KR20240008225A - Motion matching method for generating character motion based on trajectory and velocity and computer apparatus - Google Patents

Abstract

A motion matching method for generating character movement using a movement path and speed includes the steps of a computer device receiving the character's movement path and time information, the computer device preprocessing the movement path, and the computer device performing the preprocessing. Determining a characteristic of a character at a specific point of a movement path, determining, by the computer device, movement information matching in a previously prepared motion database based on the characteristic, and determining, by the computer device, movement information using the motion information. It includes the step of creating character movement that moves based on the path. The characteristics include the position of the character at that point, the position speed of the character, a plurality of future viewpoint points on the movement path, and the orientation direction of the character at the points.

Description

Motion matching method and computer device for generating character movement using movement path and speed {MOTION MATCHING METHOD FOR GENERATING CHARACTER MOTION BASED ON TRAJECTORY AND VELOCITY AND COMPUTER APPARATUS}

The technology described below relates to a motion matching technique that generates the movements of a virtual character.

In digital content such as movies and games, characters move constantly. The character moves along a path according to pre-set information, and sometimes moves along a path by referring to information entered by the user. Therefore, it is important to control the character so that it moves naturally. Motion matching is one of the techniques for synthesizing character movements using motion capture data. Motion matching creates a simple characteristic database and allows you to easily and quickly synthesize natural movements without complex calculations or artificial intelligence.

Korean Patent No. 10-0408428

The technology described below seeks to provide a motion matching technique that generates character movement based on the movement path input by the user and the speed of the character.

A motion matching method for generating character movement using a movement path and speed includes the steps of a computer device receiving the character's movement path and time information, the computer device preprocessing the movement path, and the computer device performing the preprocessing. Determining a characteristic of a character at a specific point of a movement path, and determining, by the computer device, matching movement information in a previously prepared motion database based on the characteristic; and generating, by the computer device, a character movement based on the movement path using the movement information.

A computer device that generates character movement using the movement path and speed includes an input device that receives the character's movement path and time information, a storage device that stores reference motions matching the character's characteristics, and preprocesses the movement path. , Compare the characteristics of the character at a specific point of the preprocessed movement path with the characteristics stored in the storage device to determine matching motion information among the reference motions, and move along the movement path using the motion information. It includes an arithmetic device that generates character motion information.

The characteristics include the position of the character at that point, the position speed of the character, a plurality of future viewpoint points on the movement path, and the orientation direction of the character at the points.

The technology described below generates the character's movement based on the movement path and speed, while creating movement as natural as possible along the movement path.

Figure 1 is an example of a system that generates character movement according to a movement path.
Figure 2 is an example of a path setting process for inputting character movement speed.
Figure 3 is an example of a process for generating character movement along a movement path.
Figure 5 is an example of determining a future path point by proposing the maximum speed of a character.
Figure 6 is an example of character movement when there are sharp corners in the movement path.
Figure 7 is an example of adjusting future path points at sharp corners.
Figure 8 is an example of an artificial neural network that determines the character orientation.
9 is an example of a computer device that generates character movement according to a movement path.

The technology described below may be subject to various changes and may have various embodiments, and specific embodiments will be illustrated in the drawings and described in detail. However, this is not intended to limit the technology described below to specific embodiments, and should be understood to include all changes, equivalents, and substitutes included in the spirit and scope of the technology described below.

Terms such as first, second, A, B, etc. may be used to describe various components, but the components are not limited by the terms, and are only used for the purpose of distinguishing one component from other components. It is used only as For example, a first component may be named a second component without departing from the scope of the technology described below, and similarly, the second component may also be named a first component. The term and/or includes any of a plurality of related stated items or a combination of a plurality of related stated items.

In terms used in this specification, singular expressions should be understood to include plural expressions, unless clearly interpreted differently from the context, and terms such as “including” refer to the described features, numbers, steps, operations, and components. , it means the existence of parts or a combination thereof, but should be understood as not excluding the possibility of the presence or addition of one or more other features, numbers, step operation components, parts, or combinations thereof.

Before providing a detailed description of the drawings, it would be clarified that the division of components in this specification is merely a division according to the main function each component is responsible for. That is, two or more components, which will be described below, may be combined into one component, or one component may be divided into two or more components for more detailed functions. In addition to the main functions it is responsible for, each of the components described below may additionally perform some or all of the functions handled by other components, and some of the main functions handled by each component may be performed by other components. Of course, it can also be carried out exclusively by .

In addition, when performing a method or operation method, each process forming the method may occur in a different order from the specified order unless a specific order is clearly stated in the context. That is, each process may occur in the same order as specified, may be performed substantially simultaneously, or may be performed in the opposite order.

The technology described below generates motion information of a specific object in virtual space using a computer device. The technology described below creates an action in which a specific object moves along a movement path in virtual space.

Hereinafter, the object whose operation is controlled by a computer device is called a character. Characters refer to objects such as people, animals, robots, etc. that appear in games, animations, etc.

The computer device controls the character to have constant motion while moving along the character's movement path. At this time, the movement path may be a path in two-dimensional (2D) or three-dimensional (3D) space. In order to control character movement, a computer device requires a movement path and the characteristics of a character moving along that path at a specific point in time.

Meanwhile, computer devices can be implemented as various devices capable of processing data. For example, a computer device may be implemented as a PC, a server on a network, a smart device, or a chipset with a dedicated program embedded therein.

Figure 1 is an example of a system 100 that generates character movement according to a movement path. In Figure 1, computer devices 110 and 140 generate character movements. In Figure 1, the computer device is shown in the form of a server 130 and a computer terminal 110.

The computer terminal 110 receives the character's movement path input from the user. At this time, the space may be 2D or 3D. The computer terminal 110 receives input not only of the character's movement path but also of the speed at which the character moves along the path. The computer terminal 110 extracts the characteristics of the character from the starting point according to the position and speed along the movement path, queries the motion DB 140 for the extracted features, and receives matching motion data.

The motion DB 140 stores reference motions of the character in advance. The motion DB 140 stores character characteristics and reference motions matching the characteristics. Reference motion is data captured or created in advance by the user.

The computer terminal 110 can generate and output character movements using motion data from the current viewpoint. The computer terminal 110 may determine the movement of the character moving along the movement path by repeating this process.

The server 130 receives the movement path of the character input by the user from the user terminal 120. Additionally, the server 130 receives from the user terminal 120 not only the character's movement path but also the speed at which the character moves along the path. The server 130 extracts the characteristics of the character from the starting point according to the position and speed along the movement path, queries the motion DB 140 for the extracted features, and receives matching motion data. The server 130 may generate character movement using motion data of the current viewpoint. Additionally, the server 130 may transmit the generated motion data to the user terminal 120. The user terminal 120 may output the character movement at that point in time. The server 130 may determine the movement of the character moving along the movement path by repeating this process.

The upper part of FIG. 1 illustrates character movements generated by the computer terminal 110 or the server 130 according to the movement path. Figure 1 shows an example of generating character movements at times t ₀ , t ₁ , t ₂ , and t ₃ over time.

The researcher generated 27-dimensional characteristic data for all movement frames of motion capture data. The researcher defined the characteristics as shown in Equation 1 below. However, the technology described below is not limited to 27-dimensional characteristic design, and is applicable as long as the characteristics include the position and direction of the character after a certain point in the future. The researcher studied objects that walk on two legs, like humans. However, motion matching described below is not necessarily applied to bipedal objects.

The character's position can be defined in a certain coordinate system. i refers to the ith frame. The characteristic is the position of the left foot , left foot speed , position of the right foot ,Speed of right foot and speed of root Includes. The root refers to the pelvic region of a bipedal object. The remaining properties are the root position indicating the future path of the three future (after the i-th frame) frames. and the direction the character is facing (facing) am. t _i is the position on the plane obtained by projecting the character's route onto the floor. The researcher built a motion database based on 30Hz sampling. The researcher set the three future frames as i+10 frame, i+20 frame, and i+30 frame. In the following description, it is assumed that the three future frames are the i+10 frame, the i+20 frame, and the i+30 frame.

Motion matching basically queries features in the next frame ^* The closest feature x _j to is found in the motion database. Among the features is extracted from the current character state, and (t _i , d _i ) is extracted from the path entered by the user. The computer device queries frame j ^* according to equation 2 below: You can find a matching motion according to .

A computer device can determine the character's movement by repeatedly performing motion matching at regular time intervals. When the computer device finds j ^* , the character movement is changed (jumped) to the j ^* th frame, and the character movement is reproduced using consecutive frames after the j ^* th frame in the motion database until the next matching time. The researcher set the matching interval to 0.2 seconds (6 frames).

The computer device receives a movement path input from the user. At this time, the computer device receives input of the location (point) that makes up the path and the speed at that location. The user inputs speed information while entering the movement path using an interface device (pen, mouse, touch, etc.). Figure 2 is an example of a path setting process for inputting character movement speed. The user sets a certain movement path, and at this time, information about the character's movement speed can be set at the same time. For example, when a user inputs a movement path with a mouse, the movement path is entered as a certain line. In this case, the speed at a specific location can be input by varying the drag speed of the mouse. In Figure 2, arrows indicate movement paths. Rectangular points represent points sampled at regular time intervals (eg, 6 frames). In other words, the faster the user enters the movement path, the narrower the spacing between points becomes. Conversely, the slower the user enters the movement path, the wider the spacing between points becomes. The movement path in FIG. 2 consists of a section with a constant interval from the starting point (normal speed section), a section with widening point intervals (low-speed section), and a section with narrow point intervals (high-speed section). Of course, the computer device may receive speed information on the movement path in a manner different from that of FIG. 2.

Figure 3 is an example of a process 200 for generating movement of a character along a movement path. Figure 3 is a schematic example of the process of generating character movement.

Motion data for motion matching must be established in advance in the motion database (210).

The computer device receives movement path and speed information from the user (220). Here, the speed information may be the time for each point for speed calculation. When a user enters a movement path using an interface device, the computer device records the cursor position and the time the position is entered. For convenience of explanation, in the following description, time is used in frames rather than seconds.

Although the user enters a continuous path, a real computing device may record the cursor position at discrete times. This is because the event handler of the input interface may not be called at even intervals. Accordingly, the computer device consistently preprocesses the movement path entered by the user (230). The specific preprocessing process is described later.

The computer device determines future points and orientation based on the movement path (240). As described above, the computer device determines the character's movement by considering speed. At this time, the computer device may adjust future route points to a certain extent, reflecting the factor of speed. This process corresponds to the process of determining feature values or constructing a query to query the character's features. The detailed query construction process is described later.

The computer device queries the motion database for features such as the character's state (position and speed) at the current location, future path points, and the character's orientation direction to select motion data with the most similar features (240).

The computer device repeatedly performs the process of selecting the most similar motion data from the motion database based on the characteristics at the current location until the path movement ends (240-260).

The above-described movement path preprocessing process will be described.

The computer device needs to calibrate the reference points from which the speed is calculated. Once the movement path input is completed, the computer device can interpolate consecutive points on the movement path. For example, a computer device may perform linear interpolation between two adjacent points on a movement path, resampling all points to a fixed frame rate (eg, 150 Hz). That is, the computer device can adjust the time interval between adjacent points to be constant.

Figure 4 is an example of resampling points on a movement path. Figure 4(A) is an example of points (squares) on a movement path initially input by a computer device and the times at which the points were input. Figure 4(B) is an example of points resampled to 150Hz using the linear interpolation described above. In FIG. 4(B), square boxes represent points in FIG. 4(A), and circular dots represent points interpolated at a fixed frame rate based on consecutive points.

Meanwhile, the speed of the movement path entered by the user is faster than the speed at which an object, such as an actual person, moves. Accordingly, the computer device may resample points in the movement path to 150 Hz and set the interval between those points to 30 Hz considering the speed of the object.

Additionally, the path entered by the user and the path along which an object, such as an actual person, moves may differ in shape or form. For example, the path input through a device such as a mouse may differ from the actual human path in terms of the shape near the edges and the type of speed change. This is because the weight of the interface device and the friction when moving the path are different from those of the actual object.

Accordingly, the computer device can consistently preprocess the path entered by the user by considering the movement pattern of the actual object (230). For example, a computer device can smooth a path using a Gaussian filter.

In addition, the computer device can consistently preprocess the beginning and end of the movement path to solve the problem of path shortening due to Gaussian filtering. The computer device may perform extrapolation on the first two points and the last two points of the travel path. At this time, the extrapolation target may be two or more points.

As described above, the character's characteristics for motion matching include information on the future path (position and direction of the character). The computer device extracts information about the future route from the movement route entered by the user. In order to calculate the future path, the computer device must know the desired path point p _d corresponding to the character's current location.

Ideally, if the character were positioned at the next point on the path each frame, the computing device would only need to check the next position each time. However, if the path is drawn quickly, the points on the path may be exhausted too quickly in the process of determining the character's movement. At this time, the computer device may assume the point on the movement path that is closest to the current character's location, but if the paths overlap, it may randomly skip the point or query the same path infinitely.

In order for a computer device to effectively use route points without prematurely consuming them, the current location of the character that satisfies the optimization conditions shown in Equation 3 below can be found. The computer device can use Equation 3 to find the closest point p(i) within a limited search range from the character's current location and find p _d , which is the i _d point on the path.

In Equation 3, p _c is the current position of the character root, is the frame number corresponding to the position obtained in the previous frame. The search range is limited to α and β. The researcher performed motion matching by setting α = 0 and β = 10.

This assumes that the character's position p _d is determined from the near future position rather than returning to the past position.

Motion matching requires a query for the character to follow the path after frame i _d . The computer device may use the same frame interval to extract the character's features when sampling three future path points. For example, 3 future path points , and silver , and It can be. However, this method causes the spacing between points on the future path to become too wide if the input movement path is drawn much faster than the movements in the motion database. In this case, the character may not be able to follow the path properly.

To solve this problem, the researcher limited the character's maximum speed. The character's maximum speed is limited to v _max . Resulting in 3 future path points , and location is limited.

Assume that p(i _vmax ) is a path point reached by moving at the maximum speed v _max for 30 frames from the current frame to the last frame. Last future path point in query If you are located at _{a distance greater than p(i vmax )} along this movement path, is changed to p(i _vmax ). and, and may be resampled at regular time intervals. The computer device is the last point on the future path Change to the point p(i _vmax ) moving at maximum speed, and change the first point to Change =(p(i _d +(i _vmax -i _d )/3), and change the second point to It can be changed to = (p(i _d + 2(i _vmax -i _d )/3). Figure 5 is an example of determining the future route point by proposing the maximum speed of the character. Figure 5(A) shows the user's This is a case where the movement speed of the input character is faster than the movement in the motion database. Looking at Figure 5(A), two of the future path points are farther than the point p(i _vmax ) that can be reached at the maximum speed v _max . Figure 5(B) is an example of adjusting the positions of three future route points according to the speed limit conditions described above.

Meanwhile, the maximum speed v _max can be determined by referring to the speed distribution of movement in the motion database. Additionally, the maximum speed may not use a fixed value but may use a value that changes dynamically based on the speed of the route entered by the user.

If there is a sharp corner in the travel path set by the user, it is difficult to express the sharp corner at future path points that pass through that corner. In this case, the character may have difficulty following sharp edges. Figure 6 is an example of character movement when there are sharp corners in the movement path. Figure 6 shows a case where the current path Pc and future path points are located along an edge. Looking at Figures 6(A) and 6(B), the character moves along the path indicated by the arrow without reaching the edge. A method for solving this problem is explained.

If future paths pass through a sharp edge, the computing device generates a new query by adjusting the points that pass through that edge.

The computer device determines whether future path points are located on sharp corners.

Two future path points with a sharp edge between them and It is named. A sharp corner can be determined by whether the angle of the corner exceeds a threshold. The researcher determined that the two tangent vectors and If the angle formed by exceeds 90 degrees, it was determined to be a sharp corner. If the moment when two tangent lines form the greatest angle is called i _corner , then p(i _corner ) is the point just before the sharp corner.

The computer device is a new point for the J+1th point decides. Can be calculated as in Equation 4 below.

computer devices On the straight line connecting and p(i _corner ) Decide. The computer device identifies a point at a certain distance from p(i _corner ) on the corresponding straight line. decide. k is and the distance between for time Let it be the distance traveled from .

The computer device sets a new J+1th future point and adjusts future points after the J+1th by the same distance.

Figure 7 is an example of adjusting future path points at sharp corners. The future path points in FIG. 7 may be points adjusted according to the maximum speed limit as described above. For convenience of explanation, p _d in FIG. 7 It is displayed as Figure 7 shows future route points , and This is an example showing. Figure 7(A) shows future path points and It shows a situation where there is an edge in between. Figure 7(B) shows the point p(i _corner ) just before the sharp corner. Figure 7(C) shows person About new branch This is an example of setting. Can be determined through Equation 4 described above. Figure 7(D) shows This is an example of newly adjusting the position of . is at the corresponding coordinates go As much as the distance and direction traveled This is the point where the location of was changed. Figure 7(E) is the result of changing the positions of future points in Figure 7(A).

Furthermore, in addition to the method of adjusting future path points described in FIGS. 6 and 7, the computer device may adjust future path points of sharp corners using other algorithms. The computer device has a current point P _d and a future point , and If there is a sharp edge between them, the current point P _d can be newly set. This is to solve an error where the character stays on a sharp corner.

If P _d obtained from the previous frame (i _{d_prev} ) is P(i _{d_prev} ), the computer device can newly set P _d = P(i _{d_prev+n} ). At this time, i _{d_prev+n} represents the nth frame after the previous frame (i _{d_prev} ). n may be an optimal value determined experimentally. Afterwards, the computer device can obtain future points based on the new P _d and then adjust the future points according to the method described in FIGS. 6 and 7.

As described above, motion matching requires not only future path points but also the character's orientation d _i . This explains how to determine the character's direction of movement in three future frames. Future points are continuous points along the movement path, but the character may have different facial or orientation directions while following the points. The researcher determined the direction of the character in two ways. As a result of the experiment, both methods were evaluated as appropriate for character-oriented expression.

One way is to use the tangential direction. The tangent direction vector of the input path at frame i Can be calculated as in Equation 5 below. That is, the tangent direction can be determined based on the positions of two consecutive future path points.

Another method is to use an artificial neural network to determine the character's orientation. At this time, the direction is the direction toward the point that constitutes the future path. The researcher used the Pace network structure of QuaterNet (D. Pavllo et al.i, "Quaternet: A quaternion-based recurrent model for human motion., CoRR, vol. abs/1805.06485, 2018.), which was previously studied. An artificial neural network was constructed to determine the orientation direction. Figure 8 is an example of the artificial neural network 300 for determining the character orientation direction. The artificial neural network 300 includes one GRU (Gated recurrent units, 310) and FC ( It consists of a fully connected layer, 320. The GRU (310) receives the previous point and the character orientation direction at that point. The FC (320) receives the output value of the GRU (310) and receives the character orientation at the current point. Output the direction.

As described above, when a user inputs movement path and speed information, the computer device regularly preprocesses (interpolates and extrapolates) the movement path, constructs a query at regular frame intervals, and then performs motion matching. Generating character movements like this can provide a way for the user and character movements to interact. In other words, the computer device can control future movement by inputting a new movement path while the character is moving or moving along the movement path.

For example, (i) the user can input the character's movement path based on the interface device. At this time, the character's current location corresponds to the starting point of the entered movement path, and the character moves along the movement path. This method can be useful for the user to control the character in real time. (ii) Furthermore, the user can control the character by inputting a movement path based on the space where the character is located. In other words, the character moves to the starting point of the movement path and then moves along the movement path. This method can be used to help a character reach a specific destination, or to control multiple characters simultaneously.

Figure 9 is an example of a computer device 400 that generates character movement according to a movement path. The computer device 400 is a device corresponding to the computer device 110 or 130 of FIG. 1. The computer device 400 may be physically implemented in various forms. For example, the computer device 400 may take the form of a PC, a smart device, a server on a network, or a chipset dedicated to data processing.

The computer device 400 may include a storage device 410, a memory 420, an arithmetic device 430, an interface device 440, a communication device 450, and an output device 460.

The storage device 410 may store a motion database for motion matching. The motion database stores reference images or reference movement data for motion matching.

The storage device 410 can store the character's movement path in a specific coordinate system.

The storage device 410 can store the time at which points forming the route are input along with the movement route.

The storage device 410 may store programs (code) for preprocessing the movement path, determining future path points, generating queries for motion matching, determining the character's orientation, and motion matching based on feature values.

The storage device 410 may store a neural network model that determines the character's orientation.

Additionally, the storage device 410 can store programs or source codes necessary for data processing.

The memory 420 may store data and information generated during the process of the computer device 400 determining the character's actions.

The interface device 440 is a device that receives certain commands and data from the outside. The interface device 440 may receive an input of the character's movement path set by the user. Here, the interface device 440 may be diverse, such as a mouse, tablet, pan mouse, or touch device.

The communication device 450 refers to a configuration that receives and transmits certain information through a wired or wireless network. The communication device 450 can receive the character's movement path set by the user.

The communication device 450 can receive movement information (reference movement) matching the characteristics of the character at the current time from the motion database.

The communication device 450 may transmit motion information at the current time determined by the computer device 400 or the computing device 430 to an external object.

The output device 460 is a device that outputs certain information. The output device 460 can output an interface required for the data processing process. The output device 460 can output the movement path input by the user, the movement path, and the input times of points at regular intervals. Additionally, the output device 460 may output a character moving in a certain space on the screen through motion matching.

The computing device 430 may generate character movement based on the movement path and speed information input by the user.

The computing device 430 may consistently preprocess the movement path entered by the user. Preprocessing may include linear interpolation between movement path points, Gaussian smoothing, and end extrapolation. The preprocessing process is the same as described above.

The arithmetic unit 430 may determine a matching (as close as possible) movement in the motion database based on the characteristics of the character extracted at the current viewpoint.

The computing device 430 may consistently determine feature values for a query. As described above, the computing device 430 may determine future route points based on the movement route. The calculation device 430 can find the closest point of the input movement path to the character's current location. At this time, the computing device 430 can find a nearby point on the movement path within a limited search range. Additionally, the computing device 430 can determine the future path point while limiting the character's speed to a certain level. Additionally, the computing device 430 may constantly adjust the position of future path points at sharp corners. The future route point adjustment process is as described above.

The computing device 430 may determine the character's orientation at the current point in time. As described above, the calculation device 430 can calculate the tangential direction or determine the character's current orientation using an artificial neural network.

The arithmetic device 430 extracts the characteristics of the character at regular time intervals and repeatedly determines the motion information to continuously generate the character's movements.

The computing device 430 may be a device such as a processor that processes data and performs certain operations, an AP, or a chip with an embedded program.

Additionally, the method of generating character motion information or motion control method based on the movement path as described above may be implemented as a program (or application) including an executable algorithm that can be executed on a computer. The program may be stored and provided in a temporary or non-transitory computer readable medium.

A non-transitory readable medium refers to a medium that stores data semi-permanently and can be read by a device, rather than a medium that stores data for a short period of time, such as registers, caches, and memories. Specifically, the various applications or programs described above include CD, DVD, hard disk, Blu-ray disk, USB, memory card, ROM (read-only memory), PROM (programmable read only memory), and EPROM (Erasable PROM, EPROM). Alternatively, it may be stored and provided in a non-transitory readable medium such as EEPROM (Electrically EPROM) or flash memory.

Temporarily readable media include Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDR SDRAM), and Enhanced SDRAM (Enhanced RAM). It refers to various types of RAM, such as SDRAM (ESDRAM), Synchronous DRAM (Synclink DRAM, SLDRAM), and Direct Rambus RAM (DRRAM).

This embodiment and the drawings attached to this specification only clearly show some of the technical ideas included in the above-described technology, and those skilled in the art can easily understand them within the scope of the technical ideas included in the specification and drawings of the above-described technology. It is self-evident that all inferable variations and specific embodiments are included in the scope of rights of the above-mentioned technology.

Claims (16)

A computer device receiving information on the character's movement path and time;
Preprocessing, by the computer device, the movement path;
determining, by the computer device, characteristics of a character at a specific point of the preprocessed movement path;
determining, by the computer device, matching motion information in a previously prepared motion database based on the characteristics; and
Including the step of the computer device generating a character movement based on the movement path using the movement information,
The time information includes times at which points constituting the movement route are input,
The feature is achieved by using the movement path and speed of the character including the position of the character at the corresponding point in time, the position speed of the character, a plurality of future viewpoint points on the movement path, and the orientation direction of the character at the points. Motion matching method to generate movement. According to paragraph 1,
A motion matching method for generating character movement using a movement path and speed, wherein the future viewpoint is one or more viewpoints with a certain time interval starting from the corresponding viewpoint. According to paragraph 1,
The computer device performs the preprocessing of resampling all of the points on a fixed time basis by continuously linearly interpolating between two adjacent points among the points, and generates character movement using the movement path and speed. Matching method. According to paragraph 1,
The computer device performs Gaussian smoothing on the movement path, and performs extrapolation on a plurality of points located at the beginning and a plurality of points located at the end among the points constituting the movement path for which the Gaussian smoothing has been completed. A motion matching method that generates character movement using the movement path and speed for which the preprocessing is performed. According to paragraph 1,
The computer device searches for the specific location closest to the moving path based on the current location of the character, but uses a moving path and speed that constantly limits the search range for the specific location based on the character's previous position. A motion matching method that generates character movement. According to paragraph 1,
The computer device is
Determining the speed of using the points based on the input time of the points constituting the movement path,
If the determined speed exceeds the preset maximum speed, the computer device adjusts the points at the plurality of future points of time among the features, and moves the point at the last point of the plurality of future points of view at the maximum speed. A motion matching method for generating character movement using the movement path and speed of adjusting the position of the case and adjusting the points of the remaining viewpoints except the last viewpoint among the plurality of future viewpoints at equal intervals. According to paragraph 1,
The computer device is
When the angle formed by the tangent lines to the points of the first and second consecutive viewpoints among the plurality of future viewpoints exceeds the threshold,
The character starts from the first point on a straight line connecting a first point at the first viewpoint and a specific point forming the deepest angle on the path between the first point and the second point at the second viewpoint. Adjusting the time-shifted point to the new location of the second point,
A motion matching method for generating character movement using a movement path and speed, where the predetermined time is the time from the first point to the specific point on the movement route. In clause 7,
The computer device newly sets the position of the character at a corresponding point in time to a point after the point in time, sets the plurality of future viewpoints based on the newly set position of the character, and adjusts the position of the second point. A motion matching method that creates character movement using movement path and speed. According to paragraph 1,
The computer device (i) currently sets the current location of the character as the starting point of the moving path and generates a character motion according to the moving path, or (i) currently moves the character from the current location to the starting point of the moving path. A motion matching method that generates character movement using the movement path and speed and then generates character movement according to the movement path. According to paragraph 1,
The computer device is
Regarding any one of the plurality of future points,
By inputting the point immediately before the point and the direction of the character at the point immediately before the point into a previously learned neural network model, the movement path and speed are used to determine the direction of the character at the point. Motion matching method to generate character movement. An input device that receives the character's movement path and time information;
a storage device that stores reference motions matching the characteristics of the character; and
Preprocess the movement path, compare the characteristics of the character at a specific point of the preprocessed movement path with the characteristics stored in the storage device, determine matching motion information among the reference motions, and use the motion information to determine matching motion information. Includes a computing device that generates character motion information moving along the movement path,
The time information includes times at which points constituting the movement route are input,
The feature is achieved by using the movement path and speed of the character including the position of the character at the corresponding point in time, the position speed of the character, a plurality of future viewpoint points on the movement path, and the orientation direction of the character at the points. A computer device that generates movement. According to clause 11,
The processing unit is a computer that generates character movement using the movement path and speed that performs the preprocessing of continuously linear interpolating between two adjacent points among the points and resampling all of the points on a fixed time basis. Device. According to clause 11,
The computing device performs Gaussian smoothing on the movement path and extrapolates a plurality of points located at the beginning and a plurality of points located at the end among the points constituting the movement path for which the Gaussian smoothing has been completed. A computer device that generates character movement using the movement path and speed that performs the preprocessing. According to clause 11,
The computing device determines the speed of using the points based on the input time of the points constituting the movement path,
If the determined speed exceeds the preset maximum speed, the computer device adjusts the points at the plurality of future points of time among the features, and moves the point at the last point of the plurality of future points of view at the maximum speed. A computer device that generates character movement using a movement path and speed that are adjusted to the position of the case and the points of the remaining viewpoints except the last viewpoint among the plurality of future viewpoints are adjusted at equal intervals. According to clause 11,
When the angle formed by the tangent lines to points at consecutive first and second viewpoints among the plurality of future viewpoints exceeds a threshold, the first point at the first viewpoint and the first point and the point where the character moves for a certain period of time starting from the first point on a straight line connecting a specific point forming the deepest angle on the path between the second point at the second viewpoint and the new location of the second point. Adjust,
The predetermined time is a computer device that generates character movement using a movement path and speed, which is the time from the first point to the specific point on the movement route. According to clause 11,
The computing device is
Regarding any one of the plurality of future points,
By inputting the point immediately before the point and the direction of the character at the point immediately before the point into a previously learned neural network model, the movement path and speed are used to determine the direction of the character at the point. A computer device that generates character movements.

Images