JP2001109911A - Method and device for displaying three-dimensional graphic - Google Patents

Abstract

PROBLEM TO BE SOLVED: To prevent three-dimensional plotting from taking much time and to smoothly execute display even if few moving objects exist in the three- dimensional display of a moving picture. SOLUTION: A central processor 101 divides a three-dimensional space into front and back parts. It is judged whether the object of three-dimensional graphic data in a background area moves or stops. When it is stopped, background picture data 103 which is temporarily stored as a two-dimensional picture is read and it is transferred to a graphic processor 110. When the moving object exists in the background area or originally, a foreground area where the appearance frequency of the moving object is high is converted into two-dimensional graphic data through a three-dimensional picture processing. The graphic processor 110 synthesizes and displays two-dimensional data. Since the background area has high probability that a stop object such as a background exists, a three-dimensional plotting processing taking time such as a geometry processing can remarkably be saved and the real time display of a three-dimensional moving picture can smoothly be executed when a partition plane is set to be appropriate.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a three-dimensional graphic display method, and more particularly, to a three-dimensional graphic real-time processing method.
It relates to a three-dimensional virtual space display.

[0002]

2. Description of the Related Art In three-dimensional graphics by a computer, objects 301 to 303 as shown in FIG.
In the case of displaying a three-dimensional figure, after setting display scenes, geometry processing and rasterizing processing are performed to convert the three-dimensional structure data into two-dimensional image data. At this time, the object 303 is removed by the clipping process.
As long as 2 is within the screen range, a series of three-dimensional calculation processing is repeated for each frame.

For this reason, even when the object 302 is fixed and only the object 301 moves, exactly the same calculation processing is performed on the object 302 for each frame. As described above, since all three-dimensional calculation processing is performed, including objects that do not move, in a process having many objects such as a background composed of complex polygons, an object that is actually moving is Even if the number is small, it is difficult to perform smooth real-time display.

In recent years, although the processing speed of a computer has been improved, the processing capability of low-cost general-purpose products is limited. For this reason, there is an attempt to reduce or simplify necessary calculations as much as possible, thereby realizing high-speed three-dimensional graphics without using expensive devices. For example, Japanese Unexamined Patent Publication No. Hei 8-263690 discloses that when there is the viewpoint position information generated from the three-dimensional structure data and the already stored two-dimensional images having the same viewpoint position information or within the allowable range, the second part is used. Image display is performed using the three-dimensional image data, and the three-dimensional calculation processing is omitted.

[0005]

In the method of the above cited example, when the viewpoint is equal to the processed two-dimensional image data and the two-dimensional data can be reused, the processing can be sped up. However, in a moving image of three-dimensional graphics, both the viewpoint position and the object position are always moving, and the object position is often changed with respect to the same viewpoint position. For this reason, the frequency with which the stored two-dimensional image data can be reused is low, and it is difficult to perform high-speed image display of three-dimensional data by this method. Further, when the position of the two-dimensional image is relatively moved with respect to the viewpoint, there is a proposal to use the corrected two-dimensional image data. However, regarding the change of the viewpoint angle and the movement of the viewpoint position (other than the depth direction), However, the correction process takes time and the display error increases.

SUMMARY OF THE INVENTION An object of the present invention is to overcome the above-mentioned problems of the prior art, and to provide a three-dimensional graphic display method and apparatus capable of real-time and smooth display when a moving object and a non-moving object are mixed. Is to provide.

[0007]

In order to achieve the above object, the present invention divides a three-dimensional space (view volume) into a foreground and a foreground in the depth direction (Z value), and stops the movement of the background. It is determined whether or not the background image is stored in the memory as a two-dimensional image when the image is stopped, and the rendering of the background image stopped in the subsequent frames is performed by transferring the two-dimensional image on the memory. The three-dimensional calculation process is performed on the foreground only until the movement of the background is detected next.

[0008] Further, according to the present invention, a plurality of objects (objects) defined in a three-dimensional virtual space are subjected to three-dimensional drawing processing such as geometry processing for each frame, converted into two-dimensional graphic data and displayed. In the three-dimensional graphic display method, a virtual partition plane for dividing a three-dimensional virtual space (view volume) into a foreground (front region) and a background (back region) is set at a predetermined position in a depth direction (Z direction) from the viewpoint. When the viewpoint is not moving, the two-dimensional foreground data of the previous frame is reused for the foreground, the two-dimensional foreground data is generated for the foreground by a three-dimensional rendering process, and both data are combined and displayed. Features.

The virtual partition surface is set by a fixed value and / or a variable value, and the variable value is variable according to the appearance of a moving object.

[0010] The reuse of the two-dimensional graphic data is as follows.
Only when the object in the background is not moving, if there is a moving object, a three-dimensional rendering process is also performed on the background. At this time, when the object in the background does not move continuously for a predetermined number of frames, it is determined that there is no movement.

Further, the object is defined by a plurality of polygons, and the change is examined by using an arbitrary coordinate value of the polygon as a representative point to determine whether the object has moved.

[0012] The two-dimensional graphic data generated by the three-dimensional rendering process for the plurality of objects in the three-dimensional virtual space may temporarily store the two-dimensional data of the background in a reusable manner. .

A three-dimensional graphic display device according to the present invention, which enables the application of the above method, stores three-dimensional graphic data of a plurality of objects (objects) defined in a three-dimensional virtual space; A graphic processor for converting the three-dimensional data into two-dimensional graphic data and developing the pixels in a frame buffer, a display means for reading the frame buffer and displaying the two-dimensional graphic data on a screen;
In a three-dimensional graphic display device provided with a central processing unit for controlling the reading of dimensional data and the graphic processor, etc.
Partition plane data storage means for setting a virtual partition plane position (Z value) for dividing a three-dimensional virtual space (view volume) into a front area and a rear area; viewpoint information storage means for storing viewpoint coordinates for each frame; A background image data storage unit is provided for storing two-dimensional graphic data that has been subjected to three-dimensional rendering processing for objects in the area, and the central processing unit does not move the viewpoint or the viewpoint and the object in the rear area for each frame. The two-dimensional graphic data of the rear area in the previous frame is transferred from the background image data storage means to the graphic processor when the moving object is not moved.

[0014]

Embodiments of the present invention will be described below in detail with reference to the drawings. FIG.
1 shows a system configuration of a three-dimensional graphic display device. This device is a central processing unit 1 that controls the entire system via a system bus 118.
01, the main memory 102 for storing graphic information, and the frame buffer 1
12 and a graphic processor 110 for writing to the Z buffer 113, and an input / output controller 115 for controlling input / output devices such as a keyboard 116 and a mouse 117. The contents of the data written in the frame buffer 112 are displayed on the display 114 via the bus 111.

The main memory 102 stores the background image data 103 and the background image information data 104 according to the present embodiment together with the three-dimensional graphic data 105 to be displayed. The background image data 103 includes at least the frame buffer 1
It has two-dimensional image data that can store the resolution and the number of colors according to the size of Twelve. Background image information data 1
Reference numeral 04 includes information on viewpoint coordinates, viewpoint directions, image storage addresses, and image sizes related to the background image data 103.

In this embodiment, as a display processing control register, a back-ground partition reference plane data register 106 for storing a Z-coordinate value for setting a back-ground partition reference plane, and object information storage for object movement determination. A register 107 has a viewpoint information storage register 108 for judging the movement of the viewpoint, and further has a frame counter 109 for incrementing and counting for each frame.
The object information storage register 107 stores object ID information, representative point information for each object ID, and object position (rear / front) information, and the viewpoint information storage register 108 stores viewpoint coordinate data and viewpoint direction data. are doing.

A three-dimensional image is composed of a plurality of objects 301 to 303 as shown in FIG. 2, and each object is defined by a polygon. FIG. 3 shows an example of definition using polygons. For example, the object 300 is defined as a group of polygons 301 to 308, and when drawing, polygon data is collectively defined for each object.
At this time, the first vertex of the polygon data defined first is set as the representative point 310 of the object 300. An object ID number is assigned to the defined object 300 and recorded in the object information storage register 107 together with the representative point.

FIG. 4 is a flowchart of a process of a three-dimensional image display method according to one embodiment of the present invention. This processing is executed by the central processing unit 101. However, steps S50 and S6
02 and S703 are performed by the central processing unit 1
01 is executed by the graphic processor.

In step S10, the same initial processing as that of the conventional drawing and registration of the defined object are performed, and the initial position of the background reference plane is set. The coordinates of the foreground partition reference plane are stored in the foreground partition reference plane data register 106. In step S20, a setting process of a scene to be three-dimensionally displayed is performed for each frame.

In step S30, as shown in FIG.
Before rendering one frame, the foreground area 501 and the foreground area 5
02, and the front clipping surface 30
A background partitioning reference plane 500 having an arbitrary Z value is set between No. 4 and the rear clipping plane 305. In the present embodiment, the background partitioning reference plane 500 has a fixed value by the initial setting, and may be set by the initial processing (S10). In the condition determination step S40, a condition determination for branching to each of the normal drawing process S50, the captured background drawing process S60, and the background capture process S70 is performed.

FIG. 6, FIG. 7, and FIG.
40 shows a detailed flow. In step S401 in FIG. 6, the value of the frame counter 109 is incremented. In step S402, as shown in FIG. 7, details of the previous viewpoint coordinates are stored in the viewpoint information storage register 108.
The direction is extracted (S4021), and the current viewpoint coordinates and the direction are compared (S4022). If the coordinate values and directions do not match, the current viewpoint coordinate values and directions are recorded in the viewpoint information storage register 108 as viewpoint coordinate data and viewpoint direction data (S
4023), the value of the frame counter 109 is cleared to 0 (S4
024). Next, in the case of changing the viewpoint, the captured background image is unnecessary, so the captured background image data is cleared (S4).
025).

In step S403, as shown in detail in FIG. 8, the object information is extracted from the object information storage register 107 in the order of ID numbers (S4031). If the representative point of the extracted object is ahead of the background partition plane 500, The current object representative point coordinate value is recorded as representative point coordinate data (S4033). If the representative point of the extracted object is behind the background partition plane, the same ID
The current representative point coordinates of the object having the number are compared with the previous representative point coordinates (S4034). If the coordinate values do not match, the current object representative point coordinate value is recorded as representative point coordinate data (S4035), and the frame counter 1
The value of 09 is cleared to 0 (S4036), and the captured background image information data is cleared (S4037). These processes are performed for all objects.

If the value of the frame counter 109 exceeds the value of MAXFRAME (arbitrarily defined value) at step S404 in FIG. 6, the frame is located behind the viewpoint and foreground partitioning reference plane 500 during this MAXFRAME number of frames. Since the object has not moved, it can be determined that the viewpoint position and the object position have been fixed, and the process proceeds to step S405. When the value of the frame counter 109 is less than the MAXFRAME number, the process proceeds to step S50.

In the condition determination step S405, an image matching the current viewpoint coordinates and viewpoint direction is searched from the background image information data 104 recorded at the time of capturing the background, and if a matching image exists, the process proceeds to step S60. The process proceeds to the process, and if not, the process proceeds to step S70.

In step S50, a normal three-dimensional drawing process is performed because the viewpoint position and the object position have changed. This is equivalent to the conventional method. In step S501, a coordinate conversion process from three-dimensional coordinates to two-dimensional coordinates is performed.
In 502, a clipping process for cutting off data outside the display range, in S503, a light source calculation process for calculating light hits, in S504, a rasterizing process for dropping an image in pixel units, and in S505, a screen display for actual screen display Perform processing.

FIG. 9 shows details of step S60. In step S60, since the viewpoint position and the object position do not fluctuate over a fixed number of frames and a background exists for the current viewpoint position, first, a background display process is performed (S601). That is, the background image data 103 to be currently displayed is searched from the background image information data 104 of the main memory 102 (S6011), and the image of the background image data 103 is stored in the frame buffer 112 according to an instruction from the graphic processor 110. By transferring, the background is displayed (S6012).

Next, foreground display processing is performed (S602).
That is, the representative point 310 of each object in the frame
Are sorted from the rear side in the Z direction from the background reference plane 500 (S6021), and it is determined from the coordinates of the representative points whether the object to be drawn is before or after the background partition reference plane 500.
(S6022). For an object determined to be in front of the reference plane 500, the object information storage register 10
7, the front is set as the object position (S6023), the object on the foreground screen is set as a display target, and the same three-dimensional drawing process as in step S50 is executed to display the foreground (S6024). .

FIG. 10 shows the details of step S70. In step S70, if the viewpoint position and the object position do not fluctuate over a fixed number of frames and there is no background for the current viewpoint position, first, a capturing background image rendering process is performed (S701). That is, the representative points 310 of the objects in the frame are sorted from the rear side in the Z direction (S7011), and it is determined whether the coordinates of the representative points of the object to be drawn are behind or before the background partitioning reference plane 500 (S7012). For an object determined to be behind the reference plane 500, the rear is set as the object position in the object information storage register 107 (S701).
3) After the object on the foreground screen is to be displayed, the background is displayed by executing the above-described three-dimensional drawing process (S50) (S7014).

Next, a background fetch process is performed (S70).
2). When the representative point sorted from the rear side in the Z direction in S7012 is determined to be before the background partitioning reference plane 500, the rendering process has already been completed for the object behind the reference plane 500. The process proceeds to step S7021. In step S7021, the image drawn as the background is raster-scanned and transferred to the background image data 103 of the main memory 102 in accordance with an instruction from the graphic processor 110 (S7022) to capture the image as a two-dimensional still image. Do. At the same time, the viewpoint coordinates, viewpoint direction, image storage address, and image size are recorded in the background image information data 104 (S7023). Finally, step S7031
Then, the same foreground display processing as in step S602 is performed.

As described above, the object in the frame is located behind the background partition plane set in the Z direction, and when there is no change in the viewpoint position and the object position, the background of the two-dimensional data that has already been subjected to the drawing processing is obtained. And only the object in front of the background partitioning plane is subjected to three-dimensional rendering processing. In general, since many objects behind are stationary or stopped, such as the background, according to the present embodiment, a higher-speed drawing process can be realized as compared with the conventional three-dimensional drawing method, and a smooth moving image with less discomfort can be realized in real time. Can be provided.

Next, another embodiment of the present invention will be described. In the first embodiment described above, the setting of the background reference partition plane in step S30 is based on a fixed value. On the other hand, in the second embodiment, the setting of the reference plane is dynamically changed according to the moving state of the object.

FIG. 11 shows a setting process for dynamically changing the background partition reference plane. In step S30 of the second embodiment, first, the drawing of the previous frame is performed by the normal drawing process (S50).
Is determined (S301). If it is not normal drawing processing, the current position of the partition reference plane 500 (the value in the previous frame) is maintained as it is (S302). In the case of the normal drawing processing, the representative points of the objects are sorted from the depth backward direction (S303), and the movement of the objects is determined in the sorting order (S304). If the object is fixed, the next object is sorted, and the movement of the object is determined again. If it is determined that the object is moving, a background partitioning reference plane 500 having a Z value after the representative point of the object is set (S305).

The operation of the embodiment will now be described. FIGS. 12, 13 and 14 (a) to (g) show an example of the state transition of the contents of the control register in the second embodiment. It is assumed that the contents shown in FIG. 12A are set in each of the foreground partition reference plane data register 106, the object information storage register 107, and the viewpoint information storage register 108 by the initial setting in step S10. Here, since the reference plane Z coordinate is set to 10, the position of the representative point for both the objects 1 and 2 is “forward” of the reference plane 500.

Next, in the setting of the display scene in step S20, the object 1 (ID information number = 1) is set to (1, 1,
4) If it is moved to (2, 1, 4), step S
At 30, the background partitioning reference plane 500 is changed from Z = 10 to Z = 5 as shown in FIG. Next, in the object coordinate comparison process in step S403, the movement of the coordinates of the representative point of the object 1 is determined.
As shown in (d), the object information storage register 107
Are changed.

Next, the frame counter 109 sets the MAXF
Suppose that it is equal to or larger than RAME, and assume that the foreground fetching processing step S70 has been executed. Since the position of the object 2 is determined to be backward by the processing in step S701, the object information storage register 107 is changed as shown in FIG. 13E (S7013).

Next, in the setting of the display scene in step S20, as shown in FIG. 14 (f), the viewpoint coordinates are (1, 2, 3).
⇒ If it is changed to (4,5,6), step S402
In the viewpoint information comparison process, since the moved viewpoint coordinates are recorded, the viewpoint information storage register 106 is changed as shown in FIG.

According to the second embodiment, the movement from the previous frame is monitored by the representative point of the object, so that the group of the stationary objects behind and the group of the moving objects ahead can be monitored. Since the setting of the background reference plane is performed, the reference plane can be changed according to the appearance state of the moving image. In general, a moving image has a high appearance frequency in the front, so that more efficient display is possible according to the present embodiment.

As a third embodiment of the present invention, it is possible to switch between the case where the background reference plane according to the first embodiment is fixed and the case where the background reference plane according to the second embodiment is variable. An example will be described.

FIG. 15 shows a processing flow of setting a reference background partitioning plane (S30) according to the third embodiment. In this process,
Step 30 (FIG. 11) of the second embodiment includes step S
The processing of 300 is added, and thereafter, it is the same as the second embodiment. That is, the reference plane 500 can be controlled to the fixed state or the variable state by operating the operation flag of the background partition reference plane (S300). The operation flag is a variable flag in the initial state, and is switched to a fixed flag when the variable set position of the reference surface 500 is fixed for a fixed number of times or more. In addition, when the frequency of the three-dimensional rendering processing increases in the foreground display using the fixed flag, or when the ratio of the stationary objects included in the foreground display increases, the flag is switched to the variable flag. This makes it possible to set a partition reference plane more suitable for the display state of a moving image, and thus enables more efficient three-dimensional image display.

[0040]

According to the three-dimensional graphic display of the present invention, already processed two-dimensional graphic data is transferred to an object in a scene in which the background in the Z direction of the three-dimensional virtual space is fixed, and geometry processing is performed. Since the time required for ordinary three-dimensional drawing processing such as rasterization processing can be omitted and graphic display can be speeded up, smooth display in real-time processing becomes possible.

[Brief description of the drawings]

FIG. 1 is a configuration diagram of a three-dimensional graphic display device according to an embodiment of the present invention.

FIG. 2 is an explanatory diagram showing an example of a general three-dimensional virtual space.

FIG. 3 is an explanatory diagram showing the definition of objects and their representative points by polygons.

FIG. 4 is a flowchart showing a three-dimensional graphic display method according to an embodiment of the present invention.

FIG. 5 is an explanatory diagram showing a setting method of a background partitioning reference plane in the three-dimensional virtual space of FIG. 2;

FIG. 6 is a flowchart showing details of the drawing process determination (S40) in FIG. 4;

FIG. 7 is a flowchart showing details of a viewpoint coordinate comparison process (S402) in FIG. 6;

FIG. 8 is an object coordinate comparison process in FIG. 6 (S403).
FIG. 4 is a flowchart showing details of the process.

FIG. 9 is a flowchart showing details of the foreground display (S60) in FIG. 4;

FIG. 10 is a flowchart showing details of the background capture (S70) of FIG. 4;

FIG. 11 is a flowchart showing a change unit (background partition reference plane setting method) for FIG. 4 in the three-dimensional graphic display method according to the second embodiment of the present invention.

FIG. 12 is a data transition diagram (part 1) showing a specific operation of the second embodiment.

FIG. 13 is a data transition diagram (part 2) showing a specific operation of the second embodiment.

FIG. 14 is a data transition diagram (part 3) showing a specific operation of the second embodiment.

FIG. 15 is a flowchart showing a method of setting a reference background partitioning plane corresponding to FIG. 11 of the second embodiment in the three-dimensional graphic display method according to the third embodiment of the present invention.

[Explanation of symbols]

101 central processing unit, 102 main memory, 103 background image data, 104 background image information data, 105
Three-dimensional diagram example data, 106 ... Background partition reference plane data register, 107 ... Object information storage register, 10
8: viewpoint information storage register, 109: frame counter, 110: graphic processor, 112: frame buffer, 113: Z buffer, 114: display, 115: input controller, 116: keyboard,
117: mouse, 118: system bus.

Continuing from the front page (72) Nobuo Obata, Inventor 5-2-1 Omikacho, Hitachi City, Ibaraki Prefecture Inside Hitachi Process Computer Engineering Co., Ltd. (72) Inventor Tsutomu Tayama 5-2-1 Omikacho, Hitachi City, Ibaraki Prefecture Hitachi Computer Engineering Co., Ltd. (72) Masamitsu Akatsu, Inventor 5-2-1 Omikacho, Hitachi City, Ibaraki Prefecture Hitachi Process Computer Engineering Co., Ltd. (72) Hideki Fujii, Omikacho, Hitachi City, Ibaraki Prefecture No. 1 F term in Hitachi Ltd. Omika Works (reference) 5B050 BA09 BA11 EA06 EA19 5B057 CA13 CB13 CE08 CH11 CH14 5B080 AA13 CA01 CA05 DA00 FA08 5C082 AA01 BA12 BA43 CB01 DA22 MM02 MM10

Claims (10)

[Claims] 1. A three-dimensional figure which performs three-dimensional drawing processing such as geometry processing and rasterization processing for each frame on a plurality of objects (objects) defined in a three-dimensional space, converts the object into two-dimensional figure data, and displays the two-dimensional figure data. In the display method, the foreground and the foreground are separated by a virtual partition plane set at a predetermined position in the depth (Z direction) of the three-dimensional space (view volume), and whether or not the movement of the foreground has stopped is determined. The image of the stopped background is determined and temporarily stored as a two-dimensional image, and the drawing of the stopped background in subsequent frames is displayed by transferring the temporarily stored two-dimensional image, and then the background is displayed. A three-dimensional graphic display method, wherein a three-dimensional calculation process is performed only on the foreground until a frame in which a motion is detected. 2. A three-dimensional figure which performs three-dimensional drawing processing such as geometry processing and rasterization processing on a frame-by-frame basis for a plurality of objects (objects) defined in a three-dimensional space, converts them into two-dimensional figure data, and displays them. In the display method, a virtual partition plane for dividing a three-dimensional virtual space (view volume) into a foreground and a background is set at a predetermined position in a depth direction (Z direction) from the viewpoint. Wherein the two-dimensional foreground data of the previous frame is reused, the two-dimensional foreground data is generated for the foreground by the three-dimensional rendering process, and the two data are combined and displayed. 3. The three-dimensional graphic display according to claim 2, wherein the virtual partition surface is set by a fixed value and / or a fluctuating value, and the fluctuating value is changed according to the appearance of a moving object. Method. 4. The method according to claim 2, wherein the two-dimensional background data is reused only when an object in the background is not moving, and when there is an object to be moved, the two-dimensional background data is reused. A three-dimensional drawing process. 5. The three-dimensional graphic display method according to claim 4, wherein when the object in the background does not move continuously for a predetermined number of frames, it is determined that there is no movement. 6. The method according to claim 4, wherein the object is defined by a plurality of polygons, and a change in the object is determined by using an arbitrary coordinate value of the polygon as a representative point to determine whether the object has moved. Characteristic three-dimensional figure display method. 7. The two-dimensional background data of the two-dimensional graphics data generated by the three-dimensional rendering process for a plurality of objects in the three-dimensional virtual space according to claim 2, A three-dimensional graphic display method, characterized by storing the three-dimensional figure so that it can be used. 8. A three-dimensional graphic storage means for storing three-dimensional data of a plurality of objects (objects) defined in a three-dimensional virtual space, and converting the three-dimensional data into two-dimensional graphic data and storing the pixel data in a frame buffer. A graphic processor to be expanded, and the frame buffer
A three-dimensional graphic display device comprising a display means for displaying three-dimensional graphic data on a screen and a central processing unit for controlling the readout of the three-dimensional data and the graphic processor; Partition plane data storage means for setting a virtual partition plane for dividing a virtual space (view volume) into a front area and a rear area; viewpoint information storage means for storing viewpoint coordinates for each frame; A background image data storage unit for storing two-dimensional graphic data that has been subjected to drawing processing is provided, and the central processing unit determines whether the viewpoint or the object in the viewpoint and the rear area is moving for each frame, and moves. If not, the two-dimensional graphic data of the rear area in the previous frame is stored in the background image data storage means. Three-dimensional graphic display device is characterized in that so as to transfer al the graphics processor. 9. The apparatus according to claim 8, further comprising a frame counter that counts the number of frames in which the object in the rear area has not moved and clears the frame counter to 0 when the object is moving. A three-dimensional graphic display device, wherein when the number of frames in which a certain object has not moved exceeds a predetermined value, the two-dimensional graphic data in the rear area is transferred. 10. The three-dimensional graphic display device according to claim 8, further comprising an input unit for inputting a virtual partition plane position stored in the partition plane data storage unit.