JP2001297335A - Interaction operation describing method for three- dimensional animation - Google Patents

Abstract

PROBLEM TO BE SOLVED: To change the contact of a movable character and a static object or to describe interaction operation so as not to require complicated processing without changing the data of the movable character even when the plural contacts exist. SOLUTION: The absolute coordinates of a reference position A of a movable character 10, a scheduled contact position B with a static object 20, a reference position C of the static object 20 and a scheduled contact position D with the character 10 are time sequentially expressed from the time of interaction operation start as plural variables with the coordinates on the side of the object 20 as a reference and by setting the positions of joints of the character 10 and the object 20 on the basis of these respective variables, the interaction operation is described.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to a method for describing an interaction action of a three-dimensional animation for realizing an interaction action between a movable character and a static object.

[0002]

2. Description of the Related Art In a three-dimensional animation in which a character is defined and used as a three-dimensional joint structure having a hierarchical structure, a conventional method for expressing an interaction operation between a character, which is a movable object, and a static object is a conventional method. A method of dividing an operation into a plurality of operations and expressing the operations separately has been considered.

For example, when a character moves in association with a static object, it can be divided into the following three movements assuming contact between the character and the object. One is a case where the character and the object are in contact with each other, and has an effect of moving one with the movement of the other. Next, when the character and the object are completely separated, they operate independently of each other. Finally, in an intermediate state between these, an operation is performed from when a part of the character touches the object to when the character completely touches the object. In the conventional research, the three movements are called holding, reaching, and grasping, respectively, since the movement mainly targets the movement of the character holding the object.

The prior art focuses on automatically generating a natural three-dimensional animation in which a character performs a reaching and a graping action on a general static object. For example, in consideration of a motion in which a character holds a static object with a hand, reaching corresponds to a motion immediately before the character reaches out and touches the object. At this time, the position of the hand immediately before the character touches the object is given, and this is used to make the Inverse Kinematitics (InveseKinemati).
Apply cs) to determine the pose of the arm and interpolate the movement between just before touching the object and the pose before extending the arm.

[0005] Grasping corresponds to an operation of a fingertip or the like for firmly grasping the object after a part of a hand such as a fingertip or a palm of the character touches the object. The character has a plurality of gripping methods of the object as a taxonomy (vocabulary) in advance, and in such an operation, the character obtains the shape information of the object and selects an appropriate gripping method.

[0006] Holding is regarded as a mechanism for an object to follow the hand of a character after glassing. In this case, since the object is held in the hand of the character and follows the movement of the character while being affected by the movement, a framework for dynamically establishing a hierarchical relationship between the character and the object is introduced.

[0007]

In such a conventional technique, the following two problems can be raised.

One is that a character can only come into contact with a static object at a position preset on the character side. In the above-described glassing, in order for the character to grip the object, the position at which the character contacts the object is determined as data held by the character, and the motion of the character relating to the position is prepared as a taxonomy. Therefore, when increasing the number of contact points with the static object, it is necessary to change the character data itself instead of increasing the taxonomy.

The other is that this method cannot generally realize a case where a character and an object touch at a plurality of places. In the above-mentioned holding, the object grasped by the hand is arranged under the hierarchy of the joint of the character. The hierarchical relationship of the joints must form a tree structure for the entire character, but if the object is in contact with the character at multiple locations,
An object is inserted below the hierarchy of a plurality of joints of the character, and the tree structure is broken.
Therefore, in this case, a complicated logic is required in which the static object is placed below the hierarchy of one contact point, and the other contact point follows in a pseudo manner.
An example of such a case is when the character grasps a static object with both hands. It is necessary to separately prepare dynamic logic for controlling the structure.

The present invention has been made in view of such a problem, and changes the contact point in the interaction operation. Even when there are a plurality of contact points, the data of the movable character is not changed and the data is complicated. It is an object of the present invention to provide a method for describing an interaction operation of a three-dimensional animation that does not require complicated processing.

[0011]

SUMMARY OF THE INVENTION In order to solve the above-mentioned problems, the present invention provides a method for realizing an interaction operation between a movable character and a static object in a three-dimensional animation defined by a three-dimensional joint structure. In a method for describing an interaction action of a three-dimensional animation, a reference position of the character, an expected contact position of the character with the object, an absolute position of the reference position of the object, and an expected contact position of the object with the character are respectively indicated. Coordinates, first relative coordinates indicating the amount of movement of the reference position of the object from the start of operation, second relative coordinates from the reference position on the object to the expected contact position,
Third relative coordinates from the expected contact position on the object to the expected contact position on the character, and fourth relative coordinates from the reference position of the object to the reference position of the character, as variables respectively indicating variables;
A method of describing an interaction operation of a three-dimensional animation, wherein the interaction operation is described by setting positions of joints of the character and the object based on motion data in which the variables are stored in time series. Provided.

In such a method for describing an interaction action of a three-dimensional animation, the coordinates of a characteristic portion in the action of the character and the object are represented as variables with reference to the coordinates of the static object, so that the character and the object can be described. Since the interaction behavior is described in a unified manner, even if the contact point is changed, or even if there are multiple contact points, the character's reference position and the expected contact point and the object's contact These operations can be realized without complicating the processing of the joint structure only by newly setting only the variable indicating the relationship with the scheduled point.

[0013]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 shows a form example of a movable character and a static object.

In the present invention, an interaction between a movable character 10 and a static object 20 as shown in FIG. 1 will be considered. The character 10 and the object 20 both have a three-dimensional joint structure having a hierarchical structure. The character 10 is a humanoid character, and its structure is, for example, h-anim (Humanoid Anima).
This is based on the format standardized by the Action Working Group. Here, as an example of the interaction operation, an operation when a human has a static object will be described. In FIG. 1, a bag is illustrated as the object 20.
Here, as a simple example, a case where the character 10 and the object 20 come into contact at one point will be described.

First, a reference position A of the character 10, a contact position B of the character 10 with the object 20, a reference position C of the object 20, and a contact position D of the object 20 with the character are referred to as 4.
Attention is paid to two characteristic portions, and these are determined as shown in FIG. The movements of the character 10 and the object 20 are represented by determining the positions of the joints from the reference position A and the expected contact position B of the character 10 and the reference position C and the contact position D of the object 20. For example, when the absolute coordinates of the reference position A and the expected contact position B of the character 10 are designated, the coordinates of the joint from the root joint to the reference position A and the expected contact position B are determined by inverse kinematics.

In the present invention, the reference position A and the expected contact position B of the character 10 and the reference position C and the expected contact position D of the object 20 are determined based on the coordinates on the object side for controlling the movement of the character 10 and the object 20. Are given as a plurality of variables, and values over time from the start of the interaction operation of these variables are given as motion data. That is, when expressing the interaction action, the absolute coordinates of the reference position A and the expected contact position B of the character 10 and the reference position C and the expected contact position D of the object 20 are obtained from the variables included in the given motion data. From the character 10 and the object 20
Is calculated for each joint.

Assuming that these variables are W, X, Y and Z, from the absolute coordinates of the reference position A and expected contact position B of the character 10 and the reference position C and expected contact position D of the object 20, each variable is It is expressed as follows. Here, the value of a variable s at time t is s
(T), and time t0 is the time at the start of the interaction operation.

The variable W is expressed as W = C (t) -C (t0), and indicates a movement amount of the reference position C of the object 20 from the start of the interaction operation to the time t as relative coordinates from the start position. The variable X is set to X = D−C, and indicates the relative coordinates of the expected contact position D from the reference position C of the object 20. The variable Y is set to Y = BD, and indicates a vector from the expected contact position D on the object 20 to the expected contact position B on the character 10 as relative coordinates. The variable Z is set as Z = A−C, and the character 1 is shifted from the reference position C of the object 20.
0 indicates relative coordinates to the reference position A.

Next, the character 10 and the object 20
The interaction operation with と is divided into three operations of reaching, grasping and holding, and the characteristics of the change of each variable will be described. However, it is assumed that the values of the variables at the joining point between the reaching and the graping and between the graping and the holding are equal.

FIG. 2 shows a character 10 in the case of reaching.
And the displacement of each feature position of the object 20. Reaching is a process until the movable character 10 contacts the static object 20. In FIG.
Object 20 in which humanoid character 10 is a bag
, Reaching up to reach a part of the palm and slightly touching the bag. The value of each variable in this case is as follows.

Since the object 20 does not move autonomously, the variable W always becomes 0 in this process. Further, the character 10 moves toward the expected contact position D of the object 20.
Is a moving process, so the variable X is a constant.

On the other hand, the character 10 moves autonomously, and the variables Y and Z change so as to be finally 0 or a certain constant. The change in the variable Y is the character 10
Means reaching the contact position D of the object 20. The change of the variable Z is the character 1
0 means approaching the object 20.

FIG. 3 shows a character 1 in the case of grassing.
The state of displacement of each characteristic position of 0 and the object 20 is shown. Grasping is a character 10 object 2
This is the process of adjusting the contact point while keeping the contact with zero. FIG.
5 shows an operation in which a part of the palm of the humanoid character 10 slightly contacts the object 20 which is a bag, and then the finger joint is slightly moved to firmly grip the object 20. The value of each variable in this case is as follows.

Since the entire object 20 is not moved by the character 10, the variable W is always a constant. The constant X slightly changes with the delicate movement of the finger until the character 10 firmly grips the object 20. The character 10 and the object 20
Is kept in contact, the variable Y is always 0. Further, since the character 10 does not approach the object 20 any more, the variable Z is always 0 or a constant.

FIG. 4 shows how the characteristic positions of the character 10 and the object 20 are displaced in the case of holding. Holding is a process in which the object 20 moves while being in contact with the character 10 under the influence of the character 10. In FIG. 4, a humanoid character 10
Represents a motion of moving or walking while holding the object 20 which is a bag. The value of each variable in this case is as follows.

The variable W changes under the influence of the motion of the character 10. This change means that the character 10 moves from the viewpoint of the object 20 due to the movement of the hand moving or walking, and conversely, moves the hand to move the object 20 from the viewpoint of the character 10. Means walking or walking.

The character 10 is an object 20
, The variable X is always zero.
Further, since the object 20 and the character 10 operate while being in contact with each other, the variable Y is always 0. Also, the variable Z
Changes depending on how the character 10 holds the object 20.

Next, FIG. 5 shows an embodiment in which the processing of the present invention is executed. In the present invention, as shown in FIG. 5, each of the above variables is read as motion data 51, and each variable is applied to character data 54 and object data 55 using a motion module 52 and an IK (inverse kinematics) module 53. , Character 10 and object 20 are determined. The character data 54 is data describing the shape and joint information of the character 10. When the character 10 is a humanoid, for example, ha which standardizes the joint structure of the humanoid character is used.
The format conforms to the format defined by Nim. The object data 55 is data describing the shape and joint information of the object 20.

The motion data 51 is data in which variables W, X, Y, and Z for controlling the character 10 and the object 20 as described above are stored in a time-series manner. Information is also included. The motion module 52
An apparatus for managing the motion data 51.

The motion module 52 reads the motion data 51 when the interaction operation is activated. During the operation time included in the motion data 51, a variable corresponding to the operation time is continuously set in the IK module 53. Thus, the object data 55 can receive the coordinates of the reference position C and the expected contact position D of the object 20, and the operation of the object 20 is determined.

The IK module 53 receives a variable corresponding to the operation time from the motion module 52, and receives the reference position A and the contact position B of the character 10 from the variable.
Is calculated, and the coordinates of each joint from the root of the character 10 to the corresponding position are calculated using the inverse kinematics algorithm. As a result, the character data 54 corresponding to the character
Are continuously set, and the motion of the character 10 is determined.

As an apparatus for executing the description of the interaction operation, for example, a computer such as a graphic workstation can be considered. In this case, the motion module 52 and the IK module 53 are implemented by a dedicated integrated circuit. Will be realized
Alternatively, it is described as a program that realizes these operations, and is realized by being executed by a processor or the like.

As described above, in the three-dimensional animation interaction operation description method of the present invention, the interaction operation between the movable character 10 and the static object 20 is determined based on the reference position of the character 10 based on the coordinates of the object 20. A and the contact expected position B, the reference position C and the contact expected position D of the object 20 are described by being given, and the data of the contact point is determined in advance in the data of the character 10 as in the related art. It is not necessary, and even when the contact point moves, it can be described only by the change of each variable regardless of the character data 54. For this reason, not only in the case of the above-described example in which the humanoid character 10 lifts the object 20 with one hand and moves, for example, the operation of lifting the object 20 with both hands or alternately holding the object 20 one by one, The same can be applied to the case where there are a plurality of contact points. Such a case is realized by newly setting variables corresponding to the variables X and Y for each contact point and including the variables in the motion data 51, without complicating the processing of the joint structure of the character 10 as in the related art. These actions can be expressed,
Efficient.

In the above embodiment, a bag is exemplified as the static object 20. However, the present invention is not limited to this. For example, the object 20 may be a skateboard.
It is also possible to describe the action of the humanoid character 10 riding on a skateboard. The present invention does not specifically specify a hand motion as an interaction motion. In this case, the same description can be made simply by merely setting the expected contact position B of the character 10 to a foot instead of a hand.

[0035]

As described above, in the method for describing the interaction behavior of a three-dimensional animation according to the present invention, the relationship between the coordinates of the characteristic portions in the motion of the character and the object is based on the coordinates of the static object. By expressing it as a variable, the interaction behavior of the character and the object is described in a unified manner. Therefore, even if the contact point is changed, or even if there are multiple contact points, the character's reference position is not changed without changing the character data. These operations can be realized without complicating the processing of the joint structure only by newly setting only the variables indicating the relationship between the object, the contact point, and the contact point on the object side.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating a form example of a movable character and a static object.

FIG. 2 is a diagram illustrating a state of displacement of each characteristic position of a character and an object in the case of reaching.

FIG. 3 is a diagram illustrating a state of displacement of each characteristic position of a character and an object in the case of glassing.

FIG. 4 is a diagram showing a state of displacement of each characteristic position of a character and an object in the case of holding.

FIG. 5 is a block diagram showing an example of an embodiment for executing the processing of the present invention.

[Explanation of symbols]

10 ... character, 20 ... object, 51 ...
Motion data, 52: Motion module, 5
3 ... IK module, 54 ... Character data, 5
5. Object data

Claims (3)

[Claims] 1. A method for describing an interaction operation of a three-dimensional animation for realizing an interaction operation between a movable character and a static object in a three-dimensional animation defined by a three-dimensional joint structure, comprising the steps of: The absolute coordinates indicating the expected position of the character in contact with the object, the reference position of the object, and the expected position of the object in contact with the character are respectively defined as the reference of the object from the start of the interaction operation. First relative coordinates indicating the movement of the position, second relative coordinates from a reference position on the object to the expected contact position, and third relative coordinates from the expected contact position on the object to the expected contact position on the character.
And relative coordinates of the object and fourth relative coordinates from the reference position of the object to the reference position of the character are respectively represented as variables that are shown in time series from the start of the interaction operation. A method for describing an interaction operation of a three-dimensional animation, wherein the interaction operation is described by setting positions of a joint of a character and the object. 2. The method according to claim 1, wherein the positions of the joints of the character are set by inverse kinematics based on the variables. 3. The method according to claim 1, wherein when the character contacts the object at a plurality of locations, a plurality of combinations of variables indicating the second relative coordinates and the third relative coordinates are set. The object description method of the described three-dimensional animation.