KR20200009599A - A method, server and program for apply artist's style to image and provide it - Google Patents

Abstract

The present invention relates to a method for providing an image, a server, and a program thereof, and more specifically, to a method for providing after applying the brush of an artist to an image, a server, and a program thereof. To this end, the present method comprises: a step that a server receives at least one image from an artist client; a tracing image providing step that an individual tracing model generates a tracing image after applying the brush of the artist to the image before providing the same to the artist client, and, in the case that it is the first time that the image is received from the artist, a tracing model of the artist group to which the artist belongs applies the brush of the artist group to the image in order to generate the tracing image; a retouched image receiving step that a retouched image is received from the artist client, wherein the retouched image is produced after the artist retouched the tracing image; and a learning model construction step that the retouched image is stored in the individual tracing model.

Description

A METHOD, SERVER AND PROGRAM FOR APPLY ARTIST'S STYLE TO IMAGE AND PROVIDE IT}

The present invention relates to a method, a server, and a program for providing an image, and more particularly, to a method, a server, and a program for applying an artist's style to an image.

In general, people with occupations such as cartoonists and webtoon writers spend a lot of time drawing. This is because cartoons, webtoons, and animations take a lot of time to draw.

In recent years, various IT technologies have been developed, programs, applications, and the like have been developed, but most of the people who have such jobs still spend a lot of time drawing in a classic way.

Techniques for adjusting the color, brightness, etc. of a specific photo using applications and programs have been disclosed. However, this is a simple way of correcting an image for the general public, and thus there is a problem in that a professional style such as an artist cannot be applied at all. .

The background art of the present invention is disclosed in the Republic of Korea Patent Publication No. 10-1189003, etc., but the fundamental solution to the above-mentioned problem is not presented.

The present invention for solving the problems described above can provide a method, a server and a program for applying the artist's style to the image to create and provide a tracing image by applying the artist's style to the image received from the artist. have.

The present invention also provides a method of applying the artist's painting style to an image which allows the artist to learn the painting style by constructing a learning model by storing the retouching image in the tracing model when the artist sends the painting image to the server. And programs.

In addition, the present invention may provide a method, a server, and a program for analyzing the sample image to derive the style of the artist by analyzing the sample image and applying the style of the artist to the image set to the corresponding group of artists. have.

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

According to an aspect of the present invention, there is provided a method of applying an artist's painting style to an image according to an embodiment of the present invention, which is a computer-implemented method, in which a server receives one or more images from an author's client. When the tracing model applies the artist's style to the image to generate a tracing image and provides it to the artist's client, but the first time that the image is received from the artist, the tracing model of the artist to which the artist belongs is applied to the image of the artist's group on the image. Applying a tracing image, receiving a retouching image from the writer client, wherein the retouching image is a retouching of the tracing image by the writer, and a retouching image in the personal tracing model. Learning model It includes the build phase.

The method may further include receiving, by the server, at least one sample image from the writer client, analyzing the sample image to derive the artist's style, and setting the corresponding artist as the artist group corresponding to the style.

In addition, analyzing the sample image to derive the artists of the style similar to the artist, and analyzing the similarity between the artists.

The learning model building step may include extracting a feature vector from the retouching image, and training the artificial intelligence by storing the extracted feature vector in a personal tracing model.

The providing may include: extracting a feature vector from the received image by the tracing model, and generating a tracing image by applying the artist's style to the feature vector of the image, and providing the tracing image to the writer client. Steps.

In addition, the learning model building step includes copying the artist group tracing model and generating a personal tracing model used only for the corresponding artist.

In addition, the step of receiving an image or a tracing image corresponding to the style of the artist group belonging to each artist group tracing model, and the artist group learning model construction step, the artist group tracing model extracts and stores a feature vector from the image or tracing image Include.

In addition, the server includes a writer group determination module for deriving a writer group by analyzing an image through a learning model built on the writer group tracing model.

The server for applying the artist's style to the image according to an embodiment of the present invention for solving the above-described problem, the receiving unit for receiving an image or a retouching image from the artist client, applying the artist's style to the image tracing A personal tracing model for generating an image and providing the image to a writer client, and a writer group tracing model for generating a tracing image by applying a style of the artist group to which the artist belongs to the image. And storing the retouching image in the personal tracing model to train the personal tracing model.

In addition, a computer readable recording medium for recording another method for implementing the present invention, another system, and a computer program for executing the method may be further provided.

The program for applying the artist's painting style to the image according to another embodiment of the present invention executes a method of applying the artist's painting style to the above-mentioned image in combination with a computer, which is hardware, and is stored in a medium.

According to the present invention as described above, by applying the artist's painting style to the image received from the artist to create and provide a tracing image, the artist has the effect of reducing the work time by proceeding by retouching based on the tracing image. .

In addition, according to the present invention, when the artist retouches the tracing image and transmits it to the server, the artist stores the retouching image in the tracing model to build a learning model to learn the artist's style of painting. The effect is to be able to provide a tracing image similar to.

In addition, according to the present invention, when receiving the sample image from the artist, by analyzing the sample image to derive the artist's style, and set to the corresponding group of artists, there is an effect of forming a group of artists with similar styles.

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

1 is a schematic diagram of a method of applying an artist's painting style to an image according to an embodiment of the present invention.
2 and 3 are a flow chart of a method of applying the artist's painting style to the image according to an embodiment of the present invention.
4 is an exemplary diagram illustrating setting a writer as a specific artist group by analyzing a sample image according to an embodiment of the present invention.
5 is an exemplary diagram illustrating an image input by a writer into a server according to an embodiment of the present invention.
FIG. 6 illustrates an example in which a tracing image is provided by applying the artist's painting style to FIG. 5, and the artist retouches the tracing image.
Figure 7 is a block diagram of a server providing by applying the artist's painting style to the image according to an embodiment of the present invention.

Advantages and features of the present invention, and methods for achieving them will be apparent with reference to the embodiments described below in detail in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below but may be embodied in various different forms, only the present embodiments are intended to complete the disclosure of the present invention, and those skilled in the art It is provided to fully inform the skilled person of the scope of the invention, which is defined only by the scope of the claims.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. In this specification, the singular also includes the plural unless specifically stated otherwise in the phrase. As used herein, “comprises” and / or “comprising” does not exclude the presence or addition of one or more other components in addition to the mentioned components. Like reference numerals refer to like elements throughout, and "and / or" includes each and all combinations of one or more of the mentioned components. Although "first", "second", etc. are used to describe various components, these components are of course not limited by these terms. These terms are only used to distinguish one component from another. Therefore, of course, the first component mentioned below may be the second component within the technical spirit of the present invention.

Unless otherwise defined, all terms used in the present specification (including technical and scientific terms) may be used in a sense that can be commonly understood by those skilled in the art. In addition, terms that are defined in a commonly used dictionary are not ideally or excessively interpreted unless they are specifically defined clearly.

Hereinafter, with reference to the accompanying drawings will be described an embodiment of the present invention;

Prior to the description, the meaning of terms used in the present specification will be briefly described. However, it is to be noted that the description of the terms is intended to help the understanding of the present specification, and therefore, the terms are not used to limit the technical spirit of the present invention unless explicitly stated as limiting the present invention.

Painting Style: Means a particular way, style, tendency, or characteristic of drawing. Similar styles include painting style, drawing style, and painting.

Background Image: Refers to an area image that becomes a still background on a photo or image.

Tracing: A method of copying a non-moving area of a photo or image as it is and image it for manga.

Retouching: Retouching a picture or photo. In the embodiment of the present invention, it means that the user (author) directly corrects the tracing image provided from the server.

1 is a schematic diagram of a method of applying an artist's painting style to an image according to an embodiment of the present invention.

Referring to Figure 1 to briefly describe a method of applying the artist's painting style to the image according to an embodiment of the present invention.

The artist transmits an image to be converted into a tracing image to the server 100. At this time, the image transmitted by the writer client 300 may be applied to a variety of images, such as a picture actually taken, a picture downloaded from the Internet.

Preferably, since all of the characters, stories, and the like that the artist intends to express are related to the artist's creativity, the background image that the artist intends to use for comics and webtoons may be mainly applied.

In addition, the server 100 applies the artist's painting style to the image received from the writer client 300, and the artist group tracing model 170 generates a tracing image and provides it to the writer client 300.

The artist retouches the tracing image provided from the server 100.

More specifically, since the artist is provided with a background image to which his style of painting is applied, the artist can modify the background by expressing his creativity using the tracing image as a background and draw a person, an object, a story, and the like.

The writer transmits the retouching image to the server 100 when the retouching is completed, and the server 100 stores the personal tracing model 190 to build a learning model.

By repeating the above operation several times, as the number of learning of the personal tracing model 190 increases, the similarity between the artist's actual painting style and the tracing image increases, and the artist's satisfaction increases and the artist's drawing time is shortened. Will be exercised.

2 and 3 are flowcharts of a method of applying an artist's painting style to an image according to an embodiment of the present invention, and FIG. 4 is a diagram illustrating a method of setting a artist as a specific artist group by analyzing a sample image according to an embodiment of the present invention. It is an exemplary figure which illustrates that.

Referring to Figure 2 will be described with respect to the flow of a method of applying the artist's painting style to the image according to an embodiment of the present invention.

First, server 100 receives one or more images from writer client 300. (Step S510)

The artist accesses the server 100 through a device such as a computer, a tablet, a smartphone, a PDA, a drawing board, a handheld device, and transmits an image to be converted into a tracing image to the server 100.

In this case, the character, object, etc., which the artist draws on the cartoon, the webtoon, etc. should be exhibited by the artist's creativity for the story. Therefore, an image mainly used as a background image is preferably input.

In operation S510, the server 100 generates a tracing image by applying the artist's painting style to the received image, and provides the image to the artist client 300. (Step S520)

More specifically, the server 100 generates an image by applying the artist's style to the image received from the writer client 300 using the personal tracing model 190, and provides the image to the writer client 300.

In this case, when receiving the image from the artist for the first time, the writer group tracing model to which the writer belongs to generates a tracing image by applying the artist's style to the image and provides the writer client 300.

This is because, when the artist first subscribes to the corresponding service and transmits the image to the server 100, since the artist does not learn about the artist's painting style in the tracing model, the artist creates a tracing image by applying the painting style of the artist group to which the artist belongs. Means that. In addition, after that, the artist's painting style generates a tracing image through the learned personal tracing model 190.

As another example, when the server 100 receives an image from the writer client 300, the server 100 checks the personal tracing model 190 set for the corresponding artist, and the personal tracing model 190 set for the writer is If it is not there, the artist may determine to use the service for the first time, and may further include setting to use the artist group tracing model 170.

In this case, the providing of the tracing image (S520) may include extracting a feature vector from the image received from the artist client 300 by the artist group tracing model 170 or the personal tracing model 190, and the artist group tracing model ( 170 or the personal tracing model 190 generates a tracing image by applying the artist's style to the feature vector of the image received from the writer client 300, and provides it to the writer client 300.

Accordingly, the tracing model extracts a feature vector to which the artist's style is applied from the image received from the artist client 300, and generates a tracing image by applying the artist's style to the extracted feature vector.

In operation S520, the server 100 receives the retouching image from the writer client 300. (Step S530)

The retouching image means that the artist retouches the tracing image received from the server 100.

More specifically, the artist completes the image by retouching the tracing image provided from the server 100 for comics and webtoons, and then transmits the completed retouching image to the server 100.

Accordingly, the server 100 receives the completed retouching image retouched by the artist.

In step S530, the server 100 builds a learning model through the retouching image. (Step S540)

More specifically, the server 100 stores the retouching image in the personal tracing model 190 to build a learning model.

Therefore, when the server 100 generates the tracing image of the artist for the first time, since the tracing image is applied by the artist group tracing model 170 to generate the tracing image, the rough image of the artist is applied to generate the tracing image. do.

However, after that, since the individual tracing model 190 is given to each individual, the learning becomes more and more similar to the artist's drawing style, and the effect of shortening the artist's drawing time is increased. Will be used.

At this time, the training model building step (S540 step), the step of extracting a feature vector from the retouching image, and the training model building step of training the artificial intelligence by storing the extracted feature vector in the personal tracing model 190.

Basically, the retouching image received from the writer client 300 is a writer retouching the tracing image. This is because the artist draws a character, an object, and the like on the background image to express creativity, so that the server 100 extracts and learns only the feature vector without having to learn the artist's style in the entire retouching image.

As a result, the server 100 may be prevented from being overloaded, and the personal tracing model 190 may be efficiently learned.

The learning model building step (S540) may further include generating a personal tracing model 190 which is used only for the corresponding artist by copying the artist group tracing model 170.

As described in the step of providing a tracing image (step S520), when the first receiving an image from the artist, the writer group tracing model 170 to which the writer belongs to generates a tracing image by applying the artist's style to the image. Subsequently, in order to use the personal tracing model 190 applied only to the corresponding artist, copying the artist group tracing model 170 means generating the personal tracing model 190 dedicated to the corresponding artist.

Therefore, the personal tracing model 190 generated by the server 100 is applied only to the corresponding artist, and builds a learning model by learning through a retouching image received from the corresponding artist.

FIG. 3 is a diagram for describing steps preceding before the steps described in FIG. 2 proceed.

As described above, in the step of providing a tracing image (step S520), when the artist has set the personal tracing model 190, the personal tracing model 190 set to the artist generates a tracing image by applying the artist's style to the image. , Author to client 300,

If the personal tracing model 190 is not set to the artist and the server 100 receives the image from the artist for the first time, the artist group belonging to the artist group tracing model 170 applies the artist's style to the image and traces the image. Generated and provided to the writer client (300).

Therefore, an embodiment of setting an artist group when the first image is received from the artist will be described with reference to FIG. 3.

First, server 100 receives one or more sample images from writer client 300. (Step S430)

Preferably, this means that the artist transmits the background image to be converted into the tracing image to the server 100.

At this time, when the server 100 receives the sample image to the writer client 300, the server 100 checks whether the personal tracing model 190 is set to the writer.

When the personal tracing model 190 is set for the writer, the personal tracing model 190 is set without using the writer group tracing model 170, and the subsequent steps are performed using the personal tracing model 190 set for the writer.

In step S430, the server 100 sets a group of writers. (Step S450)

In detail, the server 100 analyzes the sample image to derive the artist's painting style, and sets the artist as the artist group corresponding to the painting style.

To this end, one or more artist groups are set and stored in the server 100 as shown in FIG. 4.

Each group of artists is stored for a particular painting style. For example, the first artist group may store a heavy picture body, the second artist group a bling bling picture, and the like.

In addition, each artist group is set the author group tracing model 170, each of the artist group tracing model 170 is stored the learning data for the style of the corresponding artist group.

Next, the step S450 may include analyzing the sample image to derive the artists of the painting style similar to the artist, and analyzing the painting style similarity among the artists.

In this case, the similarity may include a drawing method, a style, a drawing style, a drawing style, and the server may analyze the sample image to derive the drawing style, style, drawing style, or drawing style. This may be implemented through pre-stored algorithms, and may further include a wind analysis module for implementing such a function.

Through such a wind analysis module, it is possible to determine the drawing method, style, drawing style, and type of the analyzed image of the sample image, and calculate each score. Therefore, by comparing these scores, it is possible to implement a function of deriving similar artists.

Accordingly, the server 100 may compare the styles of the artists by grasping the information of the artists classified into the group of writers, or classify the artists having similar styles.

In addition, according to an embodiment of the present invention, prior to the step of receiving a sample image (step S430), it may further include the author group learning model building step (step S410).

In more detail, the server 100 receives an image or a tracing image corresponding to the style of the artist group to which each artist group tracing model 170 belongs, extracting a feature vector from the image or tracing image, the artist group tracing model 170 ), Including the author group learning model construction step.

Accordingly, the server 100 may learn the artist group tracing model 170 to allow the artist group tracing model 170 to have a clear picture of the artist group.

The artist group tracing model 170 includes an artist group determination module 150 that derives an artist group by analyzing an image through the learned model.

In more detail, when the server 100 receives an image from the writer client 300, the server 100 analyzes the image, analyzes what styles the artist has, and classifies the corresponding artist into a group of artists corresponding to the style of the artist. I mean.

The server 100 may use the GAN algorithm when training the artificial intelligence by building a learning model. And, it can proceed to an unsupervised learning method such as K-mean algorithm.

5 is an exemplary diagram illustrating an image input by a writer according to an embodiment of the present invention to the server 100, FIG. 6 provides a tracing image by applying the artist's style to FIG. 5, and the artist provides a tracing image. It is an exemplary figure which illustrated what was retouched.

5 and 6, the server 100 receives the image of FIG. 5 from the artist client, generates a tracing image, and provides the tracing image to the artist client 300, and the artist retouches the tracing image to the server 100. It illustrates the transmission to.

5 and 6, FIG. 5 is a real picture taken by a camera, and FIG. 6 is a server 100 generating a tracing image by applying an artist's style using a personal tracing model 190. You can see that background images such as cartoons, webtoons, and animations have been created.

As such, when the artist receives the tracing image from the server 100, the artist only has to go through a retouching process of drawing a person, an object, or the like as shown in FIG. 6, and thus does not have to spend time drawing a background image.

7 is a block diagram of a server 100 that provides an artist's painting style to an image according to an embodiment of the present invention.

The receiver 110 receives an image or a retouching image from the writer client 300.

The personal tracing model 190 generates a tracing image by applying the artist's style to the image received from the writer client 300, and provides the tracing image to the writer client 300.

When the artist group tracing model 170 receives the image from the artist client 300 for the first time, the artist group tracing model 170 generates a tracing image by applying a style of the artist group to which the artist belongs, and provides the image to the artist client 300.

As another example, when there is no personal tracing model 190 set to the artist, the server 100 determines that the artist uses the service for the first time, and the artist group to which the artist belongs to the image using the artist group tracing model 170. The tracing image is applied to generate a tracing image, and to be provided to the artist client 300.

The server 100 stores the retouching image received from the writer client 300 in the personal tracing model 190 to learn the personal tracing model 190 to build a learning model.

In more detail about the first use of the service by the writer, the receiver 110 receives one or more sample images from the writer client 300.

Then, the server 100 analyzes the sample image to derive the artist's style, and sets the artist as the artist group corresponding to the derived style.

The analysis unit 130 analyzes the sample image to derive the artists who are similar to the artist, and analyzes the similarity between the artists.

Next, the construction of the learning model will be described.

The server 100 extracts a feature vector from the retouching image received from the writer client 300, and stores the extracted feature vector in the personal tracing model 190 to train artificial intelligence.

In addition, when the server 100 generates and provides a tracing image, the tracing model extracts the feature vector from the image received from the writer client 300, and the tracing model applies the artist's style to the feature vector of the image. The tracing image is generated and provided to the writer client 300.

In this case, the tracing model means the artist group tracing model 170 or the personal tracing model 190.

Since the server 100 does not have the personal tracing model 190 set for the artist when the artist first transmits the image to the server 100 or when the artist determines that the artist uses the service for the first time, the artist group tracing is performed. By copying the model 170, a personal tracing model 190 used only for the corresponding artist is generated, and the personal tracing model 190 is trained.

Therefore, when the artist uses the service in the future, the tracing image is provided using the personal tracing model 190 set to the artist without using the artist group tracing model 170.

In addition, the server 100 to learn the artist group tracing model 170 as follows.

The server 100 receives an image or a tracing image corresponding to the style of the artist group to which each artist group tracing model 170 belongs.

The server 100 extracts the feature vector from the image or the tracing image and stores the feature vector in the artist group tracing model 170 to build the artist group learning model.

In addition, the artist group determination module 150 derives the artist group by analyzing the image received from the artist using the learning model built in the artist group tracing model 170.

The method of applying the artist's painting style to the image according to the embodiment of the present invention described above may be implemented as a program (or an application) and stored in a medium to be executed in combination with a server which is hardware.

The above-described program may include C, C ++, JAVA, machine language, etc., which the computer's processor (CPU) can read through the computer's device interface, so that the computer reads the program and executes the methods implemented as the program. Code may be coded in the computer language of. Such code may include functional code associated with a function or the like that defines the necessary functions for executing the methods, and includes control procedure related control code necessary for the processor of the computer to execute the predetermined procedure. can do. In addition, the code may further include memory reference code for additional information or media required for the computer's processor to execute the functions at which location (address address) of the computer's internal or external memory should be referenced. have. In addition, if the processor of the computer needs to communicate with any other computer or server remotely in order to execute the functions, the code may be used to communicate with any other computer or server remotely using the communication model of the computer. It may further include a communication related code for whether to communicate, what information or media should be transmitted and received during communication.

The stored medium refers to a medium that stores data semi-permanently and can be read by a device, not a medium that stores data for a short time such as a register, a cache, a memory, and the like. Specifically, examples of the storage medium include, but are not limited to, a ROM, a RAM, a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like. That is, the program may be stored in various recording media on various servers to which the computer can access or various recording media on the computer of the user. The medium may also be distributed over network coupled computer systems so that the computer readable code is stored in a distributed fashion.

The steps of a method or algorithm described in connection with an embodiment of the present invention may be implemented directly in hardware, in a software model executed by hardware, or by a combination thereof. Software models may include random access memory (RAM), read only memory (ROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, hard disk, removable disk, CD-ROM, or It may reside in any form of computer readable recording medium well known in the art.

In the above, embodiments of the present invention have been described with reference to the accompanying drawings, but those skilled in the art to which the present invention pertains may realize the present invention in other specific forms without changing the technical spirit or essential features thereof. I can understand that. Therefore, it should be understood that the embodiments described above are exemplary in all respects and not restrictive.

100: server
110: receiver
130: analysis unit
150: artist group discrimination module
170: writer group tracing model
190: personal tracing model
300: writer and client

Claims (10)

As a computer-implemented method,
The server receiving one or more images from the writer client;
When a personal tracing model applies a writer's style to the image to generate a tracing image and provides it to the writer's client, but the first time the image is received from the writer, the writer group tracing model to which the writer belongs belongs to the style of the writer's group. Generating a tracing image by applying a tracing image;
Receiving a retouching image from the writer client, wherein the retouching image is a retouching of the tracing image by the writer; And
A method of applying a painter's style to an image, including a learning model construction step of storing a retouching image in a personal tracing model. The method of claim 1,
The server receiving one or more sample images from the writer client; And
And analyzing the sample image to derive an artist's style of painting and setting the corresponding artist as a group of artists corresponding to the style of painting. The method of claim 2,
Analyzing the sample image to derive artists of a style similar to the artist, and analyzing similarity among the artists. The method of claim 1,
The learning model building step,
Extracting a feature vector from the retouching image; And
And a learning model construction step of training the artificial intelligence by storing the extracted feature vectors in a personal tracing model. The method of claim 1,
The providing step,
Extracting a feature vector from the received image by a tracing model; And
And a tracing model generating a tracing image by applying the artist's style to the feature vector of the image, and providing the tracing image to the artist's client. The method of claim 1,
The learning model building step,
And copying the artist group tracing model to generate a personal tracing model used only for the artist. The method of claim 2,
Receiving an image or a tracing image corresponding to the painting style of the artist group to which each artist group tracing model belongs; And
And a writer group learning model building step of the artist group tracing model extracting and storing a feature vector from the image or the tracing image. The method of claim 1,
The server,
A method of applying a painter's style to an image, including an artist group discrimination module for analyzing an image through a learning model built on the artist group tracing model. A receiving unit for receiving an image or a retouching image from the writer client;
A personal tracing model that generates a tracing image by applying the artist's style to the image and provides the tracing client to the image; And
When receiving the image from the writer client for the first time, includes a writer group tracing model for generating a tracing image by applying the style of the artist group belonging to the image,
And storing the retouching image in the personal tracing model to train the personal tracing model. A program coupled with a computer, which is hardware, for applying an artist's style to an image stored in a medium for executing the method of any one of claims 1 to 8.

Images