JPH11197357A - Collision sensing method in three-dimensional video game and video game device using the same and computer-readable medium on which collision sensing program is recorded in three-dimensional video game - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method which can sense a collision between characters in a three-dimensional game precisely and on real time without the need for complicated processing, a video game device using the same, and a computer readable medium on which programs of the method are recorded. SOLUTION: One skeleton object constituting a skeleton model corresponding to a character on an attack side is converted into coordinates of one three- dimensional object constituting a three-dimensional model corresponding to a character on a defense side (s6). And a bilinear curved surface is formed between a frame before the converted skeleton object and a present frame (s7) to judge whether or not the surface intersects the three-dimensional object (s8). This process is repeated for all skeleton objects and three-dimensional objects.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a method for detecting a collision between characters in a three-dimensional video game in which a game character such as a fighting game is displayed in a three-dimensional image, a video game apparatus using the method, and a three-dimensional video. The present invention relates to a computer-readable medium storing a game collision detection program.

[0002]

2. Description of the Related Art Conventional collision detection methods of this type include:
There are a method for detecting a collision between three-dimensional models defined for each character and a method for detecting a collision between skeleton models defined for each character in order to form a three-dimensional image of the character.

[0003]

However, although accurate collision detection is possible with the former method, the computation processing is complicated and enormous, and there is a time lag between the collision between characters on an actual video screen and the detection thereof. Or the speed of movement of the character may be limited, thereby deteriorating the real-time performance of the game.Also, the latter method simplifies the arithmetic processing, but the character processing on the actual video screen is simplified. Despite the collisions, the skeleton models do not collide with each other and cannot be detected in some cases, causing a problem that a sense of discomfort occurs in the game.

[0004] It is an object of the present invention to provide an accurate and real-time detection system that does not require complicated arithmetic processing.
It is an object of the present invention to provide a collision detection method in a three-dimensional video game, a video game device using the same, and a computer-readable medium recording a collision detection program in a three-dimensional video game.

[0005]

In order to achieve the above object, according to the present invention, a collision in a three-dimensional video game for detecting collision between a plurality of characters displayed in a three-dimensional image in which a three-dimensional model and a skeleton model are defined. In the detection method, an intersection state between a three-dimensional model of one character and a skeleton model of the other character is determined, and when the three-dimensional model and the skeleton model intersect, it is determined that the characters collide with each other. .

[0006] According to the above configuration, despite the fact that the characters on the actual video screen collide with each other, it is almost impossible to detect a collision, and there is no need for an enormous amount of arithmetic processing.

Here, specifically, the skeleton model is converted into the coordinate system representation (relative coordinates) of the three-dimensional model, and the relative coordinates of the frame immediately before the skeleton model and the relative coordinates of the current frame are calculated. A bilinear surface is generated, and it is determined whether or not the bilinear surface and the three-dimensional model intersect to determine whether the three-dimensional model and the skeleton model intersect.

More specifically, one solid object is extracted from a plurality of solid objects constituting a solid model corresponding to the entire character, and one skeleton object is extracted from a plurality of skeleton objects constituting a skeleton model corresponding to the entire character. And converts the one skeleton object into a coordinate system representation (relative coordinates) of the one three-dimensional object, and calculates the relative coordinates of the previous frame of the one skeleton object and the relative coordinates of the current frame. A bilinear surface is generated, it is determined whether or not the bilinear surface intersects with the one solid object, and this is repeated for all skeleton objects, and these are repeated for all three-dimensional objects, thereby obtaining a three-dimensional model and a skeleton. Intersect with model By, can detect a collision with simple calculation processing.

The intersection between the bilinear surface and the three-dimensional object is determined by obtaining a section of the bilinear surface and calculating whether or not the section may enter the interior of the three-dimensional object. Arithmetic processing can be further simplified.

More specifically, a section of a bilinear surface is obtained, and it is calculated whether or not the section may enter the inside of the three-dimensional object. If there is a possibility, the bilinear surface is divided and the interval is obtained again, the above calculation is repeated, and if there is a possibility of intersecting until the bilinear surface becomes sufficiently small, the midpoint of the bilinear surface at that time is extracted as an intersection By doing so, the intersection between the bilinear curved surface and the three-dimensional model can be determined accurately and with a simple calculation process.

When the collision detection is applied to a fighting game, it is preferable that the attacking side be a skeleton model and the receiving side be a three-dimensional model. At this time, if collision detection is performed only during the time when at least one character is attacking, it is possible to avoid erroneous recognition of contact between characters in a scene other than the battle scene as a collision.

Further, the video game device of the present invention determines an intersection state between a three-dimensional model of one character and a skeleton model of the other character, and when the three-dimensional model and the skeleton model intersect, the characters collide with each other. Conversion means for converting the skeleton model of one character into the coordinate system representation (relative coordinates) of the three-dimensional model of the other character, and relative coordinates of the frame immediately before the skeleton model Bilinear surface generating means for generating a bilinear surface between the relative coordinates of the current frame and the current frame, and determining whether or not the bilinear surface and the three-dimensional model intersect to determine whether the three-dimensional model and the skeleton model intersect. And an intersection judging means for judging.

[0013] Further, the intersection judging means is constituted by section calculating means for obtaining a section of a bilinear curved surface, and calculating means for calculating whether or not the section may enter the inside of the three-dimensional model. Can be.

Further, the computer-readable medium storing the collision detection program in the three-dimensional video game of the present invention constitutes a three-dimensional model corresponding to the entire character in the computer when the program is read by the computer. One solid object is extracted from a plurality of three-dimensional objects, one skeleton object is extracted from a plurality of skeleton objects constituting a skeleton model corresponding to the entire character, and the one skeleton object is represented in the coordinate system representation of the one solid object ( Relative coordinates), generate a bilinear surface between the relative coordinates of the previous frame of the one skeleton object and the relative coordinates of the current frame,
It is determined whether or not the bilinear surface intersects with the one solid object, this is repeated for all skeleton objects, and these are repeated for all three-dimensional objects to determine the intersection between the three-dimensional model and the skeleton model. The operation is performed.

At this time, a specific intersection determining operation is to obtain a section of a bilinear curved surface, calculate whether or not the section is likely to enter the inside of the three-dimensional object, and determine that there is no intersection if there is no possibility. End, if possible, divide the bilinear surface, find the section again, repeat the above calculation, and if there is a possibility of intersecting until the bilinear surface becomes sufficiently small, the midpoint of the bilinear surface at that time Is extracted as an intersection.

[0016]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS FIG. 1 shows an example of a video game apparatus to which the present invention is applied, here, an outline of a home video game machine, in which 1 is a game machine main body, 2 is a plurality of operation keys. A controller for inputting various commands to the game machine main body 1; a monitor 3 such as a TV connected to the game machine main body 1 for outputting a game screen and music;
Is an external storage device which is mounted on the game machine main body 1 and stores game save data and the like, here a memory card, 5 is a game program executable by the game machine main body 1 and a medium on which data is recorded, here a CD-ROM It is.

The CD-ROM 5 is mounted on a CD-ROM drive (not shown) in which a detachable opening / closing lid 1a is arranged on the upper surface of the game machine main body 1, and its programs and data are read. 1b is a power switch, 1c is a reset switch, and 1d is an open / close button of the open / close lid 1a.

FIG. 2 shows a detailed circuit configuration inside the game machine main body 1, and the present apparatus comprises a CPU (central processing unit) 11.
And an image data generation processor (GTE) 12 directly connected to the CPU 11, a system controller 14, a main memory 15, a system ROM 16, and an image data decompression decoder (MDEC) connected to the CPU 11 via a main bus 13. 17, parallel interface (I / O) 18, serial interface (I / O) 1
9, image processor (GPU) 20, audio processor (SPU) 21, CD-ROM decoder 22, communication interface 23, frame buffer 24 connected to image processor 20, audio processor 2
1, a CD-ROM drive 26 and a CD-ROM buffer 27 connected to the CD-ROM decoder 22.

The image data generation processor 12 includes a CP
Based on the instruction from U11, a large data amount operation such as coordinate conversion or light source calculation is performed. System controller 14
Performs interrupt control, memory access control, and the like. The system ROM 16 stores basic programs and the like. The image data decompression decoder 17 decodes image data compressed and encoded by MPEG, JPEG, or the like.

The image processor 20 creates a video signal of the game screen using the frame buffer 24 in accordance with an instruction from the CPU 11 and outputs the video signal to the monitor 3. The sound processor 21 creates a sound signal including BGM and sound effects using the sound buffer 25 according to an instruction from the CPU 11, and outputs the sound signal to the monitor 3 (or another audio system or the like).
Output to

The CD-ROM decoder 22 is a CD-ROM
5 is read via the CD-ROM drive 26, and is temporarily stored in the CD-ROM buffer 27, and is decoded into the original program and data.

The communication interface 23 is the controller 2
And various signals and data are exchanged with the memory card 4.

FIG. 3 shows the flow of processing of the collision detection method of the present invention.

In the method of the present invention, both a three-dimensional model and a skeleton model are defined for each character for which a collision is to be detected, and a three-dimensional model of one (usually defense) character and another (usually attacker) character are used. Is detected with the skeleton model of the character.

Here, the three-dimensional model is denoted by reference numeral 31 in FIG.
As shown in the figure, the head, the upper half of the torso, the lower half of the torso, the left and right upper arm and forearm, the left and right thighs and the lower part of the knee, such as spheres, cylinders, A plurality of three-dimensional objects 31i such as a rectangular parallelepiped (i = 1, 2, 3,...)
In this case, all three-dimensional objects constituting the three-dimensional model corresponding to each character are created for each frame, and the current frame portion is recorded as a list.

The skeleton model is, as indicated by reference numeral 32 in FIG. 4, the head, the upper half of the torso, the lower half of the torso, the right and left upper arms and the forearm, the left and right thighs and the lower part of the knee. A plurality of skeleton objects 32j (j = 1, 2, 2) corresponding to straight lines imitating individual skeletons of characters such as
...), All skeleton objects constituting a skeleton model corresponding to each character are created for each frame, and the previous frame and the current frame are recorded as a list.

Hereinafter, the method of the present invention will be described with reference to FIG.

First, a value i representing an individual solid object constituting a solid model corresponding to one character and a value j representing an individual skeleton object constituting a skeleton object corresponding to the other character are initialized to "0". (S1). Next, "1" is added to the value i (s2), and it is determined whether or not the number of active three-dimensional objects (the number of all three-dimensional objects in the list) is equal to or more than i (s3). At this time, if it is less than i, the process is terminated, and if it is more than i, “1” is added to the value j (s4), and whether the number of active skeleton objects (the total number of skeleton objects in the list) is not less than j It is determined whether or not it is (s5). At this time, if it is less than j, "1" is added to the value i (s2), and the process returns to s3. If it is more than j, the following process is performed.

That is, the coordinate values at both ends of the j-th active skeleton object relating to the previous frame and the current frame are converted into the coordinate system expression (relative coordinates) of the i-th active solid object relating to the current frame. (S6). Next, the relative coordinates of both ends of the converted j-th active skeleton object for the previous frame and the relative coordinates of both ends of the converted j-th active skeleton object for the current frame are respectively connected, A bilinear curved surface 33 as shown in FIG. 5 is generated (s7). Further, it is determined whether or not the created bilinear curved surface intersects with the i-th solid object (s8).

Further, "1" is added to the value j (s4),
The same processing is repeated for the (j + 1) th skeleton object until the number of active skeleton objects becomes less than j, that is, for all skeleton objects in the list.

Further, "1" is added to the value i (s2),
(I + 1) -th solid object is selected, and the s3
Steps s8 to s8 are repeated until the number of active three-dimensional objects becomes less than i, that is, all three-dimensional objects in the list.

FIG. 6 shows a detailed flow of processing for judging the intersection between a bilinear curved surface and a three-dimensional object.

Here, in order to simplify the determination of the intersection between the solid object represented by the logical product of a plurality of linear or quadratic negative regions and the bilinear surface, an interval operation is used. That is, for example, a section of one bilinear curved surface as indicated by reference numeral 34 in FIG. 7, that is, a bounding box 35 is actually obtained (s8-1), and the section (bounding box) 35 is obtained.
It is determined whether or not there is a possibility of entering the inside of a three-dimensional object having a spherical shape (however, represented by a circle in the drawing) as indicated by reference numeral 36 in FIG. 7 (s8-2). If not, it is determined that they do not intersect.

If there is a possibility, it is determined whether all the lengths of the four sides of the bilinear curved surface are shorter than a predetermined length (s8-
3) If the surface is short, that is, if the curved surface is sufficiently small, the midpoint of the bilinear curved surface at this time is recorded as a contact point (s8-4).

On the other hand, if any one of the four sides of the bilinear surface is not shorter than the predetermined length, as shown in FIG. 8, the bilinear surface, for example, 37 is divided at the midpoint of each side to obtain four bilinear surfaces. The curved surfaces 37a, 37b, 37c, and 37d are used (s8-
5) The process of s8-1 to s8-5 is recursively repeated for each of the four bilinear curved surfaces 37a to 37d after the division.

Note that the characters inevitably come into contact with each other when using a throwing technique in a battle scene, and may also come into contact in a normal conversation scene, but such contact is not mistaken as a collision. Actually, the above-described collision detection is performed only during a period in which at least one character performs a punch attack or a kick attack excluding a throwing technique or the like.

FIG. 9 is a functional block diagram of a three-dimensional video game apparatus using the collision detection method of the present invention. In FIG. 9, reference numeral 41 denotes a program / character data reading / storing unit; Denotes a skeleton model calculation unit, 44 denotes a three-dimensional model calculation unit, 45 denotes a character image creation unit, and 46 denotes a collision detection unit.

The program / character data reading / storing section 41 reads out programs and data from the CD-ROM 5 in which the above-mentioned game programs and character data are stored as necessary, and temporarily stores them.

The command judging section 42 detects a key input operation on the controller 2 and, based on the program stored in the program / character data storage section 41, determines what command (attack command such as punch or kick, defense command, etc.) Command, recovery command, etc.).

The skeleton model calculation unit 43 calculates all the skeleton objects constituting the skeleton model of each character for each frame based on the program and the input command stored in the program / character data storage unit 41. The previous frame and the current frame are held as a list. The three-dimensional model calculation unit 44 calculates, for each frame, all three-dimensional objects constituting the three-dimensional model of each character based on the above-described skeleton model, and holds the current frame as a list.

The character image creating section 45 corrects the shape of the above-mentioned three-dimensional model corresponding to each character, performs image processing such as pasting a texture, etc., and finishes the character image to be actually displayed on a video screen. Note that this character image is combined with a separately created background image to form an image signal, and further output to the monitor 3 together with a separately created music signal.

The collision detecting section 46 is a main component of the present invention for detecting a collision between characters, and includes a collision detecting period determining section 461, an offense and defense determining section 462, a coordinate converting section 463, a bilinear curved surface. The generation unit 464 and the intersection determination unit 46
It consists of five.

The collision detection period determination section 461 limits the period during which collision detection during play should be performed to only the period during which at least one character is performing an attack, particularly a punch or kick attack other than throwing. The motion of each character is analyzed from the input command determined by the command determination unit 42 and the skeleton model calculated by the skeleton model calculation unit 43, and a detection period signal corresponding to the above-described period is output.

The offense / defense determining unit 462 includes the command determining unit 42
It is determined which of the characters in the battle is the attacking side or the defending side based on the input command determined in step (1).

The coordinate conversion unit 463 determines the skeleton model of one of the characters determined to be the attacking side by the offense / defense determining unit 462 as the defending side while the collision detection period determining unit 461 outputs the detection period signal. It is converted into the coordinate system expression (relative coordinates) of the three-dimensional model of the other character as described above.

The bilinear surface generating unit 464 generates a bilinear surface between the relative coordinates of the frame immediately before the skeleton model subjected to the coordinate conversion and the relative coordinates of the current frame as described above.

The intersection determining section 465 includes a section calculating section 465a.
And a calculating unit 465b. The section calculating unit 465a obtains a section of a bilinear surface, that is, a bounding box, and the calculating unit 465b calculates whether the section may enter the inside of the three-dimensional model. Thus, it is determined whether or not the generated bilinear curved surface intersects with the three-dimensional model, thereby determining the intersection between the three-dimensional model and the skeleton model. If a collision is detected, the collision is notified to the skeleton model calculation unit 43 together with the coordinate value, and the collision is reflected in the subsequent motion of the character and its image.

In the above description, an example in which the present invention is applied to a home game machine has been described. However, it is needless to say that the present invention can be applied to a business game machine installed in a game center or an amusement facility.

[0049]

As described above, according to the present invention,
The state of intersection between the three-dimensional model of one character and the skeleton model of the other character is determined. When the three-dimensional model and the skeleton model cross each other, it is determined that the characters collide with each other. Despite the above characters colliding with each other,
It is possible to detect a collision accurately and in real time without almost failing to detect a collision and without requiring a huge amount of arithmetic processing.

[Brief description of the drawings]

FIG. 1 is a diagram showing an outline of an example of a video game apparatus to which the present invention is applied.

FIG. 2 is a detailed circuit configuration diagram inside the game machine main body.

FIG. 3 is a flowchart of a process of a collision detection method according to the present invention.

FIG. 4 shows an example of a three-dimensional model and a skeleton model.

FIG. 5 is a diagram showing how a bilinear surface is generated;

FIG. 6 is a flowchart of a detailed process of determining an intersection between a bilinear surface and a three-dimensional object;

FIG. 7 is an explanatory diagram of a bounding box.

FIG. 8 is a diagram showing how a bilinear surface is divided;

FIG. 9 is a functional block diagram of a three-dimensional video game device using the collision detection method of the present invention.

[Explanation of symbols]

1: Game console body, 1a: Open / close lid, 1b: Power switch, 1c: Reset switch, 1d: Open / close button, 2:
Controller, 2a: operation keys, 3: monitor, 4: memory card, 5: CD-ROM, 11: CPU, 12: processor for generating image data, 13: main bus, 14:
System controller, 15: main memory, 16: system ROM, 17: image data decompression decoder, 1
8: Parallel interface, 19: Serial interface, 20: Image processor, 21: Audio processor, 22: CD-ROM decoder, 23: Communication interface, 24: Frame buffer, 25: Sound buffer, 26: CD-ROM drive , 27: CD
-ROM buffer, 31: solid model, 31i, 36:
Solid object, 32: skeleton model, 32j:
Skeleton objects, 33, 34, 37: bilinear surfaces, 35: bounding boxes, 37a, 37b,
37c, 37d: split bilinear surface, 41: program
Character data readout / storage unit, 42: command determination unit, 43: skeleton model calculation unit, 44: stereo model calculation unit, 45: character image creation unit, 46: collision detection unit, 461: collision detection period determination unit, 462: Offense / defense determining unit, 463: coordinate transformation unit, 464: bilinear surface generation unit,
465: intersection determination section, 465a: section calculation section, 465
b: calculation unit.

Claims (12)

[Claims] 1. A collision detection method in a three-dimensional video game for detecting a collision between a plurality of characters displayed in a three-dimensional image in which a three-dimensional model and a skeleton model are defined. Judge the state of intersection with the skeleton model, when the three-dimensional model and the skeleton model intersect,
A collision detection method in a three-dimensional video game, wherein it is determined that characters have collided. 2. Converting a skeleton model into a coordinate system representation (relative coordinates) of a three-dimensional model, and generating a bilinear surface between relative coordinates of a frame immediately before the skeleton model and relative coordinates of a current frame. 2. The collision detection method in a three-dimensional video game according to claim 1, wherein it is determined whether or not the bilinear curved surface and the three-dimensional model intersect to determine the intersection between the three-dimensional model and the skeleton model. 3. Extracting one three-dimensional object from a plurality of three-dimensional objects constituting a three-dimensional model corresponding to the entire character, extracting one skeleton object from a plurality of skeleton objects constituting a skeleton model corresponding to the entire character, The one skeleton object is converted into the coordinate system representation (relative coordinates) of the one solid object, and a bilinear surface is calculated between the relative coordinates of the previous frame of the one skeleton object and the relative coordinates of the current frame. It is determined whether or not the bilinear surface intersects with the one solid object. This is repeated for all the skeleton objects, and these are repeated for all the three-dimensional objects, whereby the intersection between the three-dimensional model and the skeleton model is obtained. Is characterized by determining The method for detecting a collision in a three-dimensional video game according to claim 1. 4. An intersection of the bilinear surface and the three-dimensional object is determined by calculating a section of the bilinear surface and calculating whether or not the section may enter the inside of the three-dimensional object. Claim 2
Or a collision detection method in a three-dimensional video game according to 3. 5. A section of a bilinear surface is calculated, and it is calculated whether or not the section is likely to enter the inside of the three-dimensional object. The section is obtained by dividing the primary surface again, and the above calculation is repeated. If there is a possibility that the bilinear surface may intersect until the bilinear surface becomes sufficiently small, the midpoint of the bilinear surface at that time is extracted as an intersection, thereby obtaining the bilinear shape. 4. The collision detection method according to claim 2, wherein an intersection between the curved surface and the three-dimensional model is determined. 6. The collision detection method in a three-dimensional video game according to claim 1, wherein, when applied to a fighting game, the attacking side is a skeleton model and the receiving side is a three-dimensional model. 7. The method according to claim 6, wherein the collision is detected only during a time when at least one character is attacking.
The collision detection method in the three-dimensional video game described in the above. 8. A character displayed as a three-dimensional image in which a three-dimensional model and a skeleton model are defined is displayed as a moving image according to a player's operation, and when a collision between the characters is detected, a corresponding image is displayed. The video game device further comprises means for determining an intersection state between the three-dimensional model of one character and a skeleton model of the other character, and determining that the characters collide when the three-dimensional model and the skeleton model intersect. A video game device characterized by the above-mentioned. 9. A conversion means for converting a skeleton model of one character into a coordinate system representation (relative coordinates) of a three-dimensional model of the other character, and a relative coordinate between a frame immediately before the skeleton model and a current frame. Bilinear surface generating means for generating a bilinear surface between coordinates, intersection determining means for determining whether or not the bilinear surface and the three-dimensional model intersect to determine the intersection between the three-dimensional model and the skeleton model; The video game device according to claim 8, further comprising: 10. The intersection determining means includes: a section calculating means for obtaining a section of a bilinear surface; and a calculating means for calculating whether or not the section is likely to enter the inside of the three-dimensional model. The video game device according to claim 9. 11. A computer-readable medium recording a collision detection program in a three-dimensional video game for detecting a collision between characters displayed in a three-dimensional image in which a three-dimensional model and a skeleton model are defined, wherein the three-dimensional video When a collision detection program in a game is read by a computer, the computer extracts one solid object from a plurality of three-dimensional objects constituting a three-dimensional model corresponding to the entire character, and generates a plurality of skeleton models corresponding to the whole character. Take out one skeleton object from the skeleton object, convert the one skeleton object into a coordinate system expression (relative coordinates) of the one solid object, and obtain a relative value of a frame immediately before the one skeleton object. A bilinear surface is generated between the target and the relative coordinates of the current frame, it is determined whether the bilinear surface intersects with the one solid object, and this is repeated for all skeleton objects. A computer-readable medium storing a collision detection program in a three-dimensional video game, wherein an operation of determining an intersection between a three-dimensional model and a skeleton model is performed by repeating for all three-dimensional objects. 12. A section of a bilinear surface is calculated, and it is calculated whether or not the section is likely to enter the inside of the three-dimensional object. If there is no possibility, the processing is terminated as not intersecting. The section is obtained by dividing the primary surface again, and the above calculation is repeated. If there is a possibility that the bilinear surface intersects until the bilinear surface becomes sufficiently small, an operation of extracting the midpoint of the bilinear surface at that time as an intersection is executed. 13. The method according to claim 11, wherein
A computer-readable medium storing a collision detection program for a three-dimensional video game.