including a function of drawing an image of the object having rough model data and detailed model data can be distributed by the aforementioned various kinds of methods as mentioned above regardless of the fact that the program is applied as a part of the video game.

0097Although the invention has been described with reference to several exemplary embodiments, it is understood that the words that have been used are words of description and illustration, rather than words of limitation. Changes may be made within the purview of the appended claims. Although the invention has been described with reference to particular means, materials and embodiments, the invention is not intended to be limited to the particulars disclosed; rather, the invention extends to all functionally equivalent structures, methods, and uses such as are within the scope of the appended claims.