Animation acquisition method, device, equipment and medium

Abstract

The present disclosure provides an animation acquisition method, device, equipment and medium, relating to the technical field of digital human, the method comprising: in response to receiving an input control signal of a target virtual character, acquiring an expected state corresponding to the input control signal to obtain a predicted trajectory of the target virtual character from a current state to the expected state; controlling the target virtual character to move based on the predicted trajectory, and collecting a character movement trajectory of the target virtual character after moving; and determining a next target animation segment of the target virtual character from a preset executable animation segment library based on the character movement trajectory. The degree of artificial dependence of the target virtual character application is reduced, the acquisition efficiency of the animation segment broadcast by the target virtual character is improved, the adaptation of the target animation segment to the target virtual character is improved, the application cost is reduced, the flexibility and controllability of the target virtual character are improved, and the fluency and scene adaptability of the animation broadcast are improved.

Description

Technical Field

[0001] This disclosure relates to the field of computer technology, and in particular to an animation acquisition method, apparatus, device, and medium. Background Technology

[0002] With the development of technology, more and more task scenarios can rely on the performance of digital humans. In the scenario where digital humans are broadcasting, it is necessary to rely on human intervention to manually set the animated clips referenced by the digital human's animated broadcast. This results in a high degree of human intervention and poor efficiency. Summary of the Invention

[0003] The purpose of this disclosure is to at least partially address one of the technical problems in the aforementioned technologies.

[0004] The first aspect of this disclosure provides an animation acquisition method, comprising: in response to receiving an input control signal from a target virtual character, acquiring a desired state corresponding to the input control signal to obtain a predicted trajectory of the target virtual character from the current state to the desired state; controlling the target virtual character to move based on the predicted trajectory, and collecting the character movement trajectory of the target virtual character that has been moved; and determining the next target animation segment to be executed for the target virtual character from a preset executable animation segment library based on the character movement trajectory.

[0005] The animation acquisition method provided in the first aspect of this disclosure also has the following technical features, including:

[0006] According to one embodiment of this disclosure, in response to receiving an input control signal from a target virtual character, obtaining the desired state corresponding to the input control signal to obtain a predicted trajectory of the target virtual character from the current state to the desired state includes: in response to receiving the input control signal, obtaining the initial acceleration of the target virtual character based on the input control signal; obtaining the environmental acceleration of the target virtual character based on its virtual environment, and obtaining the target acceleration of the target virtual character to reach the desired state based on the environmental acceleration and the initial acceleration; and obtaining the predicted trajectory of the target virtual character based on the target acceleration.

[0007] According to one embodiment of this disclosure, obtaining the environmental acceleration of a target virtual character based on its virtual environment, and obtaining the target acceleration of the target virtual character to reach a desired state based on the environmental acceleration and the initial acceleration, includes: obtaining the virtual gravity field acceleration of the target virtual character in the virtual environment; obtaining the virtual collision acceleration of the target virtual character in the virtual collision event in response to the existence of a predicted virtual collision event of the target virtual character; obtaining the environmental acceleration based on the virtual gravity field acceleration and the virtual collision acceleration, and adjusting the initial acceleration based on the environmental acceleration to obtain the target acceleration.

[0008] According to one embodiment of this disclosure, the method further includes: in response to the absence of a predicted virtual collision event for the target virtual character, obtaining environmental acceleration based on virtual gravity field acceleration.

[0009] According to one embodiment of this disclosure, obtaining the predicted trajectory of a target virtual character based on target acceleration includes: obtaining the time step of the target virtual character from its current state to a desired state; obtaining a target predicted velocity sequence, a target predicted position sequence, and a target predicted joint angle sequence of the target virtual character within the time step based on the time step and the target acceleration; and obtaining a predicted trajectory segment of the target virtual character within the time step based on the target predicted velocity sequence, the target predicted position sequence, and the target predicted joint angle sequence, so as to obtain the predicted trajectory of the target virtual character.

[0010] According to one embodiment of this disclosure, obtaining a target predicted velocity sequence of a target virtual character within a time step based on a time step and target acceleration includes: acquiring multiple prediction time points included in the time step; for any character skeleton point of the target virtual character, acquiring the initial velocity of the character skeleton point in the current state, and based on the initial velocity, target acceleration, and time step, obtaining the predicted velocity of the character skeleton point at each prediction time point among the multiple prediction time points, so as to obtain a point predicted velocity sequence of the character skeleton point; and obtaining a target predicted velocity sequence of the target virtual character within the time step based on the point predicted velocity sequences of each character skeleton point.

[0011] According to one embodiment of this disclosure, based on a time step and target acceleration, a target predicted position sequence of a virtual character within a time step is obtained, including: for any character skeleton point of the target virtual character, based on the initial velocity, target acceleration, and time step of the character skeleton point in the current state, obtaining the predicted point position of the character skeleton point at each of the multiple predicted time points, so as to obtain a predicted point position sequence of the character skeleton point; and based on the predicted point position sequence of each character skeleton point, obtaining a target predicted position sequence of the virtual character within a time step.

[0012] According to one embodiment of this disclosure, a target predicted joint angle sequence of a target virtual character within a time step is obtained based on a time step and target acceleration, including: obtaining the joint angle variables of each part of the target virtual character within the time step to obtain the predicted joint angle of each part at the end of the time step; and obtaining the target predicted joint angle sequence of the target virtual character at the end of the time step based on the predicted joint angle of each part.

[0013] According to one embodiment of this disclosure, controlling a target virtual character to move based on a predicted trajectory and collecting the character movement trajectory of the target virtual character includes: collecting the movement position and movement direction of each character skeleton point in the target virtual character when the target virtual character moves according to the predicted trajectory, so as to obtain the character movement trajectory.

[0014] According to one embodiment of this disclosure, determining the next target animation segment to be executed for a target virtual character from a preset executable animation segment library based on the character's movement trajectory includes: obtaining a reference virtual character and a reference movement trajectory of the reference virtual character in each executable animation segment in the executable animation segment library; and determining the next target animation segment to be executed from the executable animation segment library based on the target virtual character, the reference virtual character, the character's movement trajectory, and the reference movement trajectory.

[0015] According to one embodiment of this disclosure, determining the next target animation segment to be executed from an executable animation segment library based on a target virtual character, a reference virtual character, a character movement trajectory, and a reference movement trajectory includes: obtaining a character matching degree between the reference virtual character and the target virtual character; obtaining a trajectory matching degree between the reference movement trajectory and the character movement trajectory; and determining the next target animation segment to be executed from the executable animation segment library based on the character matching degree and the trajectory matching degree.

[0016] According to one embodiment of this disclosure, obtaining the character matching degree between a reference virtual character and a target virtual character includes: for any character skeletal point in the target virtual character, obtaining the associated skeletal point on the reference virtual character corresponding to the character skeletal point, so as to obtain a set of skeletal point difference parameters between the character skeletal point and the associated skeletal point, wherein the set of skeletal point difference parameters includes at least a first position feature difference parameter, a first direction feature difference parameter, and a first velocity feature difference parameter between the character skeletal point and the associated skeletal point; and obtaining the character matching degree between the target virtual character and the reference virtual character based on the set of skeletal point difference parameters for each character skeletal point.

[0017] According to one embodiment of this disclosure, obtaining the trajectory matching degree between a reference movement trajectory and a character movement trajectory includes: for any character trajectory point in the character movement trajectory, obtaining the associated trajectory point corresponding to the character trajectory point in the reference movement trajectory, so as to obtain a trajectory point difference parameter set between the character trajectory point and the associated trajectory point, wherein the trajectory point difference parameter set includes at least a second position feature difference parameter, a second direction feature difference parameter and a second velocity feature difference parameter between the character trajectory point and the associated trajectory point; and obtaining the trajectory matching degree between the reference movement trajectory and the character movement trajectory based on the trajectory point difference parameter set of each character trajectory point.

[0018] According to one embodiment of this disclosure, determining the next target animation segment to be executed from the executable animation segment library based on character matching degree and trajectory matching degree includes: for any executable animation segment, integrating the character matching degree and trajectory matching degree corresponding to the executable animation segment to obtain the target matching degree between the executable animation segment and the target virtual character; and determining the target animation segment from the executable animation segment library based on the target matching degree of each executable animation segment.

[0019] According to one embodiment of this disclosure, the method further includes: performing image frame segmentation on the original animation segment to obtain each frame image included in the original animation segment; extracting the original character skeleton points in the original virtual character of the original animation segment, and obtaining the skeleton point features of the original character skeleton points based on each frame image; obtaining the original movement trajectory of the original virtual character in the original animation segment; and obtaining the executable animation segment corresponding to the original animation segment based on the original movement trajectory and the skeleton point features of the original character skeleton points, so as to obtain an executable animation segment library.

[0020] A second aspect of this disclosure provides an animation acquisition device, comprising: a prediction module, configured to, in response to receiving an input control signal from a target virtual character, acquire a desired state corresponding to the input control signal, so as to obtain a predicted trajectory of the target virtual character from the current state to the desired state; a control acquisition module, configured to control the target virtual character to move based on the predicted trajectory, and acquire the character movement trajectory of the target virtual character that has been moved; and a determination module, configured to determine the next target animation segment to be executed for the target virtual character from a preset executable animation segment library based on the character movement trajectory.

[0021] The second aspect of this disclosure provides an animation acquisition device, which also has the following technical features, including: According to one embodiment of this disclosure, the prediction module is further configured to: in response to receiving an input control signal, acquire the initial acceleration of the target virtual character based on the input control signal; acquire the environmental acceleration of the target virtual character based on its virtual environment, and based on the environmental acceleration and the initial acceleration, acquire the target acceleration of the target virtual character to reach the desired state; and acquire the predicted trajectory of the target virtual character based on the target acceleration.

[0022] According to one embodiment of this disclosure, the prediction module is further configured to: obtain the virtual gravity field acceleration of the target virtual character in the virtual environment; in response to the existence of a predicted virtual collision event of the target virtual character, obtain the virtual collision acceleration of the target virtual character in the virtual collision event; obtain the environmental acceleration based on the virtual gravity field acceleration and the virtual collision acceleration, and adjust the initial acceleration based on the environmental acceleration to obtain the target acceleration.

[0023] According to one embodiment of this disclosure, the prediction module is further configured to: in response to the absence of a predicted virtual collision event for the target virtual character, obtain environmental acceleration based on virtual gravity field acceleration.

[0024] According to one embodiment of this disclosure, the prediction module is further configured to: obtain the time step of the target virtual character from the current state to the desired state; based on the time step and the target acceleration, obtain the target predicted velocity sequence, the target predicted position sequence and the target predicted joint angle sequence of the target virtual character within the time step; based on the target predicted velocity sequence, the target predicted position sequence and the target predicted joint angle sequence, obtain the predicted trajectory segment of the target virtual character within the time step, so as to obtain the predicted trajectory of the target virtual character.

[0025] According to one embodiment of this disclosure, the prediction module is further configured to: obtain multiple prediction time points included in the time step; for any character skeleton point of the target virtual character, obtain the initial velocity of the character skeleton point in the current state, and based on the initial velocity, target acceleration, and time step, obtain the predicted velocity of the character skeleton point at each of the multiple prediction time points to obtain a point prediction velocity sequence of the character skeleton point; and based on the point prediction velocity sequence of each character skeleton point, obtain the target prediction velocity sequence of the target virtual character within the time step.

[0026] According to one embodiment of this disclosure, the prediction module is further configured to: for any character skeleton point of the target virtual character, based on the initial velocity, target acceleration and time step of the character skeleton point in the current state, obtain the predicted point position of the character skeleton point at each of the multiple prediction time points, so as to obtain the predicted point position sequence of the character skeleton point; and based on the predicted point position sequence of each character skeleton point, obtain the target predicted position sequence of the target virtual character within the time step.

[0027] According to one embodiment of this disclosure, the prediction module is further configured to: obtain the joint angle variables of each part of the target virtual character within a time step, so as to obtain the predicted joint angle of each part at the end of the time step; and obtain the target predicted joint angle sequence of the target virtual character at the end of the time step based on the predicted joint angle of each part.

[0028] According to one embodiment of this disclosure, the control acquisition module is further configured to: acquire the movement position and movement direction of each character skeleton point in the target virtual character when the target virtual character moves according to the predicted trajectory, so as to obtain the character movement trajectory.

[0029] According to one embodiment of this disclosure, the determining module is further configured to: obtain a reference virtual character and a reference movement trajectory of the reference virtual character in each executable animation segment in the executable animation segment library; and determine the next target animation segment to be executed from the executable animation segment library based on the target virtual character, the reference virtual character, the character movement trajectory, and the reference movement trajectory.

[0030] According to one embodiment of this disclosure, the determining module is further configured to: obtain the character matching degree between the reference virtual character and the target virtual character; obtain the trajectory matching degree between the reference movement trajectory and the character movement trajectory; and determine the next target animation segment to be executed from the executable animation segment library based on the character matching degree and the trajectory matching degree.

[0031] According to one embodiment of this disclosure, the determining module is further configured to: for any character skeleton point in the target virtual character, obtain the associated skeleton point on the reference virtual character corresponding to the character skeleton point, so as to obtain a set of skeleton point difference parameters between the character skeleton point and the associated skeleton point, wherein the set of skeleton point difference parameters includes at least a first position feature difference parameter, a first direction feature difference parameter, and a first velocity feature difference parameter between the character skeleton point and the associated skeleton point; and obtain the character matching degree between the target virtual character and the reference virtual character based on the set of skeleton point difference parameters of each character skeleton point.

[0032] According to one embodiment of this disclosure, the determining module is further configured to: for any character trajectory point in the character movement trajectory, obtain the associated trajectory point corresponding to the character trajectory point in the reference movement trajectory, so as to obtain a set of trajectory point difference parameters between the character trajectory point and the associated trajectory point, wherein the set of trajectory point difference parameters includes at least a second position feature difference parameter, a second direction feature difference parameter and a second velocity feature difference parameter between the character trajectory point and the associated trajectory point; and obtain the trajectory matching degree between the reference movement trajectory and the character movement trajectory based on the set of trajectory point difference parameters for each character trajectory point.

[0033] According to one embodiment of this disclosure, the determining module is further configured to: for any executable animation segment, integrate the character matching degree and trajectory matching degree corresponding to the executable animation segment to obtain the target matching degree between the executable animation segment and the target virtual character; and determine the target animation segment from the executable animation segment library based on the target matching degree of each executable animation segment.

[0034] According to one embodiment of this disclosure, the apparatus further includes an acquisition module, configured to: segment the original animation clip into image frames to obtain each frame image included in the original animation clip; extract the original character skeleton points in the original virtual character of the original animation clip, and obtain the skeleton point features of the original character skeleton points based on each frame image; acquire the original movement trajectory of the original virtual character in the original animation clip; and obtain the executable animation clip corresponding to the original animation clip based on the original movement trajectory and the skeleton point features of the original character skeleton points, so as to obtain an executable animation clip library.

[0035] To achieve the above objectives, a third aspect of this disclosure provides an electronic device, including: a memory and a processor; wherein the processor reads executable program code stored in the memory to run a program corresponding to the executable program code, so as to implement the animation acquisition method provided in the first aspect of this disclosure.

[0036] To achieve the above objectives, the fourth aspect of this disclosure provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the animation acquisition method provided in the first aspect of this disclosure.

[0037] To achieve the above objectives, the fifth aspect of this disclosure provides a computer program product, including a computer program that, when executed by a processor, is used to implement the animation acquisition method provided in the first aspect of this disclosure.

[0038] The animation acquisition method and apparatus disclosed herein predict the trajectory of a target virtual character from its current state to the desired state based on the desired state corresponding to the input control signal, thereby obtaining a predicted trajectory. Based on the character's movement trajectory collected as the target virtual character moves along the predicted trajectory, the next executable animation segment for the target virtual character is obtained from a pre-set set of executable animation segments. This eliminates the need for manual setting of each animation segment played by the target virtual character, reducing the manual dependence on the application of the target virtual character, improving the acquisition efficiency of the animation segments played by the target virtual character, and enhancing the compatibility between the target animation segments and the target virtual character. Upon receiving an input control signal, the movement of the target virtual character is controlled by acquiring the predicted trajectory, eliminating the need for separate development of the actions required by the target virtual character. This reduces the application cost of the target virtual character, improves the flexibility and controllability of the target virtual character in application scenarios, thereby improving the smoothness and scene adaptability of the target virtual character's animation playback and optimizing the user's viewing experience.

[0039] Additional aspects and advantages of this disclosure will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of this disclosure. Attached Figure Description

[0040] The above and / or additional aspects and advantages of this disclosure will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, in which: Figure 1 This is a schematic diagram of an animation acquisition method according to an embodiment of the present disclosure; Figure 2 This is a schematic diagram of an animation acquisition method according to another embodiment of the present disclosure; Figure 3 This is a schematic diagram of the movement trajectory according to an embodiment of the present disclosure; Figure 4 This is a schematic diagram of the skeletal points of a target virtual character and a reference virtual character according to an embodiment of this disclosure; Figure 5 This is a schematic diagram of the trajectory points of a character movement trajectory and a reference movement trajectory according to an embodiment of this disclosure. Figure 6 This is a schematic diagram of an animation acquisition method according to another embodiment of the present disclosure; Figure 7 This is a schematic diagram of the structure of an animation acquisition device according to an embodiment of the present disclosure; Figure 8 This is a block diagram of an electronic device according to an embodiment of the present disclosure. Detailed Implementation

[0041] Embodiments of this disclosure are described in detail below, examples of which are illustrated in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and intended to explain this disclosure, and should not be construed as limiting this disclosure.

[0042] The acquisition, storage, use, and processing of data in this disclosed technical solution all comply with the relevant provisions of relevant laws and regulations.

[0043] The following description, with reference to the accompanying drawings, outlines an animation acquisition method, apparatus, device, and medium according to embodiments of the present disclosure.

[0044] Figure 1 This is a flowchart illustrating an embodiment of the animation acquisition method disclosed herein, as shown below. Figure 1 As shown, the method includes: S101, in response to receiving the input control signal of the target virtual character, obtain the desired state corresponding to the input control signal, so as to obtain the predicted trajectory of the target virtual character from the current state to the desired state.

[0045] In this embodiment of the disclosure, the character currently performing an animated segment can be identified as the target virtual character. The target virtual character can be a digital human currently broadcasting, or it can be other types of characters currently performing animated segments. No specific limitation is made here.

[0046] In a scenario where a target virtual character is performing an animated segment, when there is an external input control signal, the target virtual character can adjust and modify its subsequent virtual actions based on the external input control signal. In this case, the external control signal received by the target virtual character can be determined as the target virtual character's input control signal.

[0047] In this embodiment of the disclosure, the input signal can be detected during the animation performance of the target virtual character using a signal detection method in the related art, thereby obtaining the input control signal of the target virtual character.

[0048] In a scenario where an input control signal from a target virtual character is received, the input control signal can be processed using a signal parsing algorithm based on relevant technologies. The state information carried in the input control signal can be obtained based on the result of the algorithm processing. Based on this state information, the state that the target virtual character needs to achieve under the control of the input control signal can be determined. In this scenario, the state that the target virtual character needs to achieve can be determined as the desired state of the target virtual character.

[0049] In this embodiment of the disclosure, the current state of the target virtual character can be obtained and compared and analyzed with the desired state. Then, through the trajectory planning algorithm in the related technology, the movement trajectory required for the target virtual character to reach the desired state from the current state can be obtained based on the difference between the current state and the desired state. This trajectory can be determined as the predicted trajectory required for the target virtual character to reach the desired state.

[0050] S102, control the target virtual character to move based on the predicted trajectory, and collect the character movement trajectory of the target virtual character that has moved.

[0051] In this embodiment of the disclosure, a target virtual character can be controlled by a virtual character control method in the related art, so that it can move along a predicted trajectory.

[0052] During the movement of the target virtual character along the predicted trajectory, information on the trajectory that the target virtual character has already moved can be collected, and the collected trajectory can be determined as the character movement trajectory of the target virtual character.

[0053] S103, determine the next target animation segment to be executed for the target virtual character from the preset executable animation segment library based on the character's movement trajectory.

[0054] In this embodiment of the disclosure, a portion of animation clips can be obtained in advance. These animation clips match the conditions required for the target virtual character to perform the animation clips. In this scenario, these animation clips can be identified as executable animation clips, and the database obtained based on each executable animation clip can be identified as an executable animation clip library.

[0055] In scenarios where the target virtual character receives an input control signal, the animation segment that the target virtual character performs after reaching the desired state may differ from the originally determined animation segment. In such cases, it is necessary to obtain a new animation segment for the target virtual character from the executable animation segment library. This new animation segment can be identified as the animation segment that the target virtual character needs to perform after reaching the desired state, which is the next target animation segment to be executed by the target virtual character.

[0056] The animation acquisition method proposed in this disclosure predicts the trajectory of a target virtual character from its current state to the desired state based on the desired state corresponding to the input control signal, obtaining a predicted trajectory. Then, based on the character's movement trajectory collected as the target virtual character moves along the predicted trajectory, the next executable animation segment for the target virtual character is obtained from a pre-set set of executable animation segments. This eliminates the need for manual setting of each animation segment played by the target virtual character, reducing the manual dependence on the application of the target virtual character, improving the acquisition efficiency of the animation segments played by the target virtual character, and enhancing the compatibility between the target animation segments and the target virtual character. Upon receiving an input control signal, the movement of the target virtual character is controlled by acquiring the predicted trajectory, eliminating the need for separate development of the actions required by the target virtual character. This reduces the application cost of the target virtual character, improves its flexibility and controllability in application scenarios, thereby enhancing the smoothness and scene adaptability of the target virtual character's animation playback and optimizing the user's viewing experience.

[0057] In the above embodiments, the acquisition of the predicted trajectory of the target virtual character can be combined with... Figure 2 To understand further, Figure 2 This is a flowchart illustrating an animation acquisition method according to another embodiment of the present disclosure. The method includes: S201, in response to receiving the input control signal, acquire the initial acceleration of the target virtual character based on the input control signal.

[0058] In this embodiment of the present disclosure, when an input control signal of a target virtual character is received, the input control signal can be processed by an acceleration acquisition algorithm in the related technology. Based on the result of the algorithm processing, the acceleration vector obtained by the target virtual character under the control of the input control signal is obtained, and the acceleration vector is determined as the initial acceleration of the target virtual character.

[0059] As an example, the acquisition of initial acceleration can be understood in conjunction with the following: The input vector of the input control signal is set as If its corresponding mapping matrix is ​​M, then the initial acceleration The formula for obtaining it is It should be noted that the mapping matrix M needs to satisfy the physical rationality constraints defined in the relevant technologies.

[0060] S202, obtain the environmental acceleration of the target virtual character based on its virtual environment, and obtain the target acceleration of the target virtual character to reach the desired state based on the environmental acceleration and the initial acceleration.

[0061] In this embodiment of the disclosure, in order to make the performance state of the target virtual character conform to the physical rationality constraints in the related technology, a corresponding animation performance environment can be constructed for the target virtual character, that is, the virtual environment to which the target virtual character belongs. In this scenario, the initial acceleration of the target virtual character will be affected by the virtual environment. The acceleration generated by the environmental influence of the virtual environment on the target virtual character can be obtained, and this acceleration can be determined as the environmental acceleration of the virtual environment.

[0062] Among them, environmental acceleration can include the acceleration generated by the virtual gravity field in the virtual environment acting on the target virtual character, the acceleration generated by the collision event that occurs in the virtual environment, and the acceleration generated by other influences of the virtual environment on the target virtual character. No specific limitation is made here.

[0063] Optionally, the virtual gravitational field acceleration of the target virtual character in the virtual environment can be obtained.

[0064] In this embodiment of the disclosure, the gravitational acceleration of the target virtual character can be obtained through the gravitational acceleration acquisition algorithm in the related technology, and determined as the virtual gravitational field acceleration of the target virtual character in the virtual environment.

[0065] As an example, the formula for obtaining the acceleration of a virtual gravitational field can be as follows:

[0066] in, represents the virtual gravitational acceleration of the target virtual character, and g represents the gravitational constant of the virtual environment's gravitational field.

[0067] Optionally, in response to the existence of a predicted virtual collision event for the target virtual character, the virtual collision acceleration of the target virtual character in the virtual collision event is obtained.

[0068] In this embodiment of the disclosure, the target virtual character may or may not experience a collision event in the virtual environment. In this scenario, it is possible to predict whether a collision event may occur when the target virtual character moves based on the input control signal. When it is predicted that the target virtual character may experience a collision event under the influence of the input control signal, the event can be identified as the virtual collision event predicted for the target virtual character.

[0069] In this scenario, the acceleration generated by the virtual collision event acting on the target virtual character can be obtained through acceleration acquisition methods in related technologies. This acceleration can be determined as the virtual collision acceleration of the target virtual character in the virtual collision event.

[0070] As an example, virtual collision acceleration can be obtained as follows:

[0071] in, Represents virtual collision acceleration. Represents the velocity at time t+1. The damping coefficient is represented by k, and the elastic coefficient is represented by k. This represents the unit vector of the tangent plane.

[0072] Optionally, the environmental acceleration is obtained based on the virtual gravity field acceleration and the virtual collision acceleration, and the initial acceleration is adjusted based on the environmental acceleration to obtain the target acceleration.

[0073] In this embodiment of the disclosure, the virtual gravity field acceleration and the virtual collision acceleration can be integrated using an acceleration integration algorithm in the related technology, thereby achieving the integration of the two, and the result of the integration is determined as the environmental acceleration generated by the virtual environment acting on the target virtual character.

[0074] Among them, the virtual gravity field acceleration and the virtual collision acceleration can be summed, and the sum can be determined as the environmental acceleration generated by the virtual environment acting on the target virtual character.

[0075] Optionally, in response to the absence of a predicted virtual collision event for the target virtual character, environmental acceleration is obtained based on virtual gravity field acceleration.

[0076] In this embodiment of the disclosure, when the target virtual character is affected by the input control signal, a collision event may not occur in the virtual environment. That is to say, there is no predicted virtual collision event for the target virtual character. In this scenario, the environmental acceleration generated by the virtual environment on the target virtual character can be obtained based on the acceleration generated by the virtual gravity field acceleration and other environmental influences acting on the target virtual character.

[0077] It should be noted that in scenarios where environmental acceleration is based on virtual gravity field acceleration and virtual collision acceleration, when the target virtual character does not have a predicted virtual collision event, the virtual gravity field acceleration can be directly used as the environmental acceleration.

[0078] In this embodiment of the disclosure, the initial acceleration can be adjusted based on the environmental acceleration using the acceleration adjustment method in the related art, and the adjusted acceleration can be determined as the target acceleration of the target virtual character.

[0079] The environmental acceleration can be added to the initial acceleration, or other algorithms can be applied to the environmental acceleration and the initial acceleration; no specific limitations are specified here.

[0080] S203, based on the target acceleration, obtain the predicted trajectory of the target virtual character.

[0081] In this embodiment of the disclosure, a trajectory prediction algorithm in the related art can be used to predict the trajectory that the target virtual character needs to move from the current state to the desired state based on the target acceleration, thereby obtaining the corresponding predicted trajectory.

[0082] Optionally, the time step from the current state to the desired state of the target virtual character is obtained, and based on the time step and the target acceleration, the target predicted velocity sequence, target predicted position sequence and target predicted joint angle sequence of the target virtual character within the time step are obtained.

[0083] In this embodiment of the disclosure, the process of the target virtual character reaching the desired state from the current state requires a certain amount of time, which may include at least one time step. In this scenario, the trajectory of the target virtual character within any time step can be obtained, and then based on the integration of the trajectories under each time step, the predicted trajectory of the target virtual character from the current state to the desired state can be obtained.

[0084] In this scenario, for any given time step, the parameters required for trajectory prediction of the target virtual character within that time step can be obtained. Then, using trajectory prediction algorithms in related technologies, the trajectory of the target virtual character at that time step can be predicted based on these parameters, thereby obtaining the predicted trajectory segment of the target virtual character at that time step.

[0085] In some implementations, the predicted movement speed, predicted movement position, and joint angles of each part of the target virtual character within any time step can be obtained, thereby enabling the prediction of the trajectory segment at that time step. In this case, the predicted parameters of the target virtual character can be obtained by predicting the parameters of each skeletal point included in the target virtual character. The skeletal points on the target virtual character can be identified as character skeletal points.

[0086] In this scenario, multiple predicted time points can be obtained within the time step.

[0087] In this embodiment of the disclosure, any time step can be segmented based on a preset time interval, and the endpoints of each segment obtained can be determined as multiple predicted time points included in the time step.

[0088] Optionally, for any character skeleton point of the target virtual character, the initial velocity of the character skeleton point in the current state is obtained, and based on the initial velocity, target acceleration and time step, the predicted velocity of the character skeleton point at each of the multiple prediction time points is obtained, so as to obtain the point prediction velocity sequence of the character skeleton point.

[0089] For any character skeleton point, the velocity data of the character skeleton point in the current state can be collected using the velocity acquisition method in the relevant technology, and then the collected velocity data can be determined as the initial velocity of the character skeleton point.

[0090] It should be noted that the target acceleration of the target virtual character applies to all the character skeleton points included on the target virtual character.

[0091] Among them, the initial velocity, target acceleration and time step of the character's skeleton point can be processed by the velocity prediction algorithm in the relevant technology, so as to obtain the velocity of the character's skeleton point at any prediction time point based on the result of the algorithm processing, which is used as the point prediction velocity at that prediction time point.

[0092] As an example, the point prediction velocity of any character skeleton point at any prediction time point can be obtained by the following formula:

[0093] In the above formula, This represents the point prediction velocity at time t+1. This represents the point prediction velocity at time t. Indicates the target acceleration. Indicates the time step.

[0094] In this scenario, the predicted velocity of any character's skeleton point at each of the multiple prediction time points can be obtained. Then, based on the predicted velocity of each point, the velocity sequence of the character's skeleton point within the time step can be obtained, which is the predicted velocity sequence.

[0095] Optionally, the target predicted velocity sequence of the target virtual character is obtained based on the point predicted velocity sequence of each character's skeletal points.

[0096] In this embodiment of the disclosure, the predicted velocity sequence of each character's skeletal points within the time step can be integrated based on the sequence integration method in the related technology, thereby obtaining the velocity sequence of the target virtual character within the time step, which is the target predicted velocity sequence.

[0097] Optionally, for any character skeleton point of the target virtual character, based on the initial velocity, target acceleration and time step of the character skeleton point in the current state, the predicted point position of the character skeleton point at each of the multiple predicted time points is obtained to obtain the predicted point position sequence of the character skeleton point. Based on the predicted point position sequence of each character skeleton point, the target predicted position sequence of the target virtual character within the time step is obtained.

[0098] In this embodiment of the disclosure, a point position prediction algorithm can be used to perform point position prediction on the character skeleton points based on the initial velocity, target acceleration and time step of the character skeleton points, thereby obtaining the predicted position of any character skeleton point at multiple prediction time points as the predicted point position, and then obtaining the position sequence of the character skeleton point within the time step based on the predicted point position at each prediction time point, which is the predicted point position sequence.

[0099] As an example, the predicted position of any character's skeleton point can be obtained using the following formula:

[0100] in, This indicates the predicted point position at time t+1. This indicates the predicted point position at time t. Represents the velocity at time t. Indicates the target acceleration. Indicates the time step.

[0101] In this scenario, the predicted position sequences of each character's skeleton points can be integrated using sequence integration algorithms in related technologies to obtain the position sequence of the target virtual character within the time step, which is the target predicted position sequence.

[0102] Optionally, the joint angle variables of each part of the target virtual character within the time step are obtained to obtain the predicted joint angle of each part at the end of the time step.

[0103] In this embodiment of the disclosure, the target virtual character includes multiple parts, which can be identified as the character parts of the target virtual character. In the scenario where the target virtual character is a digital human, the character parts can be understood as the digital human's hands, arms, neck, and other parts that include joints.

[0104] During the movement of the target virtual character, the joint angles of the character parts may change. In this scenario, the amount of change in the joint angles of each character part during the process of moving from the current state to the desired state can be predicted, thereby obtaining the amount of change in the joint angles of each character part during the process of moving from the current state to the desired state in the target virtual character.

[0105] This allows us to obtain the changes in joint angles of each character's parts within each time step, thereby obtaining the joint angles of each character's parts at the end of each time step, and finally obtaining the target predicted joint angle sequence of the target virtual character at the end of each time step.

[0106] The algorithm shown in the following formula can be used to obtain the joint angle variable of any character part when it reaches the end of the time step from the current state.

[0107]

[0108] In the above formula, Represents the joint angle variable. Represents the Jacobian matrix, This indicates the expected position of a character part at the end of the time step. This indicates the position of a character part at the beginning of the time step. This indicates the joint angle of a character's body part at the beginning of the time step.

[0109] In this scenario, for any character part, the joint angle variable corresponding to the character part can be processed by the algorithm shown in the following formula to obtain the joint angle of the character part at the end of the corresponding time step, which is the predicted joint angle of the character part at the end of the time step.

[0110] The formula is as follows:

[0111] In the above formula, This represents the joint angle at time t+1, which is the predicted joint angle at the end of the time step. This represents the joint angle at time t, which is the joint angle at the start of the time step. Represents joint variables within a time step. This represents the relaxation factor.

[0112] Optionally, based on the predicted joint angles of each character's parts, a sequence of predicted joint angles of the target virtual character at the end of the time step is obtained.

[0113] In this embodiment of the disclosure, for any time step, the joint angle sequence of each character part at the time step can be obtained by using the sequence construction method in the related art, based on the predicted joint angle of each character part at the end of the time step, which is the target predicted joint angle sequence.

[0114] Optionally, based on the target predicted velocity sequence, target predicted position sequence, and target predicted joint angle sequence, a predicted trajectory segment of the target virtual character within the time step is obtained to obtain the predicted trajectory of the target virtual character.

[0115] In this embodiment of the disclosure, the target predicted velocity sequence, target predicted position sequence, and target predicted joint angle sequence at any time step can be processed by the virtual character movement trajectory prediction algorithm in the related technology. Based on the result of the algorithm processing, the executable predicted trajectory of the target virtual character in that time step is obtained as the predicted trajectory segment of the target virtual character in that time step.

[0116] In this scenario, the predicted trajectory segments within each time step can be integrated to obtain the integrated trajectory, which is then identified as the predicted trajectory for the target virtual character to reach the desired state from the current state.

[0117] S204. When the target virtual character moves according to the predicted trajectory, the movement position and direction of each character bone point in the target virtual character are collected to obtain the character's movement trajectory.

[0118] In this embodiment of the disclosure, the movement of the target virtual character can be controlled by the virtual character control method in the related technology, so that it can move according to the predicted trajectory, and the trajectory information of the target virtual character's movement trajectory can be collected to obtain the trajectory of the target virtual character's movement, that is, the character's movement trajectory.

[0119] Specifically, the system can collect position and direction data of each character skeleton point in the target virtual character based on a set time interval. The collected data is the movement position and direction of each character skeleton point. Then, based on the collected movement position and direction, the corresponding movement trajectory is obtained, which serves as the character movement trajectory of the target virtual character.

[0120] As an example, such as Figure 3 As shown, by considering each moving position and direction, a result can be obtained. Figure 3 The shown movement trajectory serves as the movement trajectory of the target virtual character, where curves represent movement positions and arrows represent movement directions.

[0121] S205, obtain the reference virtual character and the reference movement trajectory of the reference virtual character in each executable animation segment in the executable animation segment library.

[0122] In this embodiment of the disclosure, the system is equipped with a preset executable animation clip library, wherein the original animation clip can be segmented into image frames to obtain each frame image included in the original animation clip, the original character bone points in the original virtual character of the original animation clip can be extracted, and the bone point features of the original character bone points can be obtained based on each frame image.

[0123] It should be noted that the original animation clips can be obtained by extracting animation clips from open-source animation clip data sources, or they can be drawn in other ways; no specific limitations are made here.

[0124] One approach is to segment the original animation clip into image frames using image frame segmentation methods from related technologies, obtaining segmented frame images. In some scenarios, the original animation clip includes corresponding animated characters, which can be identified as the original virtual characters in the original animation clip.

[0125] In this scenario, the skeletal points of the original virtual character can be extracted using a virtual character skeletal point extraction algorithm from relevant technologies. Then, based on the processing results of the extraction algorithm, the skeletal points of the original virtual character can be obtained as the original character skeletal points in the original virtual character.

[0126] In this embodiment of the disclosure, the original character skeleton points can be processed by feature extraction algorithms in related technologies to obtain feature data of the original character skeleton points in the position dimension, direction dimension and velocity dimension, and the extracted multi-dimensional feature data are integrated accordingly to obtain the skeleton point features of the original character skeleton points.

[0127] Optionally, the original movement trajectory of the original virtual character in the original animation clip can be obtained.

[0128] In this embodiment of the disclosure, the movement trajectory of the original virtual character in the original animation clip can be extracted using a trajectory extraction algorithm in the related technology, and the extracted trajectory can be determined as the original movement trajectory of the original virtual character.

[0129] In this process, the original movement trajectory can be extracted from multiple consecutive frame images. That is, the position and orientation of the original virtual character in multiple consecutive frame images can be obtained, and then the original movement trajectory of the original virtual character can be obtained based on the extracted information.

[0130] Optionally, based on the original movement trajectory and the skeletal point features of the original character skeleton points, executable animation segments corresponding to the original animation segments are obtained to obtain an executable animation segment library.

[0131] In this embodiment of the disclosure, animation clips can be generated and constructed based on the original movement trajectory, the original virtual character, the original character skeletal points, and the skeletal point features of the original character skeletal points using animation clip generation algorithms in related technologies. This results in an animation clip reconstructed from the original animation clip, which serves as the executable animation clip corresponding to the original animation clip.

[0132] It should be noted that there are multiple original animation clips. In this scenario, the corresponding executable animation clip library can be obtained based on the executable animation clips of each original animation clip.

[0133] In this embodiment of the disclosure, the virtual characters in each executable animation segment in the executable animation segment library can be determined as reference virtual characters in each executable animation segment.

[0134] Specifically, for any executable animation segment, a trajectory extraction algorithm in the relevant technology can be used to extract the trajectory of the executable animation segment. The extracted trajectory is the trajectory of the reference virtual character in the executable animation segment, that is, the reference movement trajectory of the reference virtual character.

[0135] S206, based on the target virtual character, the reference virtual character, the character's movement trajectory, and the reference movement trajectory, determine the next target animation segment to be executed from the executable animation segment library.

[0136] In this embodiment of the disclosure, for any executable animation segment, the target virtual character and the target virtual character's movement trajectory can be compared with the executable animation segment from the dimensions of the virtual character and the character's movement trajectory. Based on the comparison results, it can be identified whether the executable animation segment can be used as the next target animation segment to be executed by the target virtual character.

[0137] Optionally, the character matching degree between the reference virtual character and the target virtual character can be obtained.

[0138] In some implementations, virtual character matching algorithms in related technologies can be used to process the reference virtual character and the target virtual character in any executable animation clip, and the matching degree between the two can be obtained based on the result of the algorithm processing, which is used as the character matching degree between the two.

[0139] In other implementations, for any character skeleton point in the target virtual character, the corresponding associated skeleton point on the reference virtual character is obtained to obtain a set of skeleton point difference parameters between the character skeleton point and the associated skeleton point. The set of skeleton point difference parameters includes at least a first position feature difference parameter, a first direction feature difference parameter, and a first velocity feature difference parameter between the character skeleton point and the associated skeleton point. Based on the set of skeleton point difference parameters for each character skeleton point, the character matching degree between the target virtual character and the reference virtual character is obtained.

[0140] In this embodiment of the disclosure, for any reference virtual character, the skeletal points included in the reference virtual character and the target virtual character can be matched and compared from the dimension of skeletal points to obtain the character matching degree between the two.

[0141] Specifically, for any character skeleton point included in the target virtual character, the relative positional relationship between the character skeleton point and the target virtual character can be obtained, and the skeleton point that is in the same relative positional relationship on the reference virtual character can be obtained. This skeleton point is the associated skeleton point of the character skeleton point on the reference virtual character.

[0142] In this scenario, the character's skeletal points and their associated skeletal points can be compared in terms of position, orientation, and velocity. Feature extraction algorithms in related technologies can be used to extract features in the position, orientation, and velocity dimensions of both, thereby obtaining the position, orientation, and velocity features of the character's skeletal points and their associated skeletal points.

[0143] In this scenario, feature difference algorithms from related technologies can be used to process the positional features of the character's skeletal points and associated skeletal points. Based on the algorithmic processing results, a difference parameter between the two positional features is obtained, which serves as the first positional feature difference parameter between them. Similarly, the directional features of the character's skeletal points and associated skeletal points can be processed using algorithms, and the difference parameter between the two directional features obtained from the algorithmic processing results is determined as the first directional feature difference parameter between them. Finally, the velocity features of the character's skeletal points and associated skeletal points can be processed using algorithms, and the difference parameter between the two velocity features obtained from the algorithmic processing results is determined as the first velocity feature difference parameter between them.

[0144] In this scenario, the corresponding parameter set can be obtained based on the first position feature difference parameter, the first direction feature difference parameter, and the first velocity feature difference parameter, which can be used as the set of skeletal point difference parameters between the character's skeletal point and the associated skeletal point.

[0145] It should be noted that the comparison between character skeletal points and related skeletal points can be achieved through the three dimensions given above, or by comparing character skeletal points with their corresponding related skeletal points based on other dimensions; no specific limitations are made here.

[0146] In this embodiment of the disclosure, the matching degree algorithm in the related technology can be used to obtain the matching degree between the skeleton points of the character and the related skeleton points based on the set of skeleton point difference parameters. Then, based on the matching degree of each skeleton point, the matching degree between the target virtual character and the reference virtual character can be obtained, that is, the character matching degree between the two.

[0147] As an example, such as Figure 4 As shown, the target virtual character and the reference virtual character include: Figure 4 The points shown are the head bone points, left hand bone points, right hand bone points, left foot bone points, and right foot bone points, that is, as... Figure 4 As shown, the head bone point 1 of the target virtual character is associated with the head bone point 2 on the reference virtual character; the left hand bone point 1 of the target virtual character is associated with the left hand bone point 2 on the reference virtual character; the right hand bone point 1 of the target virtual character is associated with the right hand bone point 2 on the reference virtual character; the left foot bone point 1 of the target virtual character is associated with the left foot bone point 2 on the reference virtual character; and the right foot bone point 1 of the target virtual character is associated with the right foot bone point 2 on the reference virtual character.

[0148] Then, we can obtain the feature difference parameters of each of the two head bone points, two left hand bone points, two right hand bone points, two left foot bone points, and two right foot bone points in the position dimension, direction dimension, and velocity dimension, and thus obtain the character matching degree between the two virtual characters.

[0149] The acquisition of feature difference parameters based on position, direction, and velocity dimensions, as well as the character matching degree, can be understood using the following formula: First positional feature difference parameter = (Positional feature (head bone point 1) - Positional feature (head bone point 2)) + (Positional feature (left hand bone point 1) - Positional feature (left hand bone point 2)) + (Positional feature (right hand bone point 1) - Positional feature (right hand bone point 2)) + (Positional feature (left foot bone point 1) - Positional feature (left foot bone point 2)) + (Positional feature (right foot bone point 1) - Positional feature (right foot bone point 2)) First directional feature difference parameter = (directional feature (head bone point 1) - directional feature (head bone point 2)) + (directional feature (left hand bone point 1) - directional feature (left hand bone point 2)) + (directional feature (right hand bone point 1) - directional feature (right hand bone point 2)) + (directional feature (left foot bone point 1) - directional feature (left foot bone point 2)) + (directional feature (right foot bone point 1) - directional feature (right foot bone point 2)) First velocity feature difference parameter = (Velocity feature (head bone point 1) - Velocity feature (head bone point 2)) + (Velocity feature (left hand bone point 1) - Velocity feature (left hand bone point 2)) + (Velocity feature (right hand bone point 1) - Velocity feature (right hand bone point 2)) + (Velocity feature (left foot bone point 1) - Velocity feature (left foot bone point 2)) + (Velocity feature (right foot bone point 1) - Velocity feature (right foot bone point 2)) Character matching degree = First position feature difference parameter + First direction feature difference parameter + First velocity feature difference parameter Optionally, the trajectory matching degree between the reference movement trajectory and the character's movement trajectory can be obtained.

[0150] Specifically, for any character trajectory point in the character's movement trajectory, the associated trajectory point corresponding to the character trajectory point in the reference movement trajectory is obtained to obtain a set of trajectory point difference parameters between the character trajectory point and the associated trajectory point. The set of trajectory point difference parameters includes at least the second position feature difference parameter, the second direction feature difference parameter, and the second velocity feature difference parameter between the character trajectory point and the associated trajectory point. Based on the set of trajectory point difference parameters for each character trajectory point, the trajectory matching degree between the reference movement trajectory and the character's movement trajectory is obtained.

[0151] In this embodiment of the disclosure, the trajectory points in the character's movement trajectory can be determined as character trajectory points. For any character trajectory point, the time when the target virtual character arrives at the character trajectory point can be obtained, and the trajectory points of the reference virtual character that arrive at the same time in the reference movement trajectory can be obtained. These trajectory points can then be determined as the associated trajectory points of the character trajectory points in the reference movement trajectory.

[0152] In this scenario, feature extraction methods from relevant technologies can be used to extract features from the character's trajectory points and associated trajectory points in terms of position, direction, and velocity. Then, feature difference algorithms from relevant technologies can be used to calculate the differences in features of the two trajectory points in the position, direction, and velocity dimensions. Based on the results of the algorithm processing, feature difference parameters of the two trajectory points in the position dimension (i.e., the second position feature difference parameter), feature difference parameters of the two trajectory points in the velocity dimension (i.e., the second velocity feature difference parameter), and feature difference parameters of the two trajectory points in the direction dimension (i.e., the second direction feature difference parameter) are obtained, thus obtaining the set of difference parameters between the two trajectory points, which serves as the trajectory point difference parameter set.

[0153] In this scenario, the matching degree between any character's trajectory point and its corresponding associated trajectory point can be analyzed based on the set of trajectory point difference parameters to obtain the trajectory point matching degree. Then, based on the matching degree of each trajectory point, the trajectory matching degree between the character's movement trajectory and the reference movement trajectory can be obtained.

[0154] As an example, such as Figure 5 As shown, Figure 5 The trajectory point A1 on the character's movement trajectory is associated with the trajectory point A2 in the reference movement trajectory, the trajectory point B1 on the character's movement trajectory is associated with the trajectory point B2 in the reference movement trajectory, the trajectory point C1 on the character's movement trajectory is associated with the trajectory point C2 in the reference movement trajectory, and the trajectory point D1 on the character's movement trajectory is associated with the trajectory point D2 in the reference movement trajectory.

[0155] Then in Figure 5 In the scenario shown, features can be extracted from trajectory points A1, A2, B1, B2, C1, C2, D1, and D2 based on position, direction, and velocity dimensions, respectively.

[0156] Specifically, for trajectory points A1 and A2, the feature difference algorithm can be used to obtain the feature difference parameters of the position dimension, the feature difference parameters of the velocity dimension, and the feature difference parameters of the direction dimension between trajectory points A1 and A2, which are the second position feature difference parameters, the second direction feature difference parameters, and the second velocity feature difference parameters. Then, the trajectory point matching degree between trajectory points A1 and A2 can be obtained based on the trajectory point difference parameter set obtained by the second position feature difference parameters, the second direction feature difference parameters, and the second velocity feature difference parameters.

[0157] Furthermore, the trajectory point matching degree between trajectory point B1 and trajectory point B2, the trajectory point matching degree between trajectory point C1 and trajectory point C2, and the trajectory point matching degree between trajectory point D1 and trajectory point D2 are obtained using the same method.

[0158] In this example, based on the matching degree of each trajectory point obtained above, we can obtain... Figure 5 The degree of matching between the shown character movement trajectory and the reference movement trajectory, that is, the trajectory matching degree between the two.

[0159] As another example, the acquisition of the second position feature difference parameter, the second direction feature difference parameter, the second velocity feature difference parameter, and the trajectory matching degree can be understood by combining the following calculation process: Second position feature difference parameter = (position feature (trajectory point A1) - position feature (trajectory point A2)) + (position feature (trajectory point B1) - position feature (trajectory point B2)) + (position feature (trajectory point C1) - position feature (trajectory point C2)) + (position feature (trajectory point D1) - position feature (trajectory point D2)) Second directional feature difference parameter = (directional feature (trajectory point A1) - directional feature (trajectory point A2)) + (directional feature (trajectory point B1) - directional feature (trajectory point B2)) + (directional feature (trajectory point C1) - directional feature (trajectory point C2)) + (directional feature (trajectory point D1) - directional feature (trajectory point D2)) Second velocity characteristic difference parameter = (velocity characteristic (trajectory point A1) - velocity characteristic (trajectory point A2)) + (velocity characteristic (trajectory point B1) - velocity characteristic (trajectory point B2)) + (velocity characteristic (trajectory point C1) - velocity characteristic (trajectory point C2)) + (velocity characteristic (trajectory point D1) - velocity characteristic (trajectory point D2)) Trajectory matching degree = Second position feature difference parameter + Second direction feature difference parameter + Second velocity feature difference parameter Optionally, the next target animation segment to be executed can be determined from the executable animation segment library based on the character matching degree and trajectory matching degree.

[0160] Specifically, for any executable animation segment, the character matching degree and trajectory matching degree corresponding to the executable animation segment are integrated to obtain the target matching degree between the executable animation segment and the target virtual character. Based on the target matching degree of each executable animation segment, the target animation segment is determined from the executable animation segment library.

[0161] In this embodiment of the disclosure, a matching degree integration algorithm in the related art can be used to integrate the character matching degree and trajectory matching degree between any executable animation segment and the target virtual character. Based on the result of the algorithm processing, the overall matching degree between the two is obtained as the target matching degree.

[0162] The integration of role matching degree and trajectory matching degree can be achieved by summing them, or by using other algorithms to process role matching degree and trajectory matching degree. No specific limitation is made here.

[0163] In this embodiment of the disclosure, after obtaining the target matching degree between each executable animation segment and the target virtual character and the character's movement trajectory, the matching degree that meets the preset filtering conditions can be obtained from each target matching degree, and the executable animation segment corresponding to the matching degree is determined as the next target animation segment to be executed after the target virtual character reaches the desired state.

[0164] The filtering criteria can be constructed based on the minimum matching degree among the target matching degrees, or based on a preset threshold; no specific limitation is made here.

[0165] As an example, such as Figure 6 As shown, it can be done through Figure 6 Each original animation segment in the shown original animation set undergoes preprocessing. This preprocessing may include segmenting each original animation segment into image frames, and extracting the virtual character and its skeletal points from each segmented frame image. Specifically, for any frame image in any original animation segment, the movement trajectory of the virtual character and the skeletal point features of each skeletal point in that frame image are obtained based on multiple consecutive frames corresponding to that frame image. Then, feature data corresponding to that frame image is constructed based on this extracted information. Finally, an executable animation segment corresponding to that original animation segment is obtained based on the feature data of each frame image, thus obtaining… Figure 6 The executable animation clip library is shown.

[0166] like Figure 6 As shown, when the target virtual character receives an input control signal, the trajectory of the target virtual character from its current state to the desired state can be predicted, the corresponding predicted trajectory can be obtained, and the target virtual character can be controlled to move along the predicted trajectory, thereby obtaining the target virtual character's movement trajectory, as shown. Figure 6 As shown, after obtaining the character's movement trajectory, it can be used... Figure 6 The feature matching module shown performs feature matching between each executable animation clip in the executable animation clip library and the target virtual character and the character's movement trajectory. The matching process can be understood in conjunction with the relevant content in the above embodiments.

[0167] like Figure 6 As shown, based on feature matching processing, the next target animation segment to be executed for the target virtual character can be obtained from the executable animation segment library, and the target virtual character can be animated accordingly based on the target animation segment.

[0168] The animation acquisition method proposed in this disclosure predicts the trajectory of a target virtual character from its current state to the desired state based on the desired state corresponding to the input control signal, obtaining a predicted trajectory. Then, based on the character's movement trajectory collected as the target virtual character moves along the predicted trajectory, the next executable target animation segment for the target virtual character is obtained from a pre-set set of executable animation segments. This eliminates the need for manual setting of each animation segment played by the target virtual character, reducing the reliance on manual intervention in the application of the target virtual character and improving the efficiency of acquiring the animation segments played by the target virtual character. Furthermore, by determining the target animation segment through character matching degree and trajectory matching degree, the method improves the accuracy of matching the target animation segment with the target virtual character. The improved adaptability between virtual characters enhances the smoothness of animation playback. Trajectory prediction is performed based on the target acceleration obtained from the initial acceleration corresponding to the environmental acceleration and input control signals, improving the scene adaptability and prediction accuracy of trajectory prediction. Upon receiving an input control signal, the movement of the target virtual character is controlled by acquiring the predicted trajectory. This eliminates the need for separate development of the actions required by the target virtual character, reducing application costs and increasing its flexibility and controllability in application scenarios. Consequently, the smoothness and scene adaptability of the target virtual character's animation playback are improved, optimizing the user's viewing experience.

[0169] The animation acquisition method proposed in this disclosure predicts the trajectory of a target virtual character from its current state to the desired state based on the desired state corresponding to the input control signal, obtaining a predicted trajectory. Then, based on the character's movement trajectory collected as the target virtual character moves along the predicted trajectory, the next executable target animation segment for the target virtual character is obtained from a pre-set set of executable animation segments. This eliminates the need for manual setting of each animation segment played by the target virtual character, reducing the reliance on manual intervention in the application of the target virtual character and improving the efficiency of acquiring the animation segments played by the target virtual character. Furthermore, by determining the target animation segment through character matching degree and trajectory matching degree, the method improves the accuracy of matching the target animation segment with the target virtual character. The improved adaptability between virtual characters enhances the smoothness of animation playback. Trajectory prediction is performed based on the target acceleration obtained from the initial acceleration corresponding to the environmental acceleration and input control signals, improving the scene adaptability and prediction accuracy of trajectory prediction. Upon receiving an input control signal, the movement of the target virtual character is controlled by acquiring the predicted trajectory. This eliminates the need for separate development of the actions required by the target virtual character, reducing application costs and increasing its flexibility and controllability in application scenarios. Consequently, the smoothness and scene adaptability of the target virtual character's animation playback are improved, optimizing the user's viewing experience.

[0170] Corresponding to the animation acquisition methods provided in the above embodiments, an embodiment of this disclosure also provides an animation acquisition device. Since the animation acquisition device provided in this disclosure corresponds to the animation acquisition methods provided in the above embodiments, the implementation methods of the above animation acquisition methods are also applicable to the animation acquisition device provided in this disclosure, and will not be described in detail in the following embodiments.

[0171] Figure 7 This is a schematic diagram of the structure of an animation acquisition device according to an embodiment of the present disclosure, as shown below. Figure 7 As shown, the animation acquisition device 700 includes a prediction module 71, a control acquisition module 72, and a determination module 73, wherein: The prediction module 71 is used to respond to the input control signal received from the target virtual character, obtain the expected state corresponding to the input control signal, and obtain the predicted trajectory of the target virtual character from the current state to the expected state. The control acquisition module 72 is used to control the target virtual character to move based on the predicted trajectory and to acquire the character movement trajectory of the target virtual character that has moved. The determination module 73 is used to determine the next target animation segment to be executed for the target virtual character from a preset library of executable animation segments based on the character's movement trajectory.

[0172] In this embodiment of the disclosure, the prediction module 71 is further configured to: in response to receiving an input control signal, obtain the initial acceleration of the target virtual character based on the input control signal; obtain the environmental acceleration of the target virtual character based on its virtual environment, and obtain the target acceleration of the target virtual character to reach the desired state based on the environmental acceleration and the initial acceleration; and obtain the predicted trajectory of the target virtual character based on the target acceleration.

[0173] In this embodiment of the disclosure, the prediction module 71 is further configured to: obtain the virtual gravity field acceleration of the target virtual character in the virtual environment; in response to the existence of a predicted virtual collision event of the target virtual character, obtain the virtual collision acceleration of the target virtual character in the virtual collision event; obtain the environmental acceleration based on the virtual gravity field acceleration and the virtual collision acceleration, and adjust the initial acceleration based on the environmental acceleration to obtain the target acceleration.

[0174] In this embodiment of the disclosure, the prediction module 71 is further configured to: in response to the absence of a predicted virtual collision event for the target virtual character, obtain environmental acceleration based on the virtual gravity field acceleration.

[0175] In this embodiment of the disclosure, the prediction module 71 is further configured to: obtain the time step of the target virtual character from the current state to the desired state; based on the time step and the target acceleration, obtain the target predicted velocity sequence, the target predicted position sequence and the target predicted joint angle sequence of the target virtual character within the time step; based on the target predicted velocity sequence, the target predicted position sequence and the target predicted joint angle sequence, obtain the predicted trajectory segment of the target virtual character within the time step, so as to obtain the predicted trajectory of the target virtual character.

[0176] In this embodiment of the disclosure, the prediction module 71 is further configured to: obtain multiple prediction time points included in the time step; for any character skeleton point of the target virtual character, obtain the initial velocity of the character skeleton point in the current state, and based on the initial velocity, target acceleration and time step, obtain the predicted velocity of the character skeleton point at each prediction time point in the multiple prediction time points, so as to obtain the point prediction velocity sequence of the character skeleton point; and based on the point prediction velocity sequence of each character skeleton point, obtain the target prediction velocity sequence of the target virtual character within the time step.

[0177] In this embodiment of the disclosure, the prediction module 71 is further configured to: for any character skeleton point of the target virtual character, based on the initial velocity, target acceleration and time step of the character skeleton point in the current state, obtain the predicted point position of the character skeleton point at each of the multiple prediction time points, so as to obtain the predicted point position sequence of the character skeleton point; and based on the predicted point position sequence of each character skeleton point, obtain the target predicted position sequence of the target virtual character within the time step.

[0178] In this embodiment of the disclosure, the prediction module 71 is further configured to: obtain the joint angle variables of each part of the target virtual character within the time step, so as to obtain the predicted joint angle of each part of the character at the end of the time step; and obtain the target predicted joint angle sequence of the target virtual character at the end of the time step based on the predicted joint angle of each part of the character.

[0179] In this embodiment of the disclosure, the control acquisition module 72 is further configured to: acquire the movement position and movement direction of each character skeletal point in the target virtual character when the target virtual character moves according to the predicted trajectory, so as to obtain the character movement trajectory.

[0180] In this embodiment of the present disclosure, the determining module 73 is further configured to: obtain a reference virtual character and a reference movement trajectory of the reference virtual character in each executable animation segment in the executable animation segment library; and determine the next target animation segment to be executed from the executable animation segment library based on the target virtual character, the reference virtual character, the character movement trajectory and the reference movement trajectory.

[0181] In this embodiment of the disclosure, the determining module 73 is further configured to: obtain the character matching degree between the reference virtual character and the target virtual character; obtain the trajectory matching degree between the reference movement trajectory and the character movement trajectory; and determine the next target animation segment to be executed from the executable animation segment library based on the character matching degree and the trajectory matching degree.

[0182] In this embodiment of the disclosure, the determining module 73 is further configured to: for any character bone point in the target virtual character, obtain the associated bone point on the reference virtual character corresponding to the character bone point, so as to obtain a set of bone point difference parameters between the character bone point and the associated bone point, wherein the set of bone point difference parameters includes at least a first position feature difference parameter, a first direction feature difference parameter, and a first velocity feature difference parameter between the character bone point and the associated bone point; and obtain the character matching degree between the target virtual character and the reference virtual character based on the set of bone point difference parameters of each character bone point.

[0183] In this embodiment of the disclosure, the determining module 73 is further configured to: for any character trajectory point in the character movement trajectory, obtain the associated trajectory point corresponding to the character trajectory point in the reference movement trajectory, so as to obtain a set of trajectory point difference parameters between the character trajectory point and the associated trajectory point, wherein the set of trajectory point difference parameters includes at least a second position feature difference parameter, a second direction feature difference parameter and a second velocity feature difference parameter between the character trajectory point and the associated trajectory point; and obtain the trajectory matching degree between the reference movement trajectory and the character movement trajectory based on the set of trajectory point difference parameters for each character trajectory point.

[0184] In this embodiment of the disclosure, the determining module 73 is further configured to: for any executable animation segment, integrate the character matching degree and trajectory matching degree corresponding to the executable animation segment to obtain the target matching degree between the executable animation segment and the target virtual character; and determine the target animation segment from the executable animation segment library based on the target matching degree of each executable animation segment.

[0185] In this embodiment of the present disclosure, the apparatus further includes an acquisition module, configured to: perform image frame segmentation on the original animation segment to obtain each frame image included in the original animation segment; extract the original character skeleton points in the original virtual character of the original animation segment, and obtain the skeleton point features of the original character skeleton points based on each frame image; acquire the original movement trajectory of the original virtual character in the original animation segment; and obtain the executable animation segment corresponding to the original animation segment based on the original movement trajectory and the skeleton point features of the original character skeleton points, so as to obtain an executable animation segment library.

[0186] The animation acquisition device disclosed herein predicts the trajectory of a target virtual character from its current state to the desired state based on the desired state corresponding to the input control signal, obtaining a predicted trajectory. Based on the character's movement trajectory collected as the target virtual character moves along the predicted trajectory, the device obtains the next executable animation segment for the target virtual character from a pre-set set of executable animation segments. This eliminates the need for manual setting of each animation segment played by the target virtual character, reducing the manual dependence on the application of the target virtual character, improving the acquisition efficiency of the animation segments played by the target virtual character, and enhancing the compatibility between the target animation segments and the target virtual character. Upon receiving an input control signal, the device controls the movement of the target virtual character by acquiring the predicted trajectory, eliminating the need for separate development of the actions required by the target virtual character. This reduces the application cost of the target virtual character, improves its flexibility and controllability in application scenarios, thereby enhancing the smoothness and scene adaptability of the target virtual character's animation playback and optimizing the user's viewing experience.

[0187] To achieve the above embodiments, this disclosure also provides an electronic device, a computer-readable storage medium, and a computer program product.

[0188] Figure 8 This is a block diagram of an electronic device according to an embodiment of the present disclosure, such as... Figure 8 As shown, device 800 includes processor 801 and memory 802 communicatively connected to the processor. Memory 802 stores instructions executable by at least one processor. The instructions are executed by at least one processor 801. Processor 801 runs a program corresponding to the executable program code by reading executable program code stored in memory 802, thereby implementing the functions described in this disclosure. Figures 1 to 6 An animation acquisition method of an embodiment.

[0189] To implement the above embodiments, this disclosure also proposes a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, are used to implement the present disclosure. Figures 1 to 6 An animation acquisition method of an embodiment.

[0190] To implement the above embodiments, this disclosure also proposes a computer program product, including a computer program, which, when executed by a processor, implements the features of this disclosure. Figures 1 to 6 An animation acquisition method of an embodiment.

[0191] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this disclosure. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0192] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this disclosure, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0193] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of code comprising one or more executable instructions for implementing custom logic functions or processes, and the scope of preferred embodiments of this disclosure includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order depending on the functions involved, as will be understood by those skilled in the art to which embodiments of this disclosure pertain.

[0194] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-included system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of computer-readable media include: an electrical connection having one or more wires (electronic device), a portable computer disk drive (magnetic device), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Alternatively, the computer-readable medium may be paper or other suitable media on which the program can be printed, since the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in a computer memory.

[0195] It should be understood that various parts of this disclosure can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0196] Those skilled in the art will understand that all or part of the steps of the methods in the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, the program includes one or a combination of the steps of the method embodiments.

[0197] Furthermore, the functional units in the various embodiments of this disclosure can be integrated into a processing module, or each unit can exist physically separately, or two or more units can be integrated into a module. The integrated module can be implemented in hardware or as a software functional module. If the integrated module is implemented as a software functional module and sold or used as an independent product, it can also be stored in a computer-readable storage medium.

[0198] The storage medium mentioned above can be a read-only memory, a disk, or an optical disk, etc. Although embodiments of the present disclosure have been shown and described above, it is to be understood that the above embodiments are exemplary and should not be construed as limiting the present disclosure. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present disclosure.

Claims

1. A method for acquiring animation, characterized in that, The method includes: In response to receiving an input control signal from a target virtual character, the desired state corresponding to the input control signal is obtained to obtain a predicted trajectory of the target virtual character from the current state to the desired state; Control the target virtual character to move based on the predicted trajectory, and collect the character movement trajectory of the target virtual character that has moved; Based on the character's movement trajectory, the next target animation segment to be executed for the target virtual character is determined from a preset library of executable animation segments.

2. The method according to claim 1, characterized in that, The step of responding to receiving an input control signal from a target virtual character and obtaining the desired state corresponding to the input control signal to obtain a predicted trajectory of the target virtual character from the current state to the desired state includes: In response to receiving the input control signal, the initial acceleration of the target virtual character based on the input control signal is obtained; The environmental acceleration of the target virtual character based on its virtual environment is obtained, and the target acceleration of the target virtual character to reach the desired state is obtained based on the environmental acceleration and the initial acceleration. Based on the target acceleration, the predicted trajectory of the target virtual character is obtained.

3. The method according to claim 2, characterized in that, The step of obtaining the environmental acceleration of the target virtual character based on its virtual environment, and obtaining the target acceleration of the target virtual character to reach the desired state based on the environmental acceleration and the initial acceleration, includes: Obtain the virtual gravitational field acceleration of the target virtual character in the virtual environment; In response to a predicted virtual collision event occurring in the target virtual character, the virtual collision acceleration of the target virtual character in the virtual collision event is obtained; The environmental acceleration is obtained based on the virtual gravity field acceleration and the virtual collision acceleration, and the initial acceleration is adjusted based on the environmental acceleration to obtain the target acceleration.

4. The method according to claim 3, characterized in that, The method further includes: In response to the absence of a predicted virtual collision event for the target virtual character, the environmental acceleration is obtained based on the virtual gravity field acceleration.

5. The method according to claim 2, characterized in that, The step of obtaining the predicted trajectory of the target virtual character based on the target acceleration includes: Obtain the time step of the target virtual character from its current state to the desired state; Based on the time step and the target acceleration, the target virtual character's predicted velocity sequence, predicted position sequence, and predicted joint angle sequence within the time step are obtained; Based on the target predicted velocity sequence, the target predicted position sequence, and the target predicted joint angle sequence, a predicted trajectory segment of the target virtual character within the time step is obtained to obtain the predicted trajectory of the target virtual character.

6. The method according to claim 5, characterized in that, Based on the time step and the target acceleration, a target predicted velocity sequence of the target virtual character within the time step is obtained, including: Obtain multiple predicted time points included in the time step; For any character skeleton point of the target virtual character, the initial velocity of the character skeleton point in the current state is obtained, and based on the initial velocity, the target acceleration and the time step, the predicted velocity of the character skeleton point at each of the multiple predicted time points is obtained, so as to obtain the point predicted velocity sequence of the character skeleton point. Based on the point prediction velocity sequence of each character's skeleton points, the target predicted velocity sequence of the target virtual character within the time step is obtained.

7. The method according to claim 5, characterized in that, Based on the time step and the target acceleration, a target predicted position sequence of the virtual character within the time step is obtained, including: For any character skeleton point of the target virtual character, based on the initial velocity, the target acceleration and the time step of the character skeleton point in the current state, the predicted point position of the character skeleton point at each of the multiple predicted time points is obtained, so as to obtain the predicted point position sequence of the character skeleton point. Based on the predicted point position sequence of each character's skeleton points, the target predicted position sequence of the target virtual character within the time step is obtained.

8. The method according to claim 5, characterized in that, Based on the time step and the target acceleration, a target predicted joint angle sequence of the target virtual character within the time step is obtained, including: The joint angle variables of each part of the target virtual character within the time step are obtained to obtain the predicted joint angle of each part at the end of the time step. Based on the predicted joint angles of each character part, the target predicted joint angle sequence of the target virtual character at the end of the time step is obtained.

9. The method according to claim 1, characterized in that, The process of controlling the target virtual character to move based on the predicted trajectory and collecting the character movement trajectory of the target virtual character includes: When the target virtual character moves according to the predicted trajectory, the movement position and direction of each character skeleton point in the target virtual character are collected to obtain the character's movement trajectory.

10. The method according to claim 9, characterized in that, The step of determining the next target animation segment to be executed for the target virtual character from a preset library of executable animation segments based on the character's movement trajectory includes: Obtain the reference virtual character and the reference movement trajectory of the reference virtual character in each executable animation segment in the executable animation segment library; Based on the target virtual character, the reference virtual character, the character's movement trajectory, and the reference movement trajectory, the next target animation segment to be executed is determined from the executable animation segment library.

11. The method according to claim 10, characterized in that, The step of determining the next target animation segment to be executed from the executable animation segment library based on the target virtual character, the reference virtual character, the character's movement trajectory, and the reference movement trajectory includes: Obtain the character matching degree between the reference virtual character and the target virtual character; Obtain the trajectory matching degree between the reference movement trajectory and the character's movement trajectory; Based on the character matching degree and the trajectory matching degree, the next target animation segment to be executed is determined from the executable animation segment library.

12. The method according to claim 11, characterized in that, The step of obtaining the character matching degree between the reference virtual character and the target virtual character includes: For any character skeleton point in the target virtual character, obtain the associated skeleton point corresponding to the character skeleton point on the reference virtual character to obtain a set of skeleton point difference parameters between the character skeleton point and the associated skeleton point. The set of skeleton point difference parameters includes at least a first position feature difference parameter, a first direction feature difference parameter, and a first velocity feature difference parameter between the character skeleton point and the associated skeleton point. Based on the set of skeletal point difference parameters for each character's skeletal points, the character matching degree between the target virtual character and the reference virtual character is obtained.

13. The method according to claim 11, characterized in that, The step of obtaining the trajectory matching degree between the reference movement trajectory and the character movement trajectory includes: For any character trajectory point in the character movement trajectory, obtain the associated trajectory point corresponding to the character trajectory point in the reference movement trajectory to obtain a set of trajectory point difference parameters between the character trajectory point and the associated trajectory point. The set of trajectory point difference parameters includes at least a second position feature difference parameter, a second direction feature difference parameter and a second velocity feature difference parameter between the character trajectory point and the associated trajectory point. Based on the set of trajectory point difference parameters for each character's trajectory points, the trajectory matching degree between the reference movement trajectory and the character's movement trajectory is obtained.

14. The method according to claim 11, characterized in that, The step of determining the next target animation segment to be executed from the executable animation segment library based on the character matching degree and the trajectory matching degree includes: For any executable animation segment, the character matching degree and trajectory matching degree corresponding to the executable animation segment are integrated to obtain the target matching degree between the executable animation segment and the target virtual character; The target animation segment is determined from the executable animation segment library based on the target matching degree of each executable animation segment.

15. The method according to any one of claims 1-14, characterized in that, The method further includes: The original animation clip is segmented into individual frames to obtain the original animation clip. Extract the original character bone points from the original virtual character in the original animation clip, and obtain the bone point features of the original character bone points based on each frame image; Obtain the original movement trajectory of the original virtual character in the original animation clip; Based on the original movement trajectory and the skeletal point features of the original character skeleton points, the executable animation segments corresponding to the original animation segments are obtained, so as to obtain the executable animation segment library.

16. An animation acquisition device, characterized in that, The device includes: The prediction module is used to respond to the input control signal received from the target virtual character, obtain the expected state corresponding to the input control signal, and obtain the predicted trajectory of the target virtual character from the current state to the expected state. The control and acquisition module is used to control the target virtual character to move based on the predicted trajectory, and to acquire the character movement trajectory of the target virtual character that has moved; The determination module is used to determine the next target animation segment to be executed for the target virtual character from a preset library of executable animation segments based on the character's movement trajectory.

17. An electronic device, characterized in that, Including memory and processor; The processor reads executable program code stored in the memory to run a program corresponding to the executable program code, so as to implement the method as described in any one of claims 1-15.

Images