KR102313203B1 - Artificial intelligence content creation system and method - Google Patents

Abstract

An artificial intelligence content creation system and method are provided. According to the system and the method, in order to generate one-person content which has been increasing recently, when users create dynamic content such as videos, animations, and the like without the help of a separate expert or tool, image processing and natural language processing are performed from images and texts using artificial intelligence, and dynamic content can be automatically generated by extracting and analyzing metadata.

Description

Artificial intelligence content creation system and method

The present invention relates to an artificial intelligence content creation system and method, and more particularly, image processing and natural language processing and metadata extraction using artificial intelligence without the help of a separate expert or tool when a user creates a creation and systems and methods for automatically generating content by analysis.

In the past, creation has been continued as the domain of experts for a long time, but it is expanding into the domain of fortune-telling individuals due to the development of information and communication technology, improvement of communication performance, and changes in content consumption culture.

Recently, the creative method is expanding to content with short but witty and connotative contents, such as vlogs using photos and videos, that is, a combination of video and blogs, webtoons, emoticons, and moving pictures. It is a trend that is being produced as content that can be shared and sympathized with.

In addition, with the spread of various social network service (SNS) platforms, unlike in the past, it has expanded to a one-person creative field, and an era has arrived where anyone can become a content producer through SNS platforms, such as game broadcasting and vlogs.

However, in order to create content, users have to go through all the processes from idea to sketch, image search, and editing. It should be entrusted to a professional who can use the .

In addition, even if a professional editor is used, it takes a lot of time to produce the final content, and especially in the case of a video, the amount of work is very high because image frames need to be edited and then stitched together.

Unexamined Patent Publication No. 10-2019-0035502

In order to solve the above problems, an object of the present invention is to extract metadata of images and texts by inputting specific images and texts in content production or by analyzing texts input on SNS platforms based on artificial intelligence. By creating dynamic content, especially videos and animations, from the extracted metadata, it automatically creates content from an idea level to high-quality content to support the creation of non-professionals and to produce high-quality content at low cost and time. there is

The problems to be solved of the present invention are not limited to those mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the following description.

As a means for solving the above technical problem, according to an embodiment of the present invention, an artificial intelligence content creation method includes the steps of: (a) inputting text and images related to content to be generated by a user; (b) analyzing the metadata of the text input by the user and the image data input by the user based on artificial intelligence and recording the text metadata and image metadata, respectively; (c) Any one or more of the background motion generation that changes the image input by the user into a background and the object, and changes to a dynamic image so that the background moves according to text metadata, and object motion generation that changes the object to a dynamic image so that the object moves according to the text metadata generating dynamic content through (d) the user calibrating the dynamic content to create a final work; and (e) distributing the final work generated by correction to a social platform or the like.

The text input by the user is the text input by the user through the input means of the user terminal or the text stored in the user terminal The text inputted through the input means of the user terminal on the social platform linked with the user terminal or interlocking with the user terminal It is text included in an image uploaded or downloaded by the user on a social platform.

The image input by the user is an image input by the user through an input means of the user terminal or uploaded or downloaded by the user on a social platform linked with the user terminal.

In step (b), text analysis based on artificial intelligence for the text input by the user, paragraph analysis with several compound sentences, extracting the most important sentence among paragraph analysis, and complex analysis of several sentences a text analysis step of analyzing and extracting action information, object information, background information, emotion of writing, and atmosphere of writing to generate metadata and recording it as text metadata in a database according to a predetermined format; and separating the background and foreground of the image based on artificial intelligence with respect to the image input by the user, and analyzing the image of a person's whole body, human face, animal, character, etc. in the image to obtain background and foreground information and feature points of objects and an image analysis step of extracting and generating metadata and recording it as image metadata in a database according to a predetermined format.

In the image analysis step, when there is no image input by the user, an image related to the text metadata is detected from among images stored in a user terminal or uploaded by the user on a social platform linked to the user terminal or images stored in a database Separates the background and foreground of the detected image, analyzes the whole body of a person, a human face, an animal, a character, etc. in the image to extract background and foreground information and feature points of the object to generate metadata in a predetermined format according to the image metadata in the database.

The step (c) includes: generating a background motion for changing the motion of the background among the image metadata to a dynamic image so as to move according to the general action of the background or the atmosphere indicated in the text metadata; and object motion generation for changing the body motion and facial expression of the object in the image metadata into a dynamic image to move according to the general action of the object or action information indicated in the text metadata; Dynamic content is created through one or more of these.

In step (d), adding, deleting, changing, or adding new elements of background motion and object motion using functions such as timeline and multi-layer provided by the service providing server for the generated dynamic content. Edits are made to create the final work.

In the step (e), the final work is stored in a form that is easy to upload to a social platform linked with the user terminal, and the user terminal is linked to the network so that the final work can be directly uploaded to the social platform or transmitted via a messenger. do.

Meanwhile, according to another embodiment of the present invention, in an artificial intelligence content creation system, a user receives text and images related to content to be generated, classifies data, transmits data to a service providing server, and receives the generated content , a user terminal that allows you to calibrate and upload to social platforms; Based on artificial intelligence, it analyzes the metadata of the text entered by the user and the image data entered by the user, extracts the text metadata and image metadata, respectively, and records them in the database, and uses the recorded metadata to create dynamic content. Artificial intelligence module for generating and transmitting to the user terminal; an authoring tool control unit that controls a tool used for generating the dynamic content and communicates with an artificial intelligence module; and a storage unit for storing a database such as images and animations required for creation of a creation, extracted metadata, and generated data.

The user terminal may include: a data receiving unit for receiving text and images related to content to be generated by the user from the user, and receiving the generated dynamic content from a service providing server; and a communication unit that transmits data to the service providing server, receives the generated content, and connects via wire or wireless to upload the final work created by correction by the user to the social platform.

The text input by the user is the text input by the user through the input means of the user terminal or the text stored in the user terminal The text inputted through the input means of the user terminal on the social platform linked with the user terminal or interlocking with the user terminal It is text included in an image uploaded or downloaded by the user on a social platform.

The image input by the user is an image input by the user through an input means of the user terminal or uploaded or downloaded by the user on a social platform linked with the user terminal.

The artificial intelligence module may include: a text metadata extractor that analyzes metadata of text data input by a user, extracts text metadata, and records it in the storage unit; an image metadata extracting unit that analyzes metadata of image data input by a user, extracts image metadata, and records it in the storage unit; a content automatic generation unit for generating dynamic content by using the text metadata and image metadata recorded in the storage unit and recording it in the storage unit; and a communication unit connecting the authoring tool control unit, the storage unit, and the user terminal to transmit and receive data.

The text metadata extraction unit is configured to analyze the text input by the user based on artificial intelligence, analyze a paragraph having multiple compound sentences, extract the most important sentence among paragraph analysis, and analyze several sentences in a complex manner. It analyzes and extracts action information, object information, background information, emotion of writing, and atmosphere of writing to generate metadata and records it as text metadata in the database according to a predetermined format.

The image metadata extraction unit separates the background and foreground of the image based on artificial intelligence with respect to the image input by the user, and analyzes the image of a person's whole body, human face, animal, character, etc. in the image to obtain the background and foreground Information and feature points of objects are extracted to generate metadata, and are recorded as image metadata in a database according to a predetermined format.

The image metadata extracting unit may include, when there is no image input by the user, an image related to the text metadata among images stored in a user terminal or uploaded by the user on a social platform linked to the user terminal or images stored in a database detection, separating the background and foreground of the detected image, and analyzing the image of a person's whole body, human face, animal, character, etc. in the image to extract background and foreground information and feature points of the object to generate metadata Write the image metadata to the database according to the form.

The automatic content generating unit may include: generating a background motion for changing a motion of a background among image metadata into a dynamic image so as to move according to a general action of the background or an atmosphere indicated in the text metadata; and object motion generation for changing the body motion and facial expression of the object in the image metadata into a dynamic image to move according to the general action of the object or action information indicated in the text metadata; Dynamic content is created through one or more of these.

According to the present invention,

In content creation, specific images and texts are input or text entered on the SNS platform is analyzed based on artificial intelligence to extract the metadata of images and texts, and dynamic content, especially videos and animations, through the extracted metadata By automatically creating content from an idea level to a high-quality content, it is possible to support the creation of non-professionals and to produce high-quality content with little cost and time.

In addition, by segmenting and diversifying genres in popular themes such as food, beauty, games, and cartoons, it can grow rapidly in line with users' desire for fresh and interesting content.

Effects of the present invention are not limited to those mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the following description.

1 is an exemplary diagram showing the main functions of an artificial intelligence content creation system and method according to an embodiment of the present invention;
2 is a flowchart of an artificial intelligence content creation method according to an embodiment of the present invention;
3 is a flowchart of an artificial intelligence content creation method according to an embodiment of the present invention;
4 is a flowchart of an artificial intelligence content creation method according to an embodiment of the present invention;
5 is a block diagram of an artificial intelligence content creation system according to an embodiment of the present invention.

The above objects, other objects, features and advantages of the present invention will be easily understood through the following preferred embodiments in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments described herein and may be embodied in other forms. Rather, the embodiments introduced herein are provided so that the disclosed content may be thorough and complete, and the spirit of the present invention may be sufficiently conveyed to those skilled in the art.

In this specification, when terms such as first, second, etc. are used to describe components, these components should not be limited by these terms. These terms are only used to distinguish one component from another. The embodiments described and illustrated herein also include complementary embodiments thereof.

Further, when it is stated that any element, component, device, or system includes a component consisting of a program or software, even if not explicitly stated, that element, component, device, or system means that the program or software is executed. Alternatively, it should be understood to include hardware (eg, memory, CPU, etc.) necessary for operation or other programs or software (eg, drivers necessary for operating an operating system or hardware, etc.).

Also, the terms used herein are for the purpose of describing the embodiments and are not intended to limit the present invention. As used herein, the singular also includes the plural unless specifically stated otherwise in the phrase. As used herein, the terms 'comprises' and/or 'comprising' do not exclude the presence or addition of one or more other components.

In addition, the '... wealth', '… Terms such as 'group' and 'module' mean a unit that processes at least one function or operation, which may be implemented as hardware or software or a combination of hardware and software. In addition, articles such as 'a', 'an' and 'the' in the context of describing the present invention are meant to include both the singular and the plural unless otherwise indicated herein or clearly contradicted by the context in the context of describing the present invention. can be used

In describing the specific embodiments below, various specific contents have been prepared to more specifically describe the invention and help understanding. However, a reader having enough knowledge in this field to understand the present invention may recognize that it can be used without these various specific details.

In some cases, it is mentioned in advance that in describing the invention, parts that are commonly known and not largely related to the invention are not described in order to avoid confusion for no reason in explaining the invention.

Hereinafter, specific technical contents to be practiced in the present invention will be described in detail with reference to the accompanying drawings.

1 is an exemplary diagram illustrating main functions of an artificial intelligence content creation system and method according to an embodiment of the present invention.

Referring to FIG. 1 , the artificial intelligence content creation method according to an embodiment of the present invention has an automatic authoring function based on AI by analyzing metadata of text and images based on AI.

Here, the analysis of metadata extracts metadata such as object/background information, action information, and emotion/atmosphere of text by extracting metadata based on text analysis. Through this, key points such as human body motion, facial expression, and animal motion are extracted.

In addition, the automatic authoring function can be largely divided into three types: automatic creation of dynamic content, automatic creation of sketch-based characters, and automatic conversion of emotion-based illustrations.

Here, the automatic creation of dynamic content can generate dynamic content with moving background by generating a sequence of background motion through neural net-based prediction by selecting a background from an image. By creating a sequence of object actions, you can create dynamic content in which objects move.

You can of course also create dynamic content in which both the background and the object move.

Here, in the automatic sketch-based character creation, when a user inputs a created sketch, a character model, a similar live-action image, etc. can be generated through neural net-based transformation.

Here, the emotion-based illustration automatic conversion can change the input image into a style suitable for the mood by converting the input image based on the neural net according to the mood of the text detected in the input text, and change the input image to the mood of the text detected in the input text. Accordingly, the image can be changed to a graphic effect suitable for the atmosphere through neural net-based effect selection and overlay.

Hereinafter, an embodiment of an artificial intelligence content creation system and method according to an embodiment of the present invention is shown.

2 is a flowchart of an artificial intelligence content creation method according to an embodiment of the present invention.

Referring to FIG. 2 , the artificial intelligence content creation method according to an embodiment of the present invention includes the steps of inputting text and images by the user (S100), analyzing and recording artificial intelligence-based metadata (S200), and creating a creation (S300), a step of correcting the user (S400), and a step of distributing the work (S400).

In the inputting step ( S100 ), the user inputs text and images related to content to be created for content creation.

Here, the text input by the user may be a text input by the user directly through the input means of the user terminal, text stored in the user terminal, or text input through the input means of the user terminal on a social platform such as SNS linked to the user terminal, , or it may be text included in an image uploaded or downloaded by a user on a social platform such as SNS linked with such a user terminal.

Here, the social platform may include a social network service (SNS) or a messenger.

Accordingly, such text may be a blog post written by a user, etc., a sentence used in a messenger conversation, or an article included in an e-book.

Here, the image input by the user may be any one or more of a static image including a main object and a background in order to generate dynamic content.

These images may also include images that the user directly input through the input means of the user terminal or stored in the user terminal, images uploaded or downloaded by the user on social platforms such as SNS linked with the user terminal.

Therefore, such an image may be a picture stored in the user terminal.

3 is a flowchart of an artificial intelligence content creation method according to an embodiment of the present invention.

FIG. 3 is a detailed description of the step S200 of analyzing and recording the metadata of FIG. 2 .

Here, in the step of analyzing and recording the metadata ( S200 ), based on artificial intelligence, the metadata of the text input by the user and the image data input by the user are analyzed and recorded as text metadata and image metadata, respectively.

Here, the step of analyzing and recording the metadata ( S200 ) may include a text analysis step ( S221 ) of analyzing the metadata of the text, and an image analysis step ( S222 ) of analyzing the metadata of the image.

Here, in the text analysis step (S221), analysis of the text input by the user based on artificial intelligence, paragraph analysis with multiple compound sentences, extracting the most important sentence among paragraph analysis, complex analysis of multiple sentences, etc. Through text analysis, metadata can be created by extracting action information, object information, background information, the emotion of the text, and the mood of the text.

Here, in the image analysis step (S222), the background and foreground of the image are separated based on artificial intelligence for the image input by the user, and the whole body of a person, a human face, an animal, a character, a sketch, an illustration, etc. in the image are imaged. Metadata can be generated by analyzing and extracting background and foreground information and feature points of objects.

Here, the feature point of the object may be a body motion, a pose or facial expression, and the shape of the eyes, nose, and mouth in the face.

However, in the metadata analysis and recording step (S200), it is first checked whether image data exists among the data input by the user in the input step (S100) (S210), and if there is an input image among the input data, the AI-based The text/image metadata analysis step (S220), that is, the text analysis step (S221) and the image analysis step (S222) proceeds as it is.

However, if there is no input image, the metadata of the input text is first analyzed based on artificial intelligence (S211), and an image matching the input text is searched from the database according to the object and background information derived from the analysis result. The AI-based text/image metadata analysis step (S220), that is, the text analysis step (S221) and the image analysis step (S222), is performed by storing (S212), searching and viewing the stored image as an image input by the user.

That is, the text analysis step (S221) proceeds regardless of the presence or absence of an image, and if there is no image, the image analysis step (S222) proceeds based on the image loaded from the database according to the result of the metadata analyzed in the text analysis step (S221) do.

Here, the database should collect images that do not have copyright issues.

Here, information such as objects, backgrounds, and atmospheres collected in the database may be tagged.

Here, in the text analysis step S221 and the image analysis step S222 , the generated metadata may be recorded in the database as text metadata and image metadata, respectively, according to a predetermined format.

4 is a flowchart of an artificial intelligence content creation method according to an embodiment of generating dynamic content of the present invention.

4 is a detailed description of the creative creation step (S300) of FIG.

4 illustrates a case in which dynamic content is generated among creations.

Referring to FIG. 4 , a video is generated based on the image according to the metadata analyzed in the metadata analysis step S200 ( S300 ).

First, the background and the object are separated based on the image metadata.

Here, the background may be separated according to foreground separation information and object/background information among image metadata.

Here, the motion of the separated background is changed to a dynamic image to move according to the general action of the background or action information shown in the text metadata, the emotion of the text, and the atmosphere of the text to generate a background motion video (background motion) (S311) .

Here, the object may be divided according to action information and feature point information among image metadata.

Here, the body motion, facial expression, etc. of the separated object are changed to a dynamic image to move according to the general action of the object or action information shown in text metadata, the emotion of the text, and the atmosphere of the text to generate an object motion video (object motion). generated (S312).

Both of these background motion generation (S311) and object motion generation (S312) may be performed or only one of them may be performed.

Through this generation process, dynamic content may be generated from the image (S313).

Accordingly, if you want to create a dynamic image suitable for an article analyzed in a messenger or e-book, etc., a motion video may be created through the object of the article, the background of the article, action information, etc. displayed in the text metadata.

For example, if you input “I skipped the rope today” in the messenger (S100), the action information of “jumping the rope” is extracted from the text and recorded as text metadata (S211), and the user's photo in the user's cell phone storage space is recorded (S211). It can be called (S212) to create a video of jumping rope motion (S313).

Also, for example, a picture (image) and content (text) of a children's book are extracted from a children's book (e-book) (S100), and object information is extracted from the text (S220) to generate a moving picture of the object in the picture ( S313) can be done.

5 is a flowchart of an artificial intelligence content creation method according to an embodiment of the present invention for generating sketch-based image content.

3, 4 and 5, after the creation step (S300), the user correction step (S400), the storage distribution step (S500), etc. for the creation created in the creation creation step (S300) will be performed. can

Here, in the user correction step (S400), detailed correction is possible through user editing after the creation is automatically created.

Here, users can calibrate dynamic content, sketch-based image content, illustration content, etc. using the timeline, multi-layer function, etc.

Here, the user can add, delete, and change motion elements of background motion and object motion, as well as separate images or text, using functions such as timeline and multi-layer provided by the service providing server for the generated dynamic content. The final work can be created by editing such as adding new elements, deleting backgrounds, or deleting all or part of an object.

Here, the user may confirm the generated dynamic content as a final work without correction.

Here, the work distribution step (S500) is linked so that the created work can be directly uploaded to SNS or transmitted through a messenger.

Here, you can save the work in a format that is easy to upload.

Here, storing the final work in a form that is easy to upload to a social platform interlocked with the user terminal, etc., and directly uploading the final work to the social platform or the like, or linking the user terminal to a network may include interworking with a network.

Here, the user can distribute and use the final work through SNS, etc.

5 is a block diagram of an artificial intelligence content creation system according to an embodiment of the present invention.

Referring to FIG. 5 , the artificial intelligence content creation system according to an embodiment of the present invention receives and classifies data, and a user terminal 10 that communicates with the service providing server 1000, a service providing an artificial intelligence-based automatic authoring function It consists of a providing server (1000).

Here, the user terminal 10 receives text and images related to content to be generated by the user, classifies the data, transmits the data to the service providing server 1000, and receives the generated content.

Here, the user terminal 10 includes a data receiving unit 11 that receives text and images related to content to be generated by the user from the user, and receives the generated dynamic content from a service providing server, and a data classification unit 12 that classifies data. ), a communication unit 13 that transmits data to the service providing server, receives the generated content, and connects via wire or wireless to upload the final work created by correction by the user to the social platform.

Here, the text input by the user may be a text input by the user directly through the input means of the user terminal, text stored in the user terminal, or text input through the input means of the user terminal on a social platform such as SNS linked to the user terminal, , or it may be text included in an image uploaded or downloaded by a user on a social platform such as SNS linked with such a user terminal.

Accordingly, such text may be a blog post written by a user, etc., a sentence used in a messenger conversation, or an article included in an e-book.

Here, the image entered by the user is a static image including the main object and background to create dynamic content, a sketch image created by the user to create sketch-based image content, and an artwork to transform the atmosphere to create the illustration content It may be any one or more of images.

These images may also include images that the user directly input through the input means of the user terminal or stored in the user terminal, images uploaded or downloaded by the user on social platforms such as SNS linked with the user terminal.

Therefore, such an image may be a picture stored in the user terminal.

Here, the service providing server 1000 includes an artificial intelligence module 100 , an authoring tool control unit 200 , and a storage unit 300 .

Here, the artificial intelligence module 100 analyzes the metadata of the text input by the user and the image data input by the user based on artificial intelligence, extracts the text metadata and the image metadata, respectively, and records them in the database, and records By using the meta-data, one or more creations among dynamic content, sketch-based image content, and illustration content are created and transmitted to the user terminal.

Here, the artificial intelligence module 100 includes a text metadata extraction unit 110 , an image metadata extraction unit 120 , an automatic content generation unit 130 , and a communication unit 140 .

Here, the text metadata extraction unit 110 extracts the most important sentence among the sentence analysis, the paragraph analysis with several compound sentences, and the paragraph analysis based on the artificial intelligence for the text input by the user, and analyzes several sentences in a complex manner. By extracting action information, object information, background information, the emotion of the text, and the atmosphere of the text through text analysis, the text metadata may be generated and stored in the metadata 320 of the storage unit 300 .

Here, the image metadata extraction unit 120 separates the background and foreground of the image based on artificial intelligence with respect to the image input by the user, and the whole body of a person in the image, a human face, an animal, a character, a sketch, an illustration, etc. can be image-analyzed to extract background and foreground information and feature points of an object to generate image metadata and store it in the metadata 320 of the storage unit 300 .

Here, the feature point of the object may be a body motion, a pose or facial expression, and the shape of the eyes, nose, and mouth in the face.

However, the image metadata extraction unit 120 first checks whether image data exists among the data input by the user, and when there is an input image among the input data, artificial intelligence-based image metadata extraction proceeds as it is.

However, when there is no input image, the user searches for and stores the image matching the text inputted from the database according to the object and background information derived from the text metadata extraction unit 110 (S212), and the user selects the searched and stored image. Based on the input image, artificial intelligence-based image metadata analysis is performed.

That is, the text metadata extractor 110 performs text-z analysis regardless of the presence or absence of an image, and the image metadata extractor 120 determines the result of the metadata analyzed by the text metadata extractor 110 if there is no image. It proceeds based on the image loaded from the database.

Here, the database should collect images that do not have copyright issues.

Here, information such as objects, backgrounds, and atmospheres collected in the database may be tagged.

Here, the metadata generated by the text metadata extraction unit 110 and the image metadata extraction unit 120 may be recorded in the database as text metadata and image metadata, respectively, according to a predetermined format.

In the embodiment of the present invention, it is mainly shown for automatic creation of dynamic content, but here, the automatic content generation unit 130 includes the dynamic content automatic generation unit 131, the sketch-based image automatic generation unit 132, and the emotion-based illustration automatic conversion unit. (133) may be included.

Here, when the image input by the user is a static image including a main object and a background, the dynamic content automatic creation unit 131 generates dynamic content according to the image and text metadata.

Here, the dynamic content automatic creation unit 131 first separates the background and the object based on image metadata.

Here, the background may be separated according to foreground separation information and object/background information among image metadata.

Here, the motion of the separated background is changed to a dynamic image to move according to the general action of the background or action information shown in text metadata, the emotion of the text, and the atmosphere of the text to generate a background motion video (background motion).

Here, the object may be divided according to action information and feature point information among image metadata.

Here, the body motion, facial expression, etc. of the separated object are changed to a dynamic image to move according to the general action of the object or action information shown in the text metadata, the emotion of the text, and the atmosphere of the text to generate an object motion video (object motion). create

Both the background motion generation and the object motion generation may be performed or only one of the two may be performed.

Through this creation process, dynamic content can be created from the image.

Here, when the image input by the user is a sketch image created by the user, the sketch-based image automatic generating unit 132 generates sketch-based image content according to image and text metadata.

The sketch-based image automatic generating unit 132 may receive a user's sketch and fuse it with object information of text and change it into an image or a character.

Here, by referring to the body motion of the sketch, the facial expression, the shape of the eyes, nose, and mouth among the image metadata, the style of changing the image according to the style of painting selected by the user or the atmosphere, style, etc. shown in the text metadata may be performed.

Also, a color change in which a color of a part of a sketch among image metadata is changed according to a color selected by a user or content indicated in text metadata may be performed.

Both the style change and the color change may be performed or only one of them may be performed.

Through this creation process, sketch-based image content can be created from the sketch image.

The sketch-based image content created here is an object image related to a sketch, and the image may be similar to a real-life image or may be a cartoonized image, etc. can

In addition, the desired figure can be selected by the user or automatically selected and reflected from the atmosphere of the text detected from the text, and it is possible to make more accurate changes by inferring from the text which object the sketch is about.

Here, the emotion-based illustration automatic conversion unit 133 generates changed illustration content according to the image and text metadata when the image input by the user is an illustration image to change the atmosphere.

Through this, if you want to add an image suitable for a blog post, etc., you can change the style of the text or add graphic effects to the image input by the user or the image loaded from the database according to the emotion of the text among the text metadata.

When the image input by the user is an illustration image to convert the atmosphere, the emotion-based illustration automatic conversion unit 133 selects the background and object of the illustration image from among the image metadata. A style change of changing the image may be made according to the like.

Also, a graphic overlay that changes the background of the illustration image and the graphic effect of the object among the image metadata according to the graphic effect selected by the user or the atmosphere, style of painting, and foreground separation information shown in the text metadata may be performed.

Both of these style changes and graphic overlays may be performed or only one of them may be performed.

Through this generation process, the changed illustration content can be generated from the illustration image.

In this way, by detecting the atmosphere of the writing from the text, the emotion-based illustration is automatically converted to change the style of the image according to the atmosphere.

In this case, the image may be an image that is retrieved by searching the database from an image input by the user or background/object information detected from text.

The dynamic graphic effects of the artwork can be emotionally selected and overlaid on the entire image or on the main object.

In addition, in the automatic content generation unit 130, the automatic creation of dynamic content, the automatic creation of sketch-based images, and the automatic conversion of emotion-based illustrations may be sequentially connected to each other.

That is, after creating an image with a sketch, a moving image can be created by applying motion to the object.

You can also change the style of an illustration and create a video that gives a flowing effect to the background of the illustration.

Therefore, by inputting the sketch-based image content generated by the sketch-based image automatic generation unit and the illustration content generated by the emotion-based illustration automatic conversion unit to the automatic dynamic content generation unit, dynamic content can be generated according to the image and text metadata. have.

Here, the image or image/text metadata is extracted again for the sketch-based image content generated by the sketch-based image automatic generation unit and the illustration content generated by the emotion-based illustration automatic conversion unit to generate dynamic content using the metadata. may be

Here, the communication unit 140 is connected to the authoring tool control unit 200 , the storage unit 300 , the user terminal 10 , and the like in a wired/wireless manner so as to transmit and receive data.

Referring to FIG. 5 , the authoring tool control unit 200 may include a content authoring tool basic function control unit 210 , an artificial intelligence module interworking communication unit 220 , and the like.

Here, the content authoring tool basic function control unit 210 may control the basic functions of the tool used for content creation by the automatic content creation unit 130 .

Here, the artificial intelligence module interworking communication unit 220 is connected to a network or the like so as to interwork with the artificial intelligence module 100 to communicate wired/wireless with the outside.

Referring to FIG. 5 , the storage unit 300 includes a database 310 , metadata 320 , and generated data 330 .

Here, the database 310 includes data generated in advance in order to search for an image matching the input text when an image needs to be generated using only text, such as an image or animation.

The metadata 320 includes text metadata extracted by the text metadata extraction unit 110 and image metadata extracted by the image metadata extraction unit 120 .

The generated data 330 includes dynamic content, sketch-based image content, and illustration content generated by the content automatic generator 130 .

Such generated data 330 may be distributed through SNS or messenger after user's correction.

Here, the communication unit of the artificial intelligence module 100 or the artificial intelligence module interworking communication unit 220 transmits the generated data 330 to the user terminal, and the user uses the timeline, multi-layer function, etc. through the input means of the user terminal. to correct dynamic content, sketch-based image content, and illustration content.

Here, the user can add, delete, and change motion elements of background motion and object motion, as well as separate images or text, using functions such as timeline and multi-layer provided by the service providing server for the generated dynamic content. The final work can be created by editing such as adding new elements, deleting backgrounds, or deleting all or part of an object.

Here, the user may confirm the generated dynamic content as a final work without correction.

In addition, the user can directly upload the work created through the communication unit 13 of the user terminal 10 to SNS or the like or link it so that it can be transmitted through a messenger.

Here, you can save the work in a format that is easy to upload.

The final work corrected by the user is transmitted to the service providing server 1000, and the service providing server 1000 stores the final work in a form that is easy to upload to a social platform interlocked with the user terminal, and stores the final work in the social platform. This may include linking the user terminal to the network so that it can be directly uploaded to a platform or the like or transmitted through a messenger.

In this case, the user may upload the final work through the user terminal 10 to a social platform or the like through the communication unit 13 .

Here, the service providing server 1000 may be installed and operated on a user's PC or the like or may be operated on a web-based basis.

In addition, the service providing server 1000 may operate in the user's PC or mobile.

Above, the embodiment of the present invention is not implemented only through the above-described apparatus and/or operation method, but through a program for realizing a function corresponding to the configuration of the embodiment of the present invention, a recording medium in which the program is recorded, etc. It may be implemented, and such an implementation can be easily implemented by an expert in the technical field to which the present invention pertains from the description of the above-described embodiments. Although the embodiments of the present invention have been described in detail above, the scope of the present invention is not limited thereto, and various modifications and improved forms of the present invention are also provided by those skilled in the art using the basic concept of the present invention as defined in the following claims. is within the scope of the right.

10: user terminal
11: data receiving unit
12: data classification unit
13: communication department
100: artificial intelligence module
110: text metadata extraction unit
120: image metadata extraction unit
130: content automatic creation unit
131: dynamic content automatic creation unit
132: sketch-based image automatic generation unit
133: emotion-based illustration automatic conversion unit
140: communication department
200: authoring tool control unit
210: content creation tool basic function control unit
220: artificial intelligence module interlocking communication unit
300: storage
310: database
320: metadata
330: generated data
1000: service providing server

Claims (17)

In the artificial intelligence content creation method,
(a) the user inputting text and images related to the content to be created;
(b) analyzing the metadata of the text input by the user and the image data input by the user based on artificial intelligence and recording the text metadata and image metadata, respectively;
(c) Any one or more of the background motion generation that changes the image input by the user into a background and the object, and changes to a dynamic image so that the background moves according to text metadata, and object motion generation that changes the object to a dynamic image so that the object moves according to the text metadata generating dynamic content through
(d) the user calibrating the dynamic content to create a final work; and
(e) distributing the final work created by correcting it to a social platform, etc.;
including,
Step (b) is,
For the text input by the user, based on artificial intelligence, text analysis such as sentence analysis, paragraph analysis with multiple compound sentences, extraction of the most important sentence among paragraph analysis, and complex analysis of multiple sentences to provide action information and object information , a text analysis step of extracting background information, the emotion of the text, and the atmosphere of the text to generate metadata and recording it as text metadata in a database according to a predetermined format; and
With respect to the image input by the user, the background and foreground of the image are separated based on artificial intelligence, and the background and foreground information and feature points of the object are extracted by analyzing the image of a person's whole body, a human face, an animal, a character, etc. in the image. an image analysis step of generating metadata and recording it as image metadata in a database according to a predetermined format;
Content creation method comprising a. According to claim 1,
The text entered by the user is
Text input by the user through input means of the user terminal or stored in the user terminal Text input by the user through the input means of the user terminal on a social platform linked to the user terminal or uploaded by the user on a social platform linked to the user terminal Or a content creation method, characterized in that the text included in a downloaded image or the like. According to claim 1,
The image entered by the user is
The content creation method, characterized in that the image is an image input by a user through an input means of a user terminal or uploaded or downloaded by the user on a social platform linked with the user terminal. delete According to claim 1,
The image analysis step is,
When there is no image input by the user, an image related to the text metadata is detected from among images stored in a user terminal or uploaded by the user on a social platform linked to the user terminal or images stored in a database, and Separating the background and foreground, analyzing the image of a person's whole body, human face, animal, character, etc. in the image, extracting background and foreground information and object feature points to create metadata A content creation method characterized in that it is recorded as data. According to claim 1,
Step (c) is,
generating a background motion for changing the motion of the background among the image metadata to a dynamic image so as to move according to the general action of the background or the atmosphere indicated in the text metadata; and
object motion generation for changing an object's body motion and facial expression among image metadata into a dynamic image to move according to the general action of the object or action information indicated in text metadata;
A content creation method, characterized in that the dynamic content is generated through any one or more. According to claim 1,
Step (d) is,
For the created dynamic content, the final work is created by adding, deleting, changing, or adding new elements to the background motion and object motion using the functions such as timeline and multi-layer provided by the service providing server. to do
A content creation method characterized by According to claim 1,
Step (e) is,
Storing the final work in a form that is easy to upload to a social platform linked with a user terminal, etc., and linking the user terminal to a network so that the final work can be directly uploaded to the social platform or transmitted via a messenger
A content creation method characterized by In the artificial intelligence content creation system,
a user terminal for receiving text and images related to content to be generated by a user, classifying data, transmitting data to a service providing server, receiving, correcting, and uploading the generated content to a social platform;
Based on artificial intelligence, it analyzes the metadata of the text entered by the user and the image data entered by the user, extracts the text metadata and image metadata, respectively, and records them in the database, and uses the recorded metadata to create dynamic content. Artificial intelligence module for generating and transmitting to the user terminal;
an authoring tool control unit that controls a tool used for generating the dynamic content and communicates with an artificial intelligence module; and
a storage unit for storing a database such as images and animations required for creation of a creation, extracted metadata, and generated data;
including,
The artificial intelligence module is
a text metadata extracting unit that analyzes metadata of text data input by a user, extracts text metadata, and records it in the storage unit;
an image metadata extracting unit that analyzes metadata of image data input by a user, extracts image metadata, and records it in the storage unit;
a content automatic generation unit for generating dynamic content by using the text metadata and image metadata recorded in the storage unit and recording it in the storage unit; and
a communication unit connecting the authoring tool control unit, the storage unit, and the user terminal to transmit and receive data;
Including, the text metadata extracting unit,
For the text input by the user, based on artificial intelligence, text analysis such as sentence analysis, paragraph analysis with multiple compound sentences, extraction of the most important sentence among paragraph analysis, and complex analysis of multiple sentences to provide action information and object information , background information, the emotion of the writing, and the atmosphere of the writing to generate metadata, and record the metadata in the database according to a predetermined format as text metadata. 10. The method of claim 9,
The user terminal is
a data receiving unit that receives text and images related to content to be generated by the user from the user and receives the generated dynamic content from a service providing server; and
a communication unit that transmits data to a service providing server, receives the generated content, and connects via wire or wireless to upload the final work created by correction by the user to the social platform;
Content creation system comprising a. 10. The method of claim 9,
The text entered by the user is
Text input by the user through input means of the user terminal or stored in the user terminal Text input by the user through the input means of the user terminal on a social platform linked to the user terminal or uploaded by the user on a social platform linked to the user terminal or text included in a downloaded image or the like. 10. The method of claim 9,
The image entered by the user is
The content creation system, characterized in that the image is an image input by a user through an input means of a user terminal or uploaded or downloaded by the user on a social platform linked with the user terminal. delete delete 10. The method of claim 9,
The image metadata extraction unit,
With respect to the image input by the user, the background and foreground of the image are separated based on artificial intelligence, and the background and foreground information and feature points of the object are extracted by analyzing the image of a person's whole body, a human face, an animal, a character, etc. in the image. to create metadata and record it as image metadata in the database according to a predetermined form
A content creation system characterized by 10. The method of claim 9,
The image metadata extraction unit,
When there is no image input by the user, an image related to the text metadata is detected from among images stored in a user terminal or uploaded by the user on a social platform linked to the user terminal or images stored in a database, and Separating the background and foreground, analyzing the image of a person's whole body, human face, animal, character, etc. in the image, extracting background and foreground information and object feature points to create metadata to record data
A content creation system characterized by 10. The method of claim 9,
The content automatic creation unit,
generating a background motion for changing the motion of the background among the image metadata to a dynamic image so as to move according to the general action of the background or the atmosphere indicated in the text metadata; and
object motion generation for changing an object's body motion and facial expression among image metadata into a dynamic image to move according to the general action of the object or action information indicated in text metadata;
A content creation system, characterized in that the dynamic content is generated through one or more of the following.

Images