KR101348282B1 - Method for generating animation from text, Apparatus thereof - Google Patents

Abstract

The present invention relates to a method for generating an animation, comprising the steps of: extracting an animation element necessary for generating an animation from text; and generating the animation by using the extracted animation element and an image corresponding to the animation element. The animation element may be at least one of an object, a background, an emotion, an action, and an interaction, and may automatically generate an animation by receiving text.

Description

Method and generating animation from text, Apparatus approx.

The present invention relates to a method for generating an animation from text, and more particularly, to an animation generating method, an animation generating device, and a recording medium for automatically generating animation by receiving text.

Content refers to a variety of creations generated from music, movies, cartoons, animations, games, etc. provided by books or web pages or the press and media. Recently, the digital content industry is receiving more attention due to the development of the Internet environment, and the digital content industry such as online games, animation, and e-learning is predicting a great growth in the global market in the future. In recent years, there has been a change in the paradigm of education in which infants and adolescents frequently encounter such information and communication technologies and use advanced technology as an educational medium. This is in stark contrast to the earlier closed forms of use, where there was a rather negative view of the use of computers for infant content.

In the late 1980s, educational software suitable for the age and development of infants began to be developed. In the mid-1990s, the e-learning emerged as an alternative education and learning strategy, and the content development industry, which is a central element, attracted attention. E-book is one of the representative contents of the e-learning is a typical application that can be used not only in the computer but also in mobile devices such as mobile phones, PMP, PDA. Currently, e-books for toddlers are composed of animation books composed of animations, and the narrator reads the texts of fairy tales according to the background music.

The importance of the database is essential to producing such digital media. The introduction of digital media has transformed the 21st century into the era of the database paradigm, which has brought about major changes in the production and industry of animation. Originally, animation has the feature of creating all images newly, but with the introduction of digital technology, images created once for animation are stored as they are in the database system. To create them. In other words, digital animation is pursuing efficiency and economic feasibility in the age of infinite edition by utilizing database system.

Recently, due to the development of technology related to digital media, an environment in which an individual can directly develop contents is being created, and thus the development potential of the content development industry is increasing. However, for non-experts who do not have knowledge about computer-based media production and computer language, it is not easy to try digital content creation.

The first problem to be solved by the present invention is to provide an animation generating method for automatically generating animation by receiving text.

The second problem to be solved by the present invention is to provide an animation generating device that automatically generates animation by receiving text.

It is another object of the present invention to provide a computer-readable recording medium storing a program for causing a computer to execute the above-described method.

The present invention includes the steps of extracting an animation element required for generating an animation from text, and generating the animation using the extracted animation element and the image corresponding to the animation element in order to achieve the first object. The animation element provides at least one of an object, a background, an emotion, an action, and an interaction.

According to an embodiment of the present invention, the extracting of the animation element may include generating syntax information by analyzing the vocabulary of the text, and analyzing the meaning of each syntax from the syntax information corresponding to each syntax. The method may include extracting the animation element, and among the animation elements, an object and a background are a noun, an emotion is an adjective, and an action and an interaction correspond to a verb. The generating may include analyzing a morpheme and a part-of-speech of each word constituting the text, and generating a syntax tree for the text using the analysis result.

The extracting of the animation element corresponding to each word may include extracting a candidate word that may correspond to a subject, a predicate, a background object, or an interactive object from the syntax information, and to select a user. Selecting a word corresponding to the subject, object, descriptive word, background object, or interactive object; and determining the type of the selected word to determine where the word corresponds to the animation element. And extracting a candidate word corresponding to a subject, object, descriptor, background object, or interaction entity from the syntax information, extracting a noun that may be distinguished from the subject as a candidate, and Objects are extracted as candidates for nouns that can be distinguished as objects, and the predicates are extracted as candidates for verbs and adjectives. The interactive entity extracts a noun connected to "to" or "to" as a candidate, and the background entity extracts a noun connected to "from" or "to" as a candidate. You can do

Further, the extracting of the animation element may include analyzing each word of the text by comparing each word of the text with a word stored in an animation element database, and extracting an animation element from the word using the meaning of the analyzed word. Extracting, and storing the extracted animation element together with the word of the text in the animation element database.

The generating of the animation may include generating an animation generating language using the extracted animation element, and generating the animation using the generated animation generating language. have.

In order to achieve the second object, the present invention provides an animation element extracting unit for generating an animation element from text and an animation generating unit for generating an animation using the extracted animation element and the image corresponding to the animation element. The animation element provides at least one of an object, a background, an emotion, an action, and an interaction.

In order to solve the above other technical problem, the present invention provides a computer readable recording medium having recorded thereon a program for executing the above-described animation generating method on a computer.

According to the present invention, the text can be received and the animation can be automatically generated by analyzing the text. In addition, according to the present invention, the use of the axis of movement can naturally represent the behavior of the individual. Furthermore, by using an animation generating language, the animation information can be easily shared with other users, and even if the application program is different, it is compatible.

1 is a block diagram illustrating an apparatus for generating animation according to an embodiment of the present invention.
2 is a flowchart of a method for generating animation according to an embodiment of the present invention.
3 is a flowchart of a method of extracting an animation element according to an embodiment of the present invention.
4 is a flowchart of a method of extracting an animation element from syntax information according to an embodiment of the present invention.
5 is a flowchart of a method of using an animation database to extract animation elements according to an embodiment of the present invention.
6 is a flowchart illustrating a method of generating an animation using an animation generation language according to an embodiment of the present invention.

Prior to the description of the concrete contents of the present invention, for the sake of understanding, the outline of the solution of the problem to be solved by the present invention or the core of the technical idea is first given.

An animation generating method according to an embodiment of the present invention includes the steps of extracting an animation element necessary for generating an animation from text, and generating the animation using the extracted animation element and the image corresponding to the animation element. The animation element is at least one of an object, a background, an emotion, an action, and an interaction.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. It will be apparent to those skilled in the art, however, that these examples are provided to further illustrate the present invention, and the scope of the present invention is not limited thereto.

BRIEF DESCRIPTION OF THE DRAWINGS The above and other objects, features and advantages of the present invention will become more apparent from the following detailed description of the present invention when taken in conjunction with the accompanying drawings, in which: It is to be noted that components are denoted by the same reference numerals even though they are shown in different drawings, and components of different drawings can be cited when necessary in describing the drawings. In the following detailed description of the principles of operation of the preferred embodiments of the present invention, it is to be understood that the present invention is not limited to the details of the known functions and configurations, and other matters may be unnecessarily obscured, A detailed description thereof will be omitted.

1 is a block diagram illustrating an apparatus for generating animation according to an embodiment of the present invention.

An animation generating apparatus according to an embodiment of the present invention is composed of an animation element extraction unit 110 and an animation generation unit 120 may further include an animation element database 130 and a media database 140.

The animation element extractor 110 extracts an animation element from text.

More specifically, animation elements consist of objects, backgrounds, emotions, actions, and interactions for generating animation. Korean basic sentence patterns include subject, object and predicate. It may further include features of the interactive object and the background object. Each word is given from the input text, and is divided into an object, a descriptor, an interactive object, and a background object, and the kind of the separated word is classified to extract the animation element. The subject, object corresponds to the entity, the descriptor corresponds to emotion and action, the interactive entity corresponds to the interaction, the background entity corresponds to the background. Syntax information is generated by analyzing a vocabulary of the input text, and the animation element is extracted by analyzing the meaning of each phrase from the syntax information. The text may be text described in a book or scenario including text, such as a fairy tale book or a picture book.

In extracting the animation element, morphological analysis and part-of-speech attachment are performed on each word constituting the text, and syntax information of the text is generated using the result of the morphological analysis and part-of-speech attachment. The meaning of each word may be analyzed from syntax information to extract the animation element corresponding to each word. The animation element corresponding to each word extracts a candidate word that may correspond to a subject, a predicate, a background object, or an interactive entity from the syntax information, and receives the user's selection to receive the subject, object, and predicate. A word corresponding to a background object or an interactive object is selected, and the selected word may be classified to determine whether the word corresponds to the animation element and extracted. The subject extracts a noun that can be distinguished as a subject, the object extracts a noun that can be distinguished as an object, the predicate extracts a verb and an adjective as a candidate, and the interactive entity " Nouns connected to "to" or "to" may be extracted as candidates, and the background entity may extract nouns connected to "from" or "to" as candidates.

Further, in extracting the animation element, the meaning of each word is analyzed by comparing each word of the text with a word stored in the animation element database 130 and using the meaning of the analyzed word, the animation element from the word. The extracted animation element may be stored in the animation database 130 along with the word of the text.

When there are two or more candidate candidates for animation elements extracted from each word, the user may be able to select the candidate animation elements. In addition, the animation element having a high probability of selection may be recommended using the information stored in the animation element database 130. The animation element extracting unit 110 may extract an animation element corresponding to each word in response to the user's input.

A detailed description of the method of extracting the animation element will be given with reference to FIGS. 3 to 4.

The animation generator 120 generates an animation using the animation elements extracted by the animation element extractor 110.

More specifically, when the background image corresponding to the animation element is set as a background, and the object image is expressed by operating the object image, and there is an interaction between the object and another object, the object image is different from the object image. Create animations by interacting with object images. In addition, an animation may be generated using at least one of voice, music, and effect sounds corresponding to the animation element and stored in the media database 140.

The animation element database 130 stores animation elements corresponding to syntax information. The animation element database 130 stores information on the meaning of the word and which of the animation elements corresponds to the word. The word may correspond to at least one of an animation element of an object, a background, an emotion, an action, and an interaction. Information stored in the animation element database 130 may be accumulated and stored. The information stored in the animation database 130 may be used to extract the animation element, and the result may be stored in the animation element database 130 to accumulate the information. In extracting animation elements by learning using cumulative storage of the information, accuracy can be increased. In extracting an animation element from the word, a probability according to the learning may be applied. Corresponding animation elements may vary according to parts of speech and meaning used even for the same word, and two or more candidates may be corresponding animation information. Therefore, when there are two or more candidates, a plurality of candidate animation elements are used by searching for information up to the past, so that an animation element having a high probability can be selected. Alternatively, the user may select a candidate animation element and determine the animation element according to the user's input. The user's selection may be stored in the animation element database 130 again to use the animation element for the next text for extraction. In the case of the same book, an animation element for a particular word can continue to be used when the word is next entered.

The media database 140 may store an image, voice, music, or sound effect corresponding to the animation element. Corresponding required media elements are stored according to animation elements and can be used to generate animations. The image may be any one of an image extracted by analyzing a picture input together with the text or an image stored in the media database 140. Like a fairy tale book, a picture of a book on which an illustration is drawn may be input, and the extracted image may be used by analyzing the picture. The object and the background can be extracted from the figure and used to create an animation. Alternatively, when there is no picture in the book and there is no input image, or there is no image corresponding to the background or the object, an image previously stored in the media database 140 may be used.

In addition, the media database 140 may store not only an image but also a voice, music, and an effect sound. In generating an animation, not only an image but also a voice, music, effect sound, etc. are required, and thus, a voice, music, effect sound, etc. for generating an animation may be stored. The voice may use a voice actor's voice stored in advance. If there is a speaking in the action among the animation elements, the corresponding dialogue may be expressed using the voice of the voice actor. It also stores background music, sound effects, and more, so you can add voice to your animation.

2 is a flowchart of a method for generating animation according to an embodiment of the present invention.

Step 210 is a step of extracting animation elements required for animation generation from the text.

More specifically, an animation element corresponding to an object, a background, an emotion, an action, and an interaction required for generating an animation is extracted from the input text. The detailed description of this step corresponds to the detailed description of the animation element extraction unit 110 of FIG. 1, and is replaced with the detailed description of the animation element extraction unit 110 of FIG. 1.

In operation 220, the animation is generated using the extracted animation element and an image corresponding to the animation element.

More specifically, an animation is generated using the animation element extracted in step 110 and an image corresponding to the animation element. The image may be an object and a background among the animation elements. The detailed description of this step corresponds to the detailed description of the animation generator 120 of FIG. 1, and is replaced with the detailed description of the animation generator 120 of FIG. 1.

In addition, the object and the behavior of the object is extracted from the text, and in showing the behavior of the image corresponding to the object, the portion of the image that is the axis of movement for the behavior is analyzed, and the axis of the movement The image can be modified to represent the behavior of the subject. Although a plurality of images may be used to indicate the behavior of an object, an animation may be generated to allow the image to behave naturally by using movement of joints of the image corresponding to the object. For example, when the object is a human or an animal, the movement of the human image may be analyzed, that is, a portion corresponding to a joint, and the image may be transformed around the joint to represent the movement of the human. In the case of generating an animation of walking by a person, an animation of walking of the human image may be generated by deforming the human image so that the leg may move around the knee which is the axis of the movement of the leg of the human image.

3 is a flowchart of a method of extracting an animation element according to an embodiment of the present invention.

Step 310 is a step of performing morpheme analysis and part-of-speech for each word constituting the text.

More specifically, the input text is divided into word units to perform morphological analysis and part-of-speech attachment on each word. Morphological analysis means outputting all possible analysis results for a certain word, and attaching parts of speech refers to a process of selecting one analysis result that is appropriate to a context from the results of morphological analysis. In order to perform the morphological analysis, a rule-based approach that analyzes input statements using deterministic rules, a statistics-based approach that analyzes inputs using probabilities or statistics, or a method that combines the advantages of rule-based and statistical-based approaches. A mixed approach can be used. The morpheme of each word of the input text is analyzed using any one of the above methods. Analyze whether each word is a noun, a verb, an adjective. A part-of-speech attachment for selecting an analysis candidate suitable for the text is performed for a plurality of analysis candidates according to the morphological analysis result. In order to perform the part-of-speech attachment, a Markov model, a hidden Markov model, a predictive suffix tree model, or a crystal tree model may be used.

Step 320 is a step of generating syntax information for the text using the results of the morpheme analysis and the part-of-speech attachment.

More specifically, in step 310, the syntax information is generated to be suitable for semantic analysis of the result of the morpheme analysis and the part-of-speech attachment. With respect to the result of step 310, it is checked whether the vocabulary of the text is composed according to the grammar. That is, the results of the morphological analysis and the part-of-speech attachment are checked to see if they have a grammatical meaning. When the result of the morphological analysis and the part-of-speech attachment is grammatically constructed, a syntax tree may be generated. A syntax tree is a tree diagram of the structure of text for grammatically correct text.

In operation 330, the meaning of each syntax is analyzed from the syntax information to extract the animation element corresponding to each syntax.

More specifically, it is a step of analyzing the meaning of each phrase from the syntax information generated in step 320. The syntax information is grammatically correct and is a hierarchical structure of elements in the text and does not have meaning. The meaning is analyzed using meaningless syntax information to determine which animation elements each object corresponds to: object, background, emotion, behavior, and interaction. In the syntax information generated by the syntax tree, nouns may be entities, backgrounds, and interactions, adjectives may correspond to emotions, and verbs may correspond to actions and interactions.

First, it is possible to extract the subject, object, and predicate which are important elements of text from each syntax. The subject can be an entity and extracts all nouns separated by the subject as candidates for the subject. The object can also be an entity, and all nouns separated by the object are extracted as candidates for the object. Predicates can be emotions, actions, and extract verbs and adjectives in verbs and adjectives or conjunctive verbs that belong to ending endings, tubular endings, and subordinate clauses. For example, the text "bear in the woods threw a ball to a rabbit" could be extracted by giving it a "bear", extracting "ball" as the object, and extracting "throw" as a descriptor.

When each word is extracted as the subject, object, or descriptor, the type of the word may be assigned. In the case of nouns, they can be divided into figures, backgrounds, objects, or intangibles, and in the case of verbs, they can be divided into types such as go, eat, see, sit, no action, and animation. . The word type may be set in advance, or may be newly defined when a new noun or a verb is input.

You can extract the interaction object and background from each of the following statements: Nouns connected to "to" and "to" are extracted as candidates for the interactive entity, and nouns connected to "to" and "to" are extracted as candidates for the background. For example, you can extract the "rabbit" linked to "to" from "A bear throws a ball in a forest" as an interactive entity. In addition, the "forest" associated with "from" can be extracted in the background. In the case of the background, since the same background is often used for each animation scene, one background can be extracted for each scene. In the case of books, they can be extracted in paragraph units.

In the text "The bear threw the ball to the rabbit in the forest", the entities were "bear" and "rabbit", the background was "forest", the action was "threw", and the interaction was "bear to the rabbit." Threw ". An animation is generated using the animation element and an image corresponding to the animation element.

4 is a flowchart of a method of extracting an animation element from syntax information according to an embodiment of the present invention.

In operation 410, a candidate word corresponding to a subject, a descriptor, a background object, or an interactive entity may be extracted from the syntax information.

More specifically, according to the basic sentence type of Korean, given from the syntax information, the object, the predicate and the background object, the interaction object is distinguished. There may or may not be one or more candidate words that may correspond to a subject, object, descriptor, background entity, or interaction entity. The candidate words that may correspond to each of the subject, the object, the descriptor, the background entity, or the interaction entity are extracted. The subject may extract nouns that may be distinguished as subjects, the object may extract nouns that may be distinguished as object, and the predicates may extract verbs and adjectives as candidates. For example, "from that day the farmer was plowing the field" to "the day", "the farmer" could be extracted as the subject, "the field" could be extracted as the object, and "the plowing" could be extracted as the descriptor. have.

The interactive entity may extract a noun connected to “to” or “to” as a candidate, and the background entity may extract a noun connected to “from” or “to” as a candidate. For example, you can extract "grandmother", a noun that links "to" with "grandmother and grandfather together," as an interactive entity. You can also extract "mountain", a noun linked to "from" in "lumberjack does wood on mountain" as a background object.

In operation 420, a user's selection is input and a word corresponding to the subject, the object, the descriptor, the background object, or the interaction object is selected.

More specifically, the user selects candidate words of each subject, object, descriptor, background object, or interaction object extracted in step 410, and receives the user's selection to receive the subject, object, descriptor, background object, or Select the word corresponding to the interaction entity. If no information is stored about the words, or if the word corresponds to the subject, object, descriptor, background object, or interactive entity, the subject, object, descriptor, background is selected by the user. Select the word that corresponds to the entity or interaction entity. When the word is stored in the database in advance, a word corresponding to an object, a descriptor, a background object, or an interaction object may be selected without a user's selection. The method of using the database will be described in detail in the detailed description of FIG. 5. Alternatively, a word having a high probability of selection among candidate words may be recommended to the user according to the information of the database among several candidate words.

In operation 430, the type of the selected word may be classified to determine where the word corresponds to the animation element.

More specifically, when a word corresponding to a subject, a predicate, a background object, or an interactive object is selected from the syntax information, the type of each word is classified using the information, and an animation element is determined according to the type. To extract. Determine what kind of person the subject corresponds to, and determine what the action of the predicate is. For example, the types of nouns and verbs can be divided as shown in Table 1 below.

The types of words are first divided into nouns and verbs, and then the types of nouns and verbs are distinguished. The types of nouns and verbs are stored in a database, and the types of nouns and verbs may vary according to books. The verb types VT_NOTHING, VT_ADJ, and VT_IMPOSSIBLE are a special type that can't be represented by animation. VT_NOTHING indicates no action, which can be a type of verb such as "is". VT_ADJ is a predicate but not a verb. For example, the word "pretty" in "flower is pretty" is an adjective, not a verb. In this way, the VT_ADJ type is applied to the descriptor without movement. VT_IMPOSSIBLE refers to a verb that cannot be animated at the present time, which can be changed later. The background can be changed for each animation scene, which can be extracted in paragraph units.

The user's selection is accumulated and stored in a database, and the information of the database can be used to recommend to the user what the most likely words are among the candidate words. Such a recommendation may be provided for subjects and objects, and the predicates may be used as a basis for calculating the selection scores of subjects and objects. Words with high scores become prominent words, and conditional probabilities can be used to calculate scores. The equation for calculating the conditional probability is as shown in Equation 1 below.

[Equation 1]

P (Object | Predicate) = P (Object ∩ Predicate) / P (Predicate)-①

P (Subject | Predicate) = P (Subject ∩ Predicate) / P (Predicate)-②

P (Interactor | Subject) = P (Interactor ∩ Subject) / P (Subject)-③

Equations ① and ② indicate the probability that a certain object or subject will appear when a predicate comes out. For example, the probability that the object "ball" appears for "death" in "bear throws" is the number of times the combination (["ball" + "throw") appears / "throw"). Number of times], which applies equally to the subject. The reason that a subject and object are given priority when a predicate comes out is that the subject and the object are dependent on the predicate. For example, in the text "bear throws the ball" above, nouns such as "time" can come from the object of the predicate "throw", and it's hard to come up with "book" as a subject. That is, since the predicate has the effect of limiting the range of the subject and the object, it is possible to determine the priority of the words to be selected by using the expressions ① and ②.

Equation ③ is a method of prioritizing candidates that can be selected as interactive entities. This represents the conditional probabilities of the interacting entities that can be extracted correspondingly when the subject is preselected. If "grandmother" is selected as the subject, the probability that "grandfather" is recommended as an interactive entity is [number of "grandma" + + grandfather interaction) combinations / number of "grandmother" subjects. same. The selection word recommendation function using this method suggests a word selection direction for the text presented to the user using the above-mentioned equation.

In order to extract emotions from the animation elements, an adjective is used. Emotions applied to adjectives can be divided into negative and positive, and can be expressed in multiple stages according to the degree of emotion. Emotions of affirmation can also be expressed as high, medium, or low affirmations, as are negative. Or adjectives such as "deep" may have no emotion. The emotion according to the adjective may be used to select background music.

5 is a flowchart of a method of using an animation database to extract animation elements according to an embodiment of the present invention.

In operation 510, the meaning of each word is analyzed by comparing each word of the text with a word stored in an animation element database.

More specifically, in analyzing the meaning of each word of the input text, an animation element database may be used. The method of FIG. 3 may be used to extract animation elements, but an animation element database may be used to accurately and quickly extract animation elements. The animation element database stores meanings of words and animation elements corresponding to the words. Therefore, the meaning of each word can be analyzed by comparing each word of the text with a word stored in the animation element database. For example, in the text "bear throws a ball," there can be nominal candidates "bear" and "ball" that can be subject to the predicate "throw." Using the information in the animation element database, you can see which combination is used by comparing the case of "bear" + "throw" and "ball" + "throw" as a combination of subject + predicate. The combination will be extracted as "bear" + "throw" as a combination of subject + predicate. When there are a plurality of candidates, the candidates may be selected by the user and may be selected by receiving input from the user.

In operation 520, an animation element is extracted from the word using the analyzed word meaning.

The detailed description of this step corresponds to the detailed description of step 330 of FIG. 3 and replaces the detailed description of step 330 of FIG. 3.

In operation 530, the extracted animation element is stored in the animation element database together with the words of the text.

More specifically, the step of accumulating and storing information extracted from the animation element from each word. When the input text is a book, the meaning of the same word used in one book and the animation element corresponding to the meaning of the word may be the same. Therefore, in order to not repeat the step of extracting an animation element, by accumulating and storing syntax information about the word, when the stored word is input again, the animation element may be extracted directly from the syntax information. In addition, even if the book is not the same, if the type of the book or the like is constant like a fairy tale book, the words used may be the same or similar. Therefore, even when a new fairy tale book is input, the animation element can be extracted quickly and accurately using the information stored in the animation element database. The animation element database may be implemented as a separate storage space according to the book type.

6 is a flowchart illustrating a method for generating animation using an animation generation language according to an embodiment of the present invention.

In operation 610, an animation generation language is generated using the extracted animation element.

More specifically, in generating an animation using the animation element extracted in step 210, an animation generating language may be used. XML (extensible markup language) can be used as the animation generation language. XML is an extensibility markup language that allows you to define your own tags for the content of your document and make them available to other users. As long as the rules for XML parsing between users are met, the XML used by one user can be shared among other users. In addition, XML is a language designed to transmit structured documents on the Web. Each element of the document is independent of each other. Therefore, it is advantageous over other generation languages in terms of document compatibility, independence of contents, and ease of change of elements. In this step, you can define AniXML and use it as animation creation language. The AniXML can define objects, backgrounds, emotions, behaviors, and interactions. When a background is input, an image stored in the media database and corresponding to the background is set as the background. When an emotion is input, music stored in the media database and corresponding to the emotion is set. When an entity, action, or interaction is input, the subject of the action and the interacting entity are established, and the interaction between the subject's actions or interacting entities is established. AniXML can be generated per scene. For books, you can create one per paragraph. Several AniXML codes may be generated per book, and an animated scene may be produced as many as the AniXML codes.

In operation 620, the animation is generated using the generated animation generation language.

More specifically, the step of generating an animation by a predefined method using the animation generation language generated in step 610. An animation is automatically generated using at least one of an image, a voice, a music, or an effect sound of the media database set by the generation language.

Embodiments of the present invention may be implemented in the form of program instructions that can be executed on various computer means and recorded on a computer readable medium. The computer-readable medium may include program instructions, data files, data structures, and the like, alone or in combination. The program instructions recorded on the medium may be those specially designed and constructed for the present invention or may be available to those skilled in the art of computer software. Examples of computer-readable media include magnetic media such as hard disks, floppy disks and magnetic tape; optical media such as CD-ROMs and DVDs; magnetic media such as floppy disks; Magneto-optical media, and hardware devices specifically configured to store and execute program instructions such as ROM, RAM, flash memory, and the like. Examples of program instructions include machine language code such as those produced by a compiler, as well as high-level language code that can be executed by a computer using an interpreter or the like. The hardware devices described above may be configured to operate as one or more software modules to perform the operations of the present invention, and vice versa.

As described above, the present invention has been described by specific embodiments such as specific components and the like. For those skilled in the art, various modifications and variations are possible from these descriptions.

Accordingly, the spirit of the present invention should not be construed as being limited to the embodiments described, and all of the equivalents or equivalents of the claims, as well as the following claims, belong to the scope of the present invention .

110: animation element extraction unit
120: animation generator
130: animation element database
140: media database

Claims (17)

Extracting, by the animation element extraction unit, animation elements necessary for generating an animation from the text; And
Generating an animation by an animation generator by using the extracted animation element and an image corresponding to the animation element;
Extracting the animation element,
Performing morpheme analysis and part-of-speech attachment for each word constituting the text;
Generating syntax information for the text using the results of the morpheme analysis and the part-of-speech attachment; And
Analyzing the meaning of each word from the phrase information and extracting the animation element corresponding to each word;
Extracting the animation element corresponding to each word,
Extracting candidate words that may correspond to a subject, a predicate, a background entity, or an interactive entity from the syntax information;
Calculating priority of the candidate words using information of an animation element database;
Selecting a word corresponding to the subject, object, descriptor, background object, or interaction entity among the candidate words according to the priority; And
Determining whether the word corresponds to one of the animation elements by classifying the selected word type;
The animation element is at least one of an object, a background, an emotion, an action, an interaction. delete The method of claim 1,
Extracting the animation element corresponding to each word,
And selecting a word corresponding to the subject, object, descriptor, background object, or interaction object when receiving a selection of a user when the candidate words having the same priority are plural. The method of claim 1,
Extracting a candidate word that may correspond to a subject, a predicate, a background object, or an interactive entity from the syntax information,
The subject extracts nouns that can be distinguished as subjects,
The object is extracted as a noun that can be classified as an object as a candidate,
The predicate extracts a verb and an adjective as a candidate,
The interactive entity extracts a noun linked to "to" or "to" as a candidate,
And the background object extracts a noun connected to “to” or “to” as a candidate. delete The method of claim 1,
And the image is any one of an image extracted by analyzing a picture received with the text or an image stored in a media database. The method of claim 1,
The image may include:
In representing the behavior of the image corresponding to the object of the animation element,
And deforming the image around an axis of motion on the image to represent the object's behavior. The method of claim 1,
Generating the animation,
And generating the animation using at least one of voice, music, and sound effects corresponding to the animation element and stored in a media database. The method of claim 1,
Generating the animation,
Generating an animation generation language using the extracted animation element; And
Generating the animation using the generated animation generation language. The method of claim 9,
The animation generation language is an animation generation method, characterized in that the XML (extensible markup language). A computer-readable recording medium having recorded thereon a program for executing the method of any one of claims 1, 3 to 4, and 6 to 10 on a computer. An animation element extracting unit extracting an animation element required for generating an animation from text; And
An animation generator for generating an animation using the extracted animation element and the image corresponding to the animation element,
The animation element extraction unit,
Perform morpheme analysis and part-of-speech attachment for each word constituting the text, and generate syntax information for the text by using the result of the morpheme analysis and part-of-speech attachment, and the meaning of each word from the syntax information. Analyzing and extracting the animation element corresponding to each word,
The animation element corresponding to each word,
Extracting candidate words that may correspond to subjects, descriptors, background objects, or interaction objects from the phrase information, and calculating the priority of the candidate words using information of an animation element database; Selecting a word corresponding to the subject, object, descriptor, background object, or interaction object among the candidate words, and classifying the type of the selected word to determine where the word corresponds to the animation element,
The animation element is an animation generating device, characterized in that at least one or more of the object, background, emotion, behavior, interaction. delete delete 13. The method of claim 12,
The animation generator,
Generating an animation generating language using the extracted animation element, and generating the animation using the generated animation generating language. The method of claim 1,
Computing the priority of the candidate words,
When there are a plurality of candidate words that may correspond to a subject, the priorities of candidate words that may correspond to the subject may be determined by using a probability that the candidate words are selected as subjects for the predicates of the sentences formed by the candidate words. Calculate,
When there are a plurality of candidate words that may correspond to the target word, the priority of the candidate words that may correspond to the target word may be determined using the probability that the candidate words are selected as the target words for the predicates of the sentences formed by the candidate words. Yields,
When there are a plurality of candidate words that may correspond to the interacting entity, the candidate entity may correspond to the interacting entity using the probability that each candidate word is selected as the interacting entity for the subject of the sentence formed by the candidate word. Calculating the priority of candidate words. 13. The method of claim 12,
Priority of the candidate word is,
When there are a plurality of candidate words that may correspond to a subject, the priorities of candidate words that may correspond to the subject may be determined by using a probability that the candidate words are selected as subjects for the predicates of the sentences formed by the candidate words. Calculate,
When there are a plurality of candidate words that may correspond to the target word, the priority of the candidate words that may correspond to the target word may be determined using the probability that the candidate words are selected as the target words for the predicates of the sentences formed by the candidate words. Yields,
When there are a plurality of candidate words that may correspond to the interacting entity, the candidate entity may correspond to the interacting entity using the probability that each candidate word is selected as the interacting entity for the subject of the sentence formed by the candidate word. An animation generating device, characterized in that for calculating the priority of candidate words.

Images