Behavior construction method and system for realizing cartoon character

Abstract

The invention relates to the technical field of cartoon processing, and discloses a behavior construction method and system for realizing a cartoon character, and the method comprises the steps: analyzing the emotion expression tendency of the cartoon character; the method comprises the following steps: analyzing character behavior motivations of a cartoon character in different scenes, constructing behavior frames of the cartoon character in different scenes, performing detail optimization processing on each limb movement in a limb movement frame to obtain an optimized limb movement frame, calculating the movement complexity corresponding to each limb movement in the optimized limb movement frame, and calculating the movement complexity corresponding to each limb movement in the optimized limb movement frame. Expression changes in the expression action framework are refined, a refined expression action framework is obtained, behavior construction schemes of the cartoon character are generated, visual information abundance corresponding to each scheme in the behavior construction schemes is calculated, scheme execution resources corresponding to the behavior construction schemes are allocated, and behavior construction processing of the cartoon character is executed. And obtaining a figure behavior construction result. According to the invention, the behavior construction effect of the cartoon character can be improved.

Description

Technical Field

[0001] The invention relates to a method and a system for realizing the behavior construction of cartoon characters, and belongs to the technical field of cartoon processing. Background Art

[0002] Cartoon characters occupy an extremely important position in many fields such as film, television, games, and animation. With the vigorous development of the animation industry, the audience's requirements for the performance of cartoon characters are increasing. The construction of cartoon characters' behavior, as the core link in shaping a vivid character image, its quality and efficiency directly affect the quality and production cycle of the entire animation work.

[0003] The traditional construction of animation character behavior mainly relies on animators to manually draw key frames, and determine the order and rhythm of the character's actions based on experience and simple timeline settings. After determining the sequence of behavioral actions that the animation character needs to show, the animator first draws the initial key frame to depict the starting posture and important turning points of the action. Then, relying on the understanding of the laws of motion and artistic creation intuition, the animator gradually adds intermediate frames to construct a relatively smooth action process. Then, the playback order and duration of each action clip are planned according to the timeline, and then integrated with scene background, sound effects and other elements. However, this method is extremely dependent on the personal skills and experience of the animator, and the work intensity is high and the efficiency is low. The style differences between different animators may also lead to problems in the coherence and consistency of the animation character behavior, which in turn leads to poor animation character behavior construction effects. Summary of the invention

[0004] The present invention provides a method and system for realizing the behavior construction of cartoon characters, the main purpose of which is to improve the behavior construction effect of cartoon characters.

[0005] To achieve the above object, the present invention provides a method for realizing the behavior construction of an animated character, comprising: Acquire basic setting information and plot scene information of the cartoon character, wherein the basic setting information includes the character's personality and the character's background story, and analyze the emotional expression tendency of the cartoon character based on the character's personality; Based on the plot scene information, analyzing the character behavior motivations of the cartoon character in different scenes, and based on the character behavior motivations, constructing the behavior framework of the cartoon character in different scenes, the behavior framework including a body movement framework, an expression movement framework and a language expression framework; Based on the character background story, each limb movement in the limb movement framework is optimized in detail to obtain an optimized limb movement framework, the movement complexity corresponding to each limb movement in the optimized limb movement framework is calculated, and based on the emotional expression tendency, the expression changes in the expression movement framework are refined to obtain a refined expression movement framework, and behavioral performance cases of classic cartoon characters of the same type are collected. Based on the behavioral performance cases, the language expression paradigm in the language expression framework is corrected to obtain a corrected language expression framework; In combination with the optimized limb movement framework, the refined facial expression movement framework and the corrected language expression framework, a behavior construction scheme for the cartoon character is generated, the visual information abundance corresponding to each scheme in the behavior construction scheme is calculated, and in combination with the visual information abundance and the movement complexity, scheme execution resources corresponding to the behavior construction scheme are allocated, and in combination with the scheme execution resources and the behavior construction scheme, the behavior construction processing of the cartoon character is executed to obtain a character behavior construction result.

[0006] Optionally, analyzing the emotional expression tendency of the cartoon character based on the character personality includes: Extracting characteristics of the character to obtain characteristics of the character; Performing emotion mapping on the personality traits to obtain an emotion mapping result; Based on the emotion mapping result, identifying the dominant emotion characteristics in the character's personality; Performing intensity evaluation on the dominant emotional feature to obtain an emotional intensity value; Based on the emotion intensity value, the emotion expression tendency of the cartoon character is determined.

[0007] Optionally, analyzing the character behavior motivations of the cartoon character in different scenes based on the plot scene information includes: Performing plot deconstruction on the plot scene information to obtain scene plot elements; Performing sentiment analysis on the scene plot elements to obtain the element sentiment tendency; Combined with the emotional tendency of the elements, the motivation trigger points in the plot elements of the scenes are explored; Based on the motivation trigger points, the character behavior motivations of the cartoon character in different scenarios are summarized.

[0008] Optionally, constructing a behavior framework of the cartoon character in different scenarios based on the character's behavior motivation includes: Classifying the character's behavior motivations to obtain motivation behavior categories; Obtaining a character behavior template of the cartoon character, and calculating a behavior similarity coefficient between the motivation behavior category and each template in the character behavior template; When the behavior similarity coefficient is greater than a preset similarity coefficient, matching a behavior category template corresponding to the motivation behavior category from the character behavior template; Based on the behavior category template, a behavior framework of the cartoon character in different scenarios is constructed.

[0009] Optionally, the calculating of the behavior similarity coefficient between the motivation behavior category and each template in the character behavior template includes: Performing vectorization processing on the motivation behavior category and each template in the character behavior template respectively to obtain a behavior category vector and a behavior template vector; Respectively extracting subdivision features corresponding to the behavior category vector and the behavior template vector to obtain a category subdivision feature vector and a template subdivision feature vector; Combining the category segmentation feature vector and the template segmentation feature vector, the behavior similarity coefficient between the motivation behavior category and each template in the character behavior template is calculated by the following formula: ; Among them, A represents the behavioral similarity coefficient between the motivation behavior category and each template in the character behavior template, represents the a-th category segmentation feature vector under the i-th behavior category in the motivation behavior category, It represents the ath template segmentation feature vector corresponding to the i-th behavior category in the character behavior template, a represents the serial number corresponding to the category segmentation feature vector and the template segmentation feature vector, i represents the serial number corresponding to the motivation behavior category and the character behavior template, and q represents the number of segmentation feature vectors corresponding to the i-th motivation behavior category and the i-th character behavior template.

[0010] Optionally, performing detail optimization processing on each body movement in the body movement framework based on the character background story includes: Extracting the plot of the character's background story to obtain the background story plot; Analyzing the correlation between the background storyline and each body movement in the body movement framework; Extracting background adaptation details corresponding to each body movement in the body movement framework in the background storyline according to the correlation degree; Based on the background adaptation details, each limb movement in the limb movement framework is subjected to detail optimization processing to obtain an optimized limb movement framework.

[0011] Optionally, the calculating the action complexity corresponding to each limb action in the optimized limb action framework includes: Performing action decomposition processing on each limb action in the optimized limb action framework to obtain action sub-elements, and determining the action duration period corresponding to the action sub-elements; Calculating the kinematic parameters of each element in the action sub-element, and extracting complexity-related parameters from the kinematic parameters, wherein the complexity-related parameters include: motion amplitude value, motion speed, and motion direction change frequency; Combined with the action duration period, the movement amplitude value, the movement speed parameter value and the movement direction change frequency, the action complexity corresponding to each limb action in the optimized limb action framework is calculated by the following formula: ; Among them, E represents the action complexity corresponding to each limb action in the optimized limb action framework. , , They represent the parameter weights corresponding to the motion amplitude, motion speed and frequency of motion direction changes, respectively. Indicates the motion amplitude parameter corresponding to the bth element in the action sub-element. Indicates the motion speed parameter corresponding to the bth element in the action sub-element. Indicates the frequency of change in the motion direction corresponding to the bth element in the action sub-element. It represents the action duration period corresponding to the bth element in the action sub-element, and T represents the total duration period corresponding to each limb action in the optimized limb action framework.

[0012] Optionally, based on the behavior performance case, correcting the language expression paradigm in the language expression framework to obtain a corrected language expression framework includes: Extracting discourse elements from the behavior performance case to obtain case discourse elements; Analyze the fit between the case discourse elements and each language expression paradigm in the language expression framework; Extracting, based on the degree of fit, paradigm correction details corresponding to each language expression paradigm in the language expression framework in the case discourse elements; Based on the paradigm correction details, each language expression paradigm in the language expression framework is corrected to obtain a corrected language expression framework.

[0013] Optionally, the calculating the visual information abundance corresponding to each scheme in the behavior construction scheme includes: Performing a screen rendering simulation on the behavior construction scheme to obtain a scheme rendering screen; Performing grayscale processing on the rendering picture of the scheme to obtain a grayscale scheme rendering picture; Extracting picture texture features from the grayscale rendering picture, and calculating picture texture complexity corresponding to the grayscale rendering picture based on the picture texture features; Counting the grayscale values ​​of pixels in the grayscale scheme picture, and calculating the picture information entropy of the grayscale scheme picture based on the pixel grayscale values; Combining the picture information entropy and the picture texture complexity, the visual information abundance corresponding to each scheme in the behavior construction scheme is calculated.

[0014] In order to solve the above problems, the present invention also provides a system for realizing the behavior construction of an animated character, the system comprising: An emotion expression tendency analysis module is used to obtain basic setting information and plot scene information of an animated character, wherein the basic setting information includes character personality and character background story, and analyze the emotion expression tendency of the animated character based on the character personality; A behavior framework construction module, for analyzing the character behavior motivations of the cartoon character in different scenes based on the plot scene information, and constructing the behavior framework of the cartoon character in different scenes based on the character behavior motivations, wherein the behavior framework includes a body movement framework, an expression movement framework, and a language expression framework; A framework correction module is used to optimize the details of each limb movement in the limb movement framework based on the character background story to obtain an optimized limb movement framework, calculate the movement complexity corresponding to each limb movement in the optimized limb movement framework, refine the expression changes in the expression movement framework based on the emotional expression tendency to obtain a refined expression movement framework, collect behavioral performance cases of classic cartoon characters of the same type, and based on the behavioral performance cases, correct the language expression paradigm in the language expression framework to obtain a corrected language expression framework; The behavior construction module is used to combine the optimized limb movement framework, the refined facial expression movement framework and the corrected language expression framework to generate a behavior construction plan for the cartoon character, calculate the visual information abundance corresponding to each plan in the behavior construction plan, allocate the plan execution resources corresponding to the behavior construction plan in combination with the visual information abundance and the movement complexity, and perform behavior construction processing on the cartoon character in combination with the plan execution resources and the behavior construction plan to obtain a character behavior construction result.

[0015] Compared with the problems described in the background technology, the present invention analyzes the emotional expression tendency of the cartoon character based on the character personality, and can obtain the emotional response pattern of the cartoon character in different situations, thereby laying a foundation for the subsequent plot development and character interaction of the cartoon character. Further, the present invention analyzes the character behavior motivation of the cartoon character in different scenes based on the plot scene information, and can deeply understand the driving factors behind the cartoon character's behavior, thereby improving the rationality and accuracy of constructing the behavior framework of the cartoon character in different scenes. Further, the present invention optimizes the details of each limb movement in the limb movement framework based on the character background story, so that the limb movement of the cartoon character can be more in line with its character background setting, enrich the action connotation, and lay the foundation for shaping a distinctive cartoon character. Further, the present invention generates the behavior construction scheme of the cartoon character by combining the optimized limb movement framework, the refined expression movement framework and the corrected language expression framework, which can shape the behavior performance of the cartoon character in all directions and lay the foundation for creating high-quality cartoon works. Therefore, the behavior construction method and system for realizing the cartoon character provided by the embodiment of the present invention can improve the behavior construction effect of the cartoon character. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic diagram of a flow chart of a method for implementing a behavior construction method of an animated character provided by an embodiment of the present invention; Figure 2 A schematic diagram of modules for implementing the method for constructing the behavior of an animated character provided by an embodiment of the present invention.

[0017] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings in conjunction with the embodiments. DETAILED DESCRIPTION

[0018] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0019] The embodiment of the present application provides a method for implementing the behavior construction of an animated character. The execution subject of the method for implementing the behavior construction of an animated character includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided by the embodiment of the present application. In other words, the method for implementing the behavior construction of an animated character can be executed by software or hardware installed on a terminal device or a server device. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0020] Embodiment 1: Reference Figure 1FIG. 1 is a flow chart of a method for implementing a behavior construction method for an animated character according to an embodiment of the present invention. In this embodiment, the method for implementing a behavior construction method for an animated character includes: S1. Obtain basic setting information and plot scene information of a cartoon character, wherein the basic setting information includes character personality and character background story, and analyze the emotional expression tendency of the cartoon character based on the character personality.

[0021] The present invention analyzes the emotional expression tendency of the cartoon character based on the character personality, and can obtain the emotional response pattern of the cartoon character in different situations, thereby laying a foundation for the subsequent creation of the plot development and character interaction of the cartoon character.

[0022] It should be explained that the basic setting information is the core element for constructing the image of an animated character, the character's personality determines the character's behavior and emotional tone, the character's background story provides the background basis for the formation of the character's personality, and the plot scene information is the specific situational environment in which the animated character is located.

[0023] In one embodiment of the present invention, analyzing the emotional expression tendency of the cartoon character based on the character personality includes: Extracting characteristics of the character to obtain characteristics of the character; Performing emotion mapping on the personality traits to obtain an emotion mapping result; Based on the emotion mapping result, identifying the dominant emotion characteristics in the character's personality; Performing intensity evaluation on the dominant emotional feature to obtain an emotional intensity value; Based on the emotion intensity value, the emotion expression tendency of the cartoon character is determined.

[0024] It should be explained that the personality traits are the key attributes of the character's personality, the emotion mapping result is the result of matching personality traits with emotion types, the dominant emotion traits are emotion-related attributes that play a dominant role in the character's personality, and the emotion intensity value is a numerical value that measures the degree of obviousness of the dominant emotion traits.

[0025] Furthermore, the feature extraction of the character's personality can be achieved through text analysis technology; the emotional mapping of the personality characteristics can be achieved through an emotional mapping model; based on the emotional mapping results, the dominant emotional characteristics in the character's personality are identified, for example, certain personality characteristics may indicate emotional tendencies such as optimism, pessimism, irritability, and gentleness, which can be identified as dominant emotional characteristics; the intensity evaluation of the dominant emotional characteristics can be achieved through an emotional intensity evaluation algorithm, such as an emotional intensity evaluation algorithm based on machine learning; based on the emotional intensity value, the emotional expression tendency of the cartoon character is determined. If the emotional intensity value shows that a cartoon character has a high irritability emotional intensity, when encountering a conflict scene, it may quickly show strong emotional reactions such as anger and excitement, and the frequency and degree of such reactions will be higher than other cartoon characters with weaker emotions.

[0026] S2. Based on the plot scene information, analyze the character behavior motivations of the cartoon character in different scenes, and based on the character behavior motivations, construct a behavior framework of the cartoon character in different scenes, wherein the behavior framework includes a body movement framework, an expression movement framework, and a language expression framework.

[0027] By analyzing the character behavior motivations of the cartoon character in different scenes based on the plot scene information, the present invention can deeply understand the driving factors behind the cartoon character's behavior, thereby improving the rationality and accuracy of constructing the cartoon character's behavior framework in different scenes.

[0028] It should be explained that the plot scene information is a collection of relevant information such as the story background, plot development, and environmental setting of the cartoon character, and the character behavior motivation is the internal psychological reason that prompts the cartoon character to take action in a specific scene.

[0029] In one embodiment of the present invention, analyzing the character behavior motivations of the cartoon character in different scenes based on the plot scene information includes: Performing plot deconstruction on the plot scene information to obtain scene plot elements; Performing sentiment analysis on the scene plot elements to obtain the element sentiment tendency; Combined with the emotional tendency of the elements, the motivation trigger points in the plot elements of the scenes are explored; Based on the motivation trigger points, the character behavior motivations of the cartoon character in different scenarios are summarized.

[0030] It should be explained that the scene plot elements are key plot fragments separated from the plot scene information, the element emotional tendency is the emotional color or emotional orientation contained in the scene plot elements, the motivation trigger point is the key factor in the scene plot elements that can trigger the behavioral motivation of the cartoon characters, and the character behavior motivation is a summary of the root causes of the cartoon character's behavior.

[0031] Furthermore, the plot deconstruction of the plot scene information can be achieved through a text segmentation algorithm; the sentiment tendency analysis of the scene plot elements can be achieved through a sentiment analysis model, such as a sentiment classification model based on deep learning; when mining the motivation trigger points in the scene plot elements, matching searches can be performed based on a pre-set motivation trigger rule library. For example, when dangerous information related to something the character cherishes appears in the scene plot elements, protection motivation may be triggered; the induction of the character behavior motivations of the cartoon characters in different scenes can be achieved through a motivation induction algorithm, such as a rule-based reasoning algorithm.

[0032] The present invention constructs a behavior framework of the cartoon character in different scenarios based on the character's behavior motivation, so as to comprehensively display the behavior performance of the cartoon character in different situations and provide strong support for character portrayal in animation creation. It should be explained that the behavior framework is a collection of comprehensive behavior norms and performance patterns of the cartoon character in different scenarios.

[0033] In one embodiment of the present invention, constructing the behavior framework of the cartoon character in different scenes based on the character's behavior motivation includes: Classifying the character's behavior motivations to obtain motivation behavior categories; Obtaining a character behavior template of the cartoon character, and calculating a behavior similarity coefficient between the motivation behavior category and each template in the character behavior template; When the behavior similarity coefficient is greater than a preset similarity coefficient, matching a behavior category template corresponding to the motivation behavior category from the character behavior template; Based on the behavior category template, a behavior framework of the cartoon character in different scenarios is constructed.

[0034] It should be explained that the motivational behavior category is the result of dividing the character's behavior motivation according to the behavior type, and the character behavior template is a collection of typical behavior patterns of the cartoon character in terms of body movements, facial expressions and language expressions under different behavior motivations. The behavior similarity coefficient represents the degree of similarity between the motivational behavior category and each template in the character behavior template. The preset similarity coefficient is the comparison standard value of the behavior similarity coefficient, which can be set to 0.8 or can be set according to the actual business scenario. The behavior category template is a set of specific behavior patterns in the character behavior template that matches the motivational behavior category, including body movements, facial expressions and language expression patterns, which are used to construct a complete behavior framework that conforms to the motivational behavior.

[0035] Furthermore, the behavior classification of the character's behavior motivation can be achieved through a behavior classification model, such as a decision tree classification model; the matching of corresponding body movement templates, facial expression movement templates and language expression templates can be achieved through a template matching algorithm, such as a template matching algorithm based on similarity calculation, which calculates the similarity between the motivation behavior category and each template, and selects the template with the highest similarity as the matching result.

[0036] For example, if the motivation of the cartoon character's behavior is to seek help, its body movement framework may include leaning forward, earnest eyes, and extending hands to make a request; the facial expression movement framework may show expressions such as slightly frowning brows, expectant eyes, and slightly open lips; the language expression framework may be the use of polite words with a requesting tone, such as "Excuse me, can you help me?" If the motivation of the behavior is to conceal a secret, the body movement framework may include defensive movements such as slightly leaning back and unconsciously crossing the arms in front of the chest; the facial expression movement framework may include nervous expressions such as flickering eyes and slightly drooping corners of the mouth; the language expression framework may be expressions such as speaking vaguely and deliberately changing the subject.

[0037] Further, as an optional embodiment of the present invention, the calculating of the behavior similarity coefficient between the motivation behavior category and each template in the character behavior template includes: Performing vectorization processing on the motivation behavior category and each template in the character behavior template respectively to obtain a behavior category vector and a behavior template vector; Respectively extracting subdivision features corresponding to the behavior category vector and the behavior template vector to obtain a category subdivision feature vector and a template subdivision feature vector; Combining the category segmentation feature vector and the template segmentation feature vector, the behavior similarity coefficient between the motivation behavior category and each template in the character behavior template is calculated by the following formula: ; Among them, A represents the behavioral similarity coefficient between the motivation behavior category and each template in the character behavior template, represents the a-th category segmentation feature vector under the i-th behavior category in the motivation behavior category, It represents the ath template segmentation feature vector corresponding to the i-th behavior category in the character behavior template, a represents the serial number corresponding to the category segmentation feature vector and the template segmentation feature vector, i represents the serial number corresponding to the motivation behavior category and the character behavior template, and q represents the number of segmentation feature vectors corresponding to the i-th motivation behavior category and the i-th character behavior template.

[0038] Among them, the behavior category vector and the behavior template vector are respectively data representation forms for measurement and comparison formed after quantifying and vectorizing the relevant behavior characteristics of each template in the motivation behavior category and the character behavior template, and the category segmentation feature vector and the template segmentation feature vector are respectively data vector forms for more accurate similarity analysis obtained after further refinement and decomposition of the segmentation characteristics under each behavior feature dimension corresponding to the behavior category vector and the behavior template vector. Furthermore, the motivation behavior category and each template in the character behavior template can be vectorized by a coding algorithm, such as a one-hot coding algorithm, to obtain a behavior category vector and a behavior template vector; the segmentation characteristics corresponding to the behavior category vector and the behavior template vector can be extracted by a feature parsing algorithm, to obtain a category segmentation feature vector and a template segmentation feature vector, such as an association rule algorithm.

[0039] S3. Based on the character's background story, each body movement in the body movement framework is optimized in detail to obtain an optimized body movement framework, the movement complexity corresponding to each body movement in the optimized body movement framework is calculated, and based on the emotional expression tendency, the expression changes in the expression action framework are refined to obtain a refined expression action framework, and behavioral performance cases of classic cartoon characters of the same type are collected. Based on the behavioral performance cases, the language expression paradigm in the language expression framework is corrected to obtain a corrected language expression framework.

[0040] The present invention optimizes the details of each body movement in the body movement framework based on the character background story, so that the body movement of the cartoon character can be more in line with the character background setting, enrich the movement connotation, and lay a foundation for shaping cartoon characters with distinct personalities. At the same time, by refining the expression changes in the expression action framework based on the emotional expression tendency, the expression and the emotional state of the character can be conveyed more accurately and vividly, thereby enhancing the character's expressiveness.

[0041] It should be explained that the character background story is an important basis for constructing the personality, experience and behavioral logic of the cartoon character. The body movement framework is a set of body behavior norms and patterns of the cartoon character in a specific scenario. The expression action framework is a set of expression change norms in the corresponding scenario. The optimized body movement framework is a set of body movements after detail optimization. The refined expression action framework is a set of expression changes after refinement.

[0042] In one embodiment of the present invention, the step of performing detail optimization processing on each limb movement in the limb movement framework based on the character background story includes: Extracting the plot of the character's background story to obtain the background story plot; Analyzing the correlation between the background storyline and each body movement in the body movement framework; Extracting background adaptation details corresponding to each body movement in the body movement framework in the background storyline according to the correlation degree; Based on the background adaptation details, each limb movement in the limb movement framework is subjected to detail optimization processing to obtain an optimized limb movement framework.

[0043] It should be explained that the background story plot is the plot segment in the character's background story that has a key impact on the character's behavior, the correlation degree indicates the degree of logical connection between the background story plot and each physical movement in the physical movement framework, and the background adaptation details are action elements that are supplemented to the physical movements according to the character's background story and are consistent with his experience and personality.

[0044] Furthermore, the plot extraction of the character's background story can be achieved through a text analysis algorithm, such as an algorithm based on keyword extraction and semantic understanding; the analysis of the correlation can be achieved by constructing a correlation model, such as using a neural network model to train the corresponding relationship data between the background story plot and body movements, so as to determine the correlation; based on the correlation, the background story plot is analyzed using text analysis and semantic understanding technology to locate relevant paragraphs, and then through keyword extraction and context analysis, details such as "jumping to avoid stones, leaning downhill, and going around the passage sideways" in the running scene in the adventure experience are excavated, and these extracted details are sorted according to the sequence of actions and strength requirements. And the degree of fit with the character's personality is sorted and optimized, so as to obtain the background adaptation details corresponding to each body movement in the body movement framework in the background story; based on the background adaptation details, each body movement in the body movement framework is optimized in detail to obtain an optimized body movement framework. If the character's background story contains the key plot of "having undergone arduous martial arts training", for the "punching action in the battle scene", background adaptation details such as "quickly adjust breathing and pace before punching, keep the center of gravity stable, quickly resume the defensive posture after punching and make clean movements" can be added to obtain an optimized body movement framework.

[0045] The present invention can quantitatively evaluate the complexity of limb movements by calculating the action complexity corresponding to each limb movement in the optimized limb movement framework, and provide a reference for action design and resource allocation in animation production. It should be explained that the action complexity represents the quantitative value of the complexity corresponding to each limb movement in the optimized limb movement framework.

[0046] In one embodiment of the present invention, the step of calculating the motion complexity corresponding to each limb motion in the optimized limb motion framework includes: Performing action decomposition processing on each limb action in the optimized limb action framework to obtain action sub-elements, and determining the action duration period corresponding to the action sub-elements; Calculating the kinematic parameters of each element in the action sub-element, and extracting complexity-related parameters from the kinematic parameters, wherein the complexity-related parameters include: motion amplitude value, motion speed, and motion direction change frequency; Combined with the action duration period, the movement amplitude value, the movement speed parameter value and the movement direction change frequency, the action complexity corresponding to each limb action in the optimized limb action framework is calculated by the following formula: ; Among them, E represents the action complexity corresponding to each limb action in the optimized limb action framework. , , They represent the parameter weights corresponding to the motion amplitude, motion speed and frequency of motion direction changes, respectively. Indicates the motion amplitude parameter corresponding to the bth element in the action sub-element. Indicates the motion speed parameter corresponding to the bth element in the action sub-element. Indicates the frequency of change in the motion direction corresponding to the bth element in the action sub-element. It represents the action duration period corresponding to the bth element in the action sub-element, and T represents the total duration period corresponding to each limb action in the optimized limb action framework.

[0047] Among them, the action sub-element is the basic constituent unit obtained after each limb action in the optimized limb action framework is disassembled, the action duration period is the time range from the start to the end corresponding to the action sub-element, the kinematic parameter is a quantitative indicator used by each element in the action sub-element to describe its motion characteristics in the space and time dimensions, the complexity-related parameter is the part of the kinematic parameter that is closely related to measuring the complexity of the action and plays a key role, and the motion amplitude value is the spatial displacement or angle change involved in the motion process of each element in the action sub-element, etc., which reflects the motion. The movement speed is the displacement of each element in the action sub-element per unit time to reflect the speed of movement. The movement direction change frequency is the number of times each element in the action sub-element changes its movement direction during movement. The parameter weight represents the importance of the movement amplitude value, movement speed and movement direction change frequency when comprehensively considering the complexity of the movement. The total duration period is the sum of the duration periods of the action sub-elements corresponding to each limb movement in the optimized limb movement framework.

[0048] Furthermore, the motion decomposition processing of each limb motion in the optimized limb motion framework can be achieved through the analysis and processing of motion capture data, for example, the joint motion trajectory in the three-dimensional motion capture data is split to obtain motion sub-elements; the motion duration period corresponding to the motion sub-elements can be determined by the motion capture device, and when a professional motion capture system is used, these devices will generate data with timestamps for each limb motion and its sub-elements; the kinematic parameters of each element in the action sub-elements can be calculated by the parameter measurement tool in the animation production software, for example, for the action sub-element "character jump", in the animation key frame, the software can measure the vertical displacement of the character's body center of gravity between the take-off frame and the highest point frame as the motion amplitude value, and the motion speed can be calculated through the time difference and displacement between the two frames. As for the frequency of changes in the motion direction, the software can analyze the character's body direction vector in different The changes between the same key frames are counted, and the number of direction changes during the jump process is counted to comprehensively obtain the kinematic parameters of the action sub-element; the complexity-related parameters can be extracted from the kinematic parameters according to the style requirements of the action, the character's personality characteristics, and the requirements for creating the scene atmosphere. For example, in anime with a hot-blooded fighting style, for the attack actions of power-type characters, the complexity-related parameters may be determined more based on kinematic parameters such as movement amplitude and movement speed, because the actions of such characters emphasize the presentation of strength and impact, and a larger movement amplitude and a faster movement speed can reflect the high complexity and powerful power of the action; and in scenes with delicate emotional expression, some subtle actions of the character, such as trembling and flicking, may be mainly based on kinematic parameters such as the frequency of changes in movement direction and relatively small changes in movement amplitude to extract complexity-related parameters; the parameter weights can be set through the hierarchical analysis method.

[0049] The present invention can make the emotion transmission of the cartoon character more accurate and realistic by refining the expression changes in the expression action framework based on the emotion expression tendency, collects behavioral performance cases of classic cartoon characters of the same type, and corrects the language expression paradigm in the language expression framework based on the behavioral performance cases, so as to improve the accuracy of the subsequent language expression of the cartoon character. It should be explained that the refined expression action framework is a more accurate and delicate expression action set obtained by refining the expression changes in the expression action framework based on the emotion expression tendency, and the corrected language expression framework is a more accurate expression set obtained by correcting the language expression paradigm in the language expression framework based on the behavioral performance cases. Further The refinement process of the expression changes in the expression action framework is as follows: the emotion expression tendency is graded by the emotion analysis model to obtain the emotion intensity level, such as the emotion classification model based on deep learning, and the expression changes in the expression action framework are refined and adjusted according to the emotion intensity level to obtain the refined expression action framework. When the expression changes in the expression action framework are refined and adjusted according to the emotion intensity level, if the emotion expression tendency is at the "high anger" level, for the "angry expression", the expression elements such as the frowning degree is deepened, the eyes are widened and the eyes are more aggressive, the mouth is opened and the teeth are gnashed can be refined and adjusted to obtain the refined expression action framework; the behavior performance cases of the same type of classic cartoon characters can be collected from the cartoon database.

[0050] In one embodiment of the present invention, the language expression paradigm in the language expression framework is corrected based on the behavior performance case to obtain a corrected language expression framework, including: Extracting discourse elements from the behavior performance case to obtain case discourse elements; Analyze the fit between the case discourse elements and each language expression paradigm in the language expression framework; Extracting, based on the degree of fit, paradigm correction details corresponding to each language expression paradigm in the language expression framework in the case discourse elements; Based on the paradigm correction details, each language expression paradigm in the language expression framework is corrected to obtain a corrected language expression framework.

[0051] It should be explained that the case discourse elements are the discourse fragments in the behavioral performance cases that can reflect the character's language style, word usage habits and expression logic. The degree of fit indicates the degree of match between the case discourse elements and each language expression paradigm in the language expression framework. The paradigm correction details are the language expression adjustment elements that are supplemented to the language expression paradigm based on the behavioral performance cases and are in line with the characteristics of the scene and role.

[0052] Furthermore, the extraction of discourse elements of the behavioral performance cases can be achieved through natural language processing technology, such as algorithms based on sentence structure analysis and word frequency statistics; the analysis of the degree of fit can be achieved by constructing a similarity model, such as using a deep learning model to train the associated data of the case discourse elements and language expression paradigms, so as to judge the degree of fit; based on the degree of fit, text mining and semantic association technology are used to analyze the case discourse elements, locate related sentences, and then through vocabulary comparison and contextual reasoning, dig out details such as the "firm eyes and loud shouts of 'We fight side by side, nothing can stop us from moving forward!'" when the protagonist in the adventure animation encourages his companions, and these extracted details are sorted according to the emotional intensity of expression and sentence structure. The structure and the degree of fit with the role identity are sorted and optimized, so as to obtain the paradigm correction details corresponding to each language expression paradigm in the language expression framework in the case discourse elements; based on the paradigm correction details, each language expression paradigm in the language expression framework is corrected to obtain a corrected language expression framework. If the behavioral performance case contains the key element of "the language style of the wise mentor role when teaching disciples", for the "explanation language paradigm in the knowledge transfer scenario", paradigm correction details such as "start with a thought-provoking question, then use easy-to-understand and philosophical words to gradually analyze, and end with encouraging words to guide disciples to think deeply" can be added to obtain a corrected language expression framework.

[0053] S4. Combining the optimized limb movement framework, the refined facial expression movement framework and the corrected language expression framework, generate a behavior construction plan for the cartoon character, calculate the visual information abundance corresponding to each plan in the behavior construction plan, combine the visual information abundance and the movement complexity, allocate the plan execution resources corresponding to the behavior construction plan, combine the plan execution resources and the behavior construction plan, execute the behavior construction processing of the cartoon character, and obtain the character behavior construction result.

[0054] The present invention generates the behavior construction scheme of the cartoon character by combining the optimized limb action framework, the refined facial expression action framework and the corrected language expression framework, which can shape the behavior performance of the cartoon character in all directions and lay the foundation for creating high-quality cartoon works. At the same time, the present invention calculates the visual information abundance corresponding to the behavior construction scheme and allocates the execution resources of the scheme in combination with the action complexity, which can realize the reasonable allocation of resources and improve the efficiency of animation production and the quality presentation effect.

[0055] It should be explained that the behavior construction plan is a complete behavior plan for cartoon characters that integrates body movements, facial expressions and language expressions. The visual information richness is a quantitative indicator of the richness and detail of visual perception in the behavior construction plan. The plan execution resources are the human, material, time and other resource inputs required to implement the behavior construction plan. The character behavior construction result is the complete behavior performance of the cartoon character in a specific scene that is finally presented. Furthermore, the generation steps of the behavior construction plan for the cartoon character are as follows: first, in-depth analysis of the action details and plot adaptability in the optimized body movement framework, clarifying the role and connection method of body movements in the cartoon story; secondly, analyzing the emotional logic of expression changes in the refined facial expression action framework, and echoing it with the emotional context in the plot development; finally, considering the degree of integration of the language paradigm and the scene in the corrected language expression framework, organically combining the three, and based on the role positioning, plot direction and emotional tone, comprehensively constructing a behavior construction plan that can fully display the personality and behavior logic of the cartoon character.

[0056] In one embodiment of the present invention, the calculating the visual information abundance corresponding to each scheme in the behavior construction scheme includes: Performing a screen rendering simulation on the behavior construction scheme to obtain a scheme rendering screen; Performing grayscale processing on the rendering picture of the scheme to obtain a grayscale scheme rendering picture; Extracting picture texture features from the grayscale rendering picture, and calculating picture texture complexity corresponding to the grayscale rendering picture based on the picture texture features; Counting the grayscale values ​​of pixels in the grayscale scheme picture, and calculating the picture information entropy of the grayscale scheme picture based on the pixel grayscale values; Combining the picture information entropy and the picture texture complexity, the visual information abundance corresponding to each scheme in the behavior construction scheme is calculated.

[0057] Among them, the scheme rendering picture is a visual presentation generated by simulating rendering of each scheme in the behavior construction scheme through professional graphics rendering software or tools; the grayscale scheme rendering picture is a single-channel grayscale image obtained after the scheme rendering picture is processed by an image grayscale conversion algorithm; the picture texture feature is the type, distribution, scale, shape and other related visual characteristics of texture elements in the grayscale scheme rendering picture; the picture texture complexity is a complexity index corresponding to the grayscale scheme rendering picture, which is quantified after comprehensively considering various aspects of the texture characteristics; the pixel grayscale value is the grayscale value of each pixel in the grayscale scheme picture; the picture information entropy is a numerical value reflecting the amount of picture information and the degree of disorder based on the distribution of pixel grayscale values ​​of the grayscale scheme picture according to the information entropy calculation formula.

[0058] Furthermore, the behavior construction scheme can be simulated by rendering the screen through professional animation production software or graphics rendering engine to obtain the scheme rendering screen; the scheme rendering screen can be gray-scaled by image color space conversion algorithm (such as weighted average method, etc.) to obtain gray-scale scheme rendering screen; the texture features of the screen in the gray-scale scheme rendering screen can be extracted by texture analysis algorithm (such as based on gray-scale co-occurrence matrix, wavelet transform, etc.), and based on the screen texture features, the screen texture complexity corresponding to the gray-scale scheme rendering screen is calculated by considering multiple factors such as texture element type, distribution and spatial hierarchy. First, the number of types of texture elements in the screen texture features is analyzed and assigned The more types there are, the higher the accumulated weight is. Secondly, the uniformity of texture element distribution is considered. The higher the unevenness is, the higher the complexity score is. Finally, various factors are comprehensively calculated to obtain the value of the texture complexity of the grayscale scheme rendering picture. The pixel grayscale values ​​in the grayscale scheme picture can be counted by a histogram statistical method. Based on the pixel grayscale values, the picture information entropy of the grayscale scheme picture is calculated according to the information entropy calculation formula. The picture information entropy and the picture texture complexity are standardized to obtain the standard information entropy and standard complexity. The standard information entropy and the standard complexity are added to obtain the visual information abundance corresponding to each scheme in the behavior construction scheme.

[0059] The present invention can accurately allocate the solution execution resources corresponding to the behavior construction scheme by combining the visual information abundance and the action complexity, improve the subsequent animation production efficiency, and perform the behavior construction processing of the animation character in combination with the solution execution resources and the behavior construction scheme, thereby improving the behavior construction effect of the animation character. It should be explained that the solution execution resources are the resources required for the behavior construction scheme in the production process. Further, in combination with the visual information abundance and the action complexity, the solution execution resources corresponding to the behavior construction scheme are allocated. First, the complexity and refinement of the picture are judged according to the visual information abundance, and the required rendering technology level and art manpower investment are determined in combination with historical data of the same type. Then, the difficulty of action design and arrangement is analyzed according to the action complexity, and the number of action design professionals and time consumption required are clarified. Then, the two conditions are comprehensively considered, and the matching animation production software permissions, hardware equipment and facilities are allocated, and the time period of each link is determined, thereby completing the allocation of solution execution resources corresponding to the behavior construction scheme.

[0060] Compared with the problems described in the background technology, the present invention analyzes the emotional expression tendency of the cartoon character based on the character personality, and can obtain the emotional response pattern of the cartoon character in different situations, thereby laying a foundation for the subsequent plot development and character interaction of the cartoon character. Further, the present invention analyzes the character behavior motivation of the cartoon character in different scenes based on the plot scene information, and can deeply understand the driving factors behind the cartoon character's behavior, thereby improving the rationality and accuracy of constructing the behavior framework of the cartoon character in different scenes. Further, the present invention optimizes the details of each limb movement in the limb movement framework based on the character background story, so that the limb movement of the cartoon character can be more in line with its character background setting, enrich the action connotation, and lay the foundation for shaping a distinctive cartoon character. Further, the present invention generates the behavior construction scheme of the cartoon character by combining the optimized limb movement framework, the refined expression movement framework and the corrected language expression framework, which can shape the behavior performance of the cartoon character in all directions and lay the foundation for creating high-quality cartoon works. Therefore, the behavior construction method and system for realizing the cartoon character provided by the embodiment of the present invention can improve the behavior construction effect of the cartoon character.

[0061] Embodiment 2: like Figure 2 The figure shows a functional module diagram of a system for realizing the behavior construction of cartoon characters according to the present invention.

[0062] The behavior construction system 200 for realizing an animated character described in the present invention can be installed in an electronic device. According to the functions to be realized, the behavior construction system for realizing an animated character can include an emotion expression tendency analysis module 201, a behavior framework construction module 202, a framework correction module 203 and a behavior construction module 204. The module described in the present invention can also be called a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0063] In the embodiment of the present invention, the functions of each module / unit are as follows: The emotion expression tendency analysis module 201 is used to obtain basic setting information and plot scene information of the cartoon character, wherein the basic setting information includes the character's personality and the character's background story, and analyze the emotion expression tendency of the cartoon character based on the character's personality; The behavior framework construction module 202 is used to analyze the character behavior motivations of the cartoon character in different scenes based on the plot scene information, and to construct the behavior framework of the cartoon character in different scenes based on the character behavior motivations, wherein the behavior framework includes a body movement framework, an expression movement framework, and a language expression framework; The framework correction module 203 is used to optimize the details of each limb movement in the limb movement framework based on the character background story to obtain an optimized limb movement framework, calculate the movement complexity corresponding to each limb movement in the optimized limb movement framework, refine the expression changes in the expression movement framework based on the emotional expression tendency to obtain a refined expression movement framework, collect behavioral performance cases of classic cartoon characters of the same type, and based on the behavioral performance cases, correct the language expression paradigm in the language expression framework to obtain a corrected language expression framework; The behavior construction module 204 is used to combine the optimized limb movement framework, the refined facial expression movement framework and the corrected language expression framework to generate a behavior construction plan for the cartoon character, calculate the visual information abundance corresponding to each plan in the behavior construction plan, combine the visual information abundance and the movement complexity, allocate the plan execution resources corresponding to the behavior construction plan, combine the plan execution resources and the behavior construction plan, perform behavior construction processing on the cartoon character, and obtain a character behavior construction result.

[0064] In detail, the modules in the system 200 for implementing the behavior construction of an animated character in the embodiment of the present invention are used in the same manner as described above. Figure 1 The same technical means as described in the method for realizing the behavior construction of cartoon characters can produce the same technical effects, so I will not go into details here.

[0065] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0066] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A method for realizing the behavior construction of an animated character, characterized in that: The method comprises: Acquire basic setting information and plot scene information of the cartoon character, wherein the basic setting information includes the character's personality and the character's background story, and analyze the emotional expression tendency of the cartoon character based on the character's personality; Based on the plot scene information, analyzing the character behavior motivations of the cartoon character in different scenes, and based on the character behavior motivations, constructing the behavior framework of the cartoon character in different scenes, the behavior framework including a body movement framework, an expression movement framework and a language expression framework; Based on the character background story, each limb movement in the limb movement framework is optimized in detail to obtain an optimized limb movement framework, the movement complexity corresponding to each limb movement in the optimized limb movement framework is calculated, and based on the emotional expression tendency, the expression changes in the expression movement framework are refined to obtain a refined expression movement framework, and behavioral performance cases of classic cartoon characters of the same type are collected. Based on the behavioral performance cases, the language expression paradigm in the language expression framework is corrected to obtain a corrected language expression framework; In combination with the optimized limb movement framework, the refined facial expression movement framework and the corrected language expression framework, a behavior construction scheme for the cartoon character is generated, the visual information abundance corresponding to each scheme in the behavior construction scheme is calculated, and in combination with the visual information abundance and the movement complexity, scheme execution resources corresponding to the behavior construction scheme are allocated, and in combination with the scheme execution resources and the behavior construction scheme, the behavior construction processing of the cartoon character is executed to obtain a character behavior construction result.

2. The method for realizing the behavior construction of an animated character according to claim 1, characterized in that: The analyzing the emotional expression tendency of the cartoon character based on the character personality includes: Extracting characteristics of the character to obtain characteristics of the character; Performing emotion mapping on the personality traits to obtain an emotion mapping result; Based on the emotion mapping result, identifying the dominant emotion characteristics in the character's personality; Performing intensity evaluation on the dominant emotional feature to obtain an emotional intensity value; Based on the emotion intensity value, the emotion expression tendency of the cartoon character is determined.

3. The method for realizing the behavior construction of an animated character according to claim 1, characterized in that: The analyzing the character behavior motivations of the cartoon character in different scenes based on the plot scene information includes: Performing plot deconstruction on the plot scene information to obtain scene plot elements; Performing sentiment analysis on the scene plot elements to obtain the element sentiment tendency; Combined with the emotional tendency of the elements, the motivation trigger points in the plot elements of the scenes are explored; Based on the motivation trigger points, the character behavior motivations of the cartoon character in different scenarios are summarized.

4. The method for realizing the behavior construction of an animated character according to claim 1, characterized in that: The step of constructing a behavior framework of the cartoon character in different scenarios based on the character's behavior motivation includes: Classifying the character's behavior motivations to obtain motivation behavior categories; Obtaining a character behavior template of the cartoon character, and calculating a behavior similarity coefficient between the motivation behavior category and each template in the character behavior template; When the behavior similarity coefficient is greater than a preset similarity coefficient, matching a behavior category template corresponding to the motivation behavior category from the character behavior template; Based on the behavior category template, a behavior framework of the cartoon character in different scenarios is constructed.

5. The method for realizing the behavior construction of an animated character according to claim 4, characterized in that: The calculating of the behavior similarity coefficient between the motivation behavior category and each template in the character behavior template includes: Performing vectorization processing on the motivation behavior category and each template in the character behavior template respectively to obtain a behavior category vector and a behavior template vector; Respectively extracting subdivision features corresponding to the behavior category vector and the behavior template vector to obtain a category subdivision feature vector and a template subdivision feature vector; Combining the category segmentation feature vector and the template segmentation feature vector, the behavior similarity coefficient between the motivation behavior category and each template in the character behavior template is calculated by the following formula: ； Among them, A represents the behavioral similarity coefficient between the motivation behavior category and each template in the character behavior template, represents the a-th category segmentation feature vector under the i-th behavior category in the motivation behavior category, It represents the ath template segmentation feature vector corresponding to the i-th behavior category in the character behavior template, a represents the serial number corresponding to the category segmentation feature vector and the template segmentation feature vector, i represents the serial number corresponding to the motivation behavior category and the character behavior template, and q represents the number of segmentation feature vectors corresponding to the i-th motivation behavior category and the i-th character behavior template.

6. The method for realizing the behavior construction of an animated character according to claim 1, characterized in that: The step of performing detail optimization processing on each body movement in the body movement framework based on the character background story includes: Extracting the plot of the character's background story to obtain the background story plot; Analyzing the correlation between the background storyline and each body movement in the body movement framework; Extracting background adaptation details corresponding to each body movement in the body movement framework in the background storyline according to the correlation degree; Based on the background adaptation details, each limb movement in the limb movement framework is subjected to detail optimization processing to obtain an optimized limb movement framework.

7. The method for realizing the behavior construction of an animated character according to claim 1, characterized in that: The calculating the action complexity corresponding to each limb action in the optimized limb action framework includes: Performing action decomposition processing on each limb action in the optimized limb action framework to obtain action sub-elements, and determining the action duration period corresponding to the action sub-elements; Calculating the kinematic parameters of each element in the action sub-element, and extracting complexity-related parameters from the kinematic parameters, wherein the complexity-related parameters include: motion amplitude value, motion speed, and motion direction change frequency; Combined with the action duration period, the movement amplitude value, the movement speed parameter value and the movement direction change frequency, the action complexity corresponding to each limb action in the optimized limb action framework is calculated by the following formula: ； Among them, E represents the action complexity corresponding to each limb action in the optimized limb action framework. , , They represent the parameter weights corresponding to the motion amplitude, motion speed and frequency of motion direction changes, respectively. Indicates the motion amplitude parameter corresponding to the bth element in the action sub-element. Indicates the motion speed parameter corresponding to the bth element in the action sub-element. Indicates the frequency of change in the motion direction corresponding to the bth element in the action sub-element. It represents the action duration period corresponding to the bth element in the action sub-element, and T represents the total duration period corresponding to each limb action in the optimized limb action framework.

8. The method for realizing the behavior construction of an animated character according to claim 1, characterized in that: The correcting the language expression paradigm in the language expression framework based on the behavior performance case to obtain a corrected language expression framework includes: Extracting discourse elements from the behavior performance case to obtain case discourse elements; Analyze the fit between the case discourse elements and each language expression paradigm in the language expression framework; Extracting, based on the degree of fit, paradigm correction details corresponding to each language expression paradigm in the language expression framework in the case discourse elements; Based on the paradigm correction details, each language expression paradigm in the language expression framework is corrected to obtain a corrected language expression framework.

9. The method for realizing the behavior construction of an animated character according to claim 1, characterized in that: The calculating the visual information abundance corresponding to each scheme in the behavior construction scheme comprises: Performing a screen rendering simulation on the behavior construction scheme to obtain a scheme rendering screen; Performing grayscale processing on the rendering picture of the scheme to obtain a grayscale scheme rendering picture; Extracting picture texture features from the grayscale rendering picture, and calculating picture texture complexity corresponding to the grayscale rendering picture based on the picture texture features; Counting the grayscale values ​​of pixels in the grayscale scheme picture, and calculating the picture information entropy of the grayscale scheme picture based on the pixel grayscale values; Combining the picture information entropy and the picture texture complexity, the visual information abundance corresponding to each scheme in the behavior construction scheme is calculated.

10. A system for implementing the behavior construction of anime characters, characterized in that: The system comprises: An emotion expression tendency analysis module is used to obtain basic setting information and plot scene information of an animated character, wherein the basic setting information includes character personality and character background story, and analyze the emotion expression tendency of the animated character based on the character personality; A behavior framework construction module, for analyzing the character behavior motivations of the cartoon character in different scenes based on the plot scene information, and constructing the behavior framework of the cartoon character in different scenes based on the character behavior motivations, wherein the behavior framework includes a body movement framework, an expression movement framework, and a language expression framework; A framework correction module is used to optimize the details of each limb movement in the limb movement framework based on the character background story to obtain an optimized limb movement framework, calculate the movement complexity corresponding to each limb movement in the optimized limb movement framework, refine the expression changes in the expression movement framework based on the emotional expression tendency to obtain a refined expression movement framework, collect behavioral performance cases of classic cartoon characters of the same type, and based on the behavioral performance cases, correct the language expression paradigm in the language expression framework to obtain a corrected language expression framework; The behavior construction module is used to combine the optimized limb movement framework, the refined facial expression movement framework and the corrected language expression framework to generate a behavior construction plan for the cartoon character, calculate the visual information abundance corresponding to each plan in the behavior construction plan, allocate the plan execution resources corresponding to the behavior construction plan in combination with the visual information abundance and the movement complexity, and perform behavior construction processing on the cartoon character in combination with the plan execution resources and the behavior construction plan to obtain a character behavior construction result.

Images