KR20210062274A - Device and method for image automatic generation - Google Patents

Abstract

According to an embodiment of the present invention, a device for automatically generating an image comprises at least one processor. The at least one processor generates a second image of a different domain based on a first image, and extracts a preset parameter from the generated second image, and generates a third image based on the extracted parameter. It is possible to automatically generate an image of a domain different from an acquired image.

Description

Automatic image generation device and generation method {DEVICE AND METHOD FOR IMAGE AUTOMATIC GENERATION}

The following embodiments relate to an apparatus and a method for automatically generating an image.

Machine learning is a field of artificial intelligence, a field that has evolved from the study of pattern recognition and computer learning theory, and refers to the field of developing algorithms and technologies that enable computers to learn.

The core of machine learning lies in representation and generalization. Representation is the evaluation of data, and generalization is the processing of data that is not yet known. It is also a field of computational learning theory.

Deep learning is defined as a set of machine learning algorithms that attempt a high level of abstraction through a combination of several nonlinear transducers. It can be said that it is a field.

Generative Adversarial Network (GAN) learns through competition between two neural network models and produces results. The two models, called “Generator” and “Discriminator,” have opposite purposes. The constructor learns real data and generates false data based on it. The purpose is to generate false data that is close to the real thing. The discriminator learns to determine whether the data presented by the generator is real or false.

According to an embodiment of the present invention, it is possible to provide an image automatic generating apparatus and method capable of automatically generating an image of a domain different from that of an acquired image.

In addition, according to another embodiment of the present invention, it is possible to provide an image automatic generating apparatus and method capable of automatically generating an image in consideration of differences according to different domains.

In addition, according to another embodiment of the present invention, it is possible to provide an image automatic generating apparatus and method capable of automatically generating an image in consideration of an average and a deviation of colors constituting an acquired image.

In addition, according to another embodiment of the present invention, it is possible to provide an apparatus and a method for automatically generating an image in which a parameter extracted from the acquired image is different according to the acquired image.

According to an embodiment of the present invention, in the automatic image generation apparatus, including at least one processor, the at least one processor generates a second image of a different domain based on the first image, and the generated A preset parameter is extracted from the second image, and a third image is generated based on the extracted parameter.

In addition, the at least one processor may detect a preset area from the first image and extract a reference area including facial landmarks from the detected area.

In addition, the at least one processor detects differences between features of a domain to which the first image belongs and features of a domain to which the second image belongs, and assigns a weight to the detected difference. can do.

Also, the at least one processor may detect an average and a deviation of colors constituting the first image.

In addition, the preset parameter may be changed based on the first image.

Also, the at least one processor may modify the generated third image.

According to another embodiment of the present invention, generating a second image of a different domain based on the first image, extracting a preset parameter from the generated second image, and a third image based on the extracted parameter. It includes the operation of generating.

In addition, the operation of generating the second image may include an operation of detecting a preset region from the first image and an operation of extracting a reference region including facial landmarks from the detected region. have.

In addition, the generating of the second image includes an operation of detecting a difference between features of a domain to which the first image belongs and a feature of a domain to which the second image belongs, and the detected difference. May include an operation of weighting

In addition, the operation of generating the second image may include an operation of detecting an average and a deviation of colors constituting the first image.

In addition, the preset parameter may be changed based on the first image.

In addition, the method of automatically generating an image may further include an operation of modifying the generated third image.

According to an embodiment of the present invention, there is an effect of automatically generating an image of a domain different from the acquired image.

In addition, there is an effect of automatically generating an image in consideration of differences according to different domains.

In addition, there is an effect of automatically generating an image in consideration of the average and deviation of colors constituting the acquired image.

In addition, there is an effect that parameters extracted from the acquired image vary according to the acquired image.

1 is a diagram illustrating a configuration of an apparatus for automatically generating an image according to an exemplary embodiment.
2 is a flowchart illustrating a method of automatically generating an image according to an exemplary embodiment.
3 is a diagram illustrating a state in which a face region is detected from a photo including a face of a real person and a reference region is extracted from the detected face region, according to an exemplary embodiment.
4 is a diagram illustrating an example of extracting a parameter from a face image according to an exemplary embodiment.
5 is a diagram illustrating a state in which an image is generated based on an extracted parameter according to an embodiment.
6 is a diagram illustrating a state in which a plurality of game characters are generated from a photo including a face of a real person according to an exemplary embodiment.

Specific structural or functional descriptions of the embodiments according to the concept of the present invention disclosed in the present specification are exemplified only for the purpose of describing the embodiments according to the concept of the present invention, and the embodiments according to the concept of the present invention are It may be implemented in various forms and is not limited to the embodiments described herein.

Since the embodiments according to the concept of the present invention can apply various changes and have various forms, the embodiments will be illustrated in the drawings and described in detail in the present specification. However, this is not intended to limit the embodiments according to the concept of the present invention to specific disclosed forms, and includes all changes, equivalents, or substitutes included in the spirit and scope of the present invention.

Terms such as first or second may be used to describe various constituent elements, but the constituent elements should not be limited by the terms. The terms are only for the purpose of distinguishing one component from other components, for example, without departing from the scope of the rights according to the concept of the present invention, the first component may be referred to as the second component, and similarly The second component may also be referred to as a first component.

When a component is referred to as being "connected" or "connected" to another component, it is understood that it may be directly connected or connected to the other component, but other components may exist in the middle. It should be. On the other hand, when a component is referred to as being "directly connected" or "directly connected" to another component, it should be understood that there is no other component in the middle. Other expressions describing the relationship between components, such as "between" and "directly between" or "adjacent to" and "directly adjacent to" should be interpreted as well.

The terms used in the present specification are only used to describe specific embodiments, and are not intended to limit the present invention. Singular expressions include plural expressions unless the context clearly indicates otherwise.

In the present specification, terms such as "comprise" or "have" are intended to designate that the described feature, number, step, action, component, part, or combination thereof is present, but one or more other features or numbers It is to be understood that the possibility of addition or presence of, steps, actions, components, parts, or combinations thereof is not preliminarily excluded.

Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs.

Terms as defined in a commonly used dictionary should be construed as having a meaning consistent with the meaning of the related technology, and should not be interpreted as an ideal or excessively formal meaning unless explicitly defined in the present specification. Does not.

In the following description, the same identification symbols mean the same configuration, and unnecessary redundant descriptions and descriptions of known technologies will be omitted.

In an embodiment of the present invention,'communication','communication network', and'network' may be used with the same meaning. The three terms refer to wired/wireless local and wide area data transmission/reception networks capable of transmitting and receiving files between a user terminal, a terminal of other users, and a download server.

Hereinafter, the present invention will be described in detail by describing a preferred embodiment of the present invention with reference to the accompanying drawings.

1 is a diagram illustrating a configuration of an apparatus for automatically generating an image according to an exemplary embodiment.

Referring to FIG. 1, the apparatus 100 for automatically generating an image includes a processor 110, an input/output interface module 120, and a memory 130.

The processor 110, the input/output interface module 120, and the memory 130 constituting the image generating apparatus 100 are interconnected, and data can be transmitted to each other.

The processor 110 may execute programs or instructions stored in the memory 130. In this case, an operation program (eg, OS) for operating the image automatic generating device 100 may be stored in the memory 130.

The processor 110 may execute a program for managing information on the apparatus 100 for automatically generating an image.

The processor 110 may execute a program for managing the operation of the apparatus 100 for automatically generating an image.

The processor 110 may execute a program for managing the operation of the input/output interface module 120.

Ⅰ) Creating a second image

The processor 110 may acquire the first image through the input/output interface module 120. In this case, the first image may be a photograph including a face of a real person, but the first image is not limited thereto.

The processor 110 may generate a second image of a different domain based on the acquired first image.

The processor 110 may detect a preset area in the acquired first image using a detector (eg, a facial landmark detector). In this case, the preset area may vary according to an object included in the first image.

According to an embodiment, when the first image is a photo including a face of a real person, the processor 110 may detect a region including the face of the real person in the first image as a preset region. .

The processor 110 may acquire facial landmarks in the detected area using a detector (eg, a facial landmark detector).

The processor 110 may determine, as a facial feature point, each of a plurality of points that can be easily identified even if the shape, size, and location of the object included in the detected area changes using a detector (eg, a facial landmark detector). Facial landmarks may be generated based on the feature points for each of the plurality of points.

According to an embodiment, the processor 110 may generate facial feature point information based on a landmark point determined using a detector (eg, a facial landmark detector), but generates the facial feature point information. Information that can be used to do so is not limited thereto.

The processor 110 may extract a reference area including facial landmarks acquired from the detected area using a detector (eg, a facial landmark detector). In this case, the reference region may have a rectangular shape, but the shape of the reference region is not limited thereto.

The processor 110 may not detect an object included in the first image through the input/output interface module 120 (eg, there is no face in the photo input as the first image), or the plurality of Objects overlap each other so that it is difficult to identify each of the plurality of objects included in the first image (eg, another object covers the face included in the photo input as the first image), or included in the first image When it is difficult to determine facial landmarks on an object (for example, when the picture input as the first image is a picture where only the side face is taken), the input of a new first image can be requested through the input/output interface module 120. have.

The processor 110 may acquire first domain features from a domain (eg, a photo) to which the first image belongs using an encoder (eg, a deep neural network (DNN)). . In this case, when the first image is a photo including a face of a real person, the first domain features include a face shape of the person included in the photo, eyes, nose, mouth constituting the face, It may be a shape of an ear or the like or a texture of the skin (eg, whether a beard is present), but the first domain features are not limited thereto.

The processor 110 may acquire second domain features from a domain (eg, game style) to which the second image belongs using an encoder (eg, a deep neural network (DNN)). have. In this case, when the second image is a game style character image, the second domain features are the shape of the face of the game style character, the shape of the eyes, nose, mouth, and ears constituting the face, or of the skin. It may be a texture (eg, whether a beard is present), but the second domain features are not limited thereto.

The processor 110 may detect a difference between the acquired first domain features and the acquired second domain features.

The processor 110 may generate a feature map for the detected difference.

The processor 110 may assign a weight to the generated feature map. In this case, the weight may be referred to as attention.

The processor 110 may generate a loss function for an attention feature map, which is a feature map to which the weight is assigned, in order to generate a more natural second image based on the first image. .

The processor 110 uses a decoder (e.g., a deep neural network (DNN)) based on a loss function for the reference region and the generated attention feature map. , Face image) can be created.

The processor 110 may detect an average and a deviation of colors constituting the first image.

The processor 110 may detect an average and a deviation of colors constituting the reference area.

The processor 110 uses a decoder (e.g., a deep neural network (DNN)) based on an average and a deviation of the reference region and the detected colors constituting the reference region. , Face image) can be created. In this case, in order to reduce the difference between the colors constituting the reference area and the colors constituting the second image, the processor 110 includes an average and a deviation of the colors constituting the detected reference area and the second image (e.g. The second image (eg, a face image) may be generated so that the average and the deviation of the colors constituting the face image) are the same.

The processor 110 uses a decoder (e.g., a deep neural network (DNN)) to determine the reference region, a loss function for the generated attention feature map, and the detected reference region. A second image (eg, an image of a face portion) may be generated based on an average and a deviation of the compose colors. In this case, in order to reduce the difference between the colors constituting the reference area and the colors constituting the second image (eg, a face image), the processor 110 is an average and a deviation of the colors constituting the detected reference area. And the second image may be generated such that an average and a deviation of colors constituting the second image (eg, a face image) are the same.

Ii) Parameter extraction

The processor 110 may extract a preset parameter from the generated second image (eg, a face image) using an encoder (eg, a deep neural network (DNN)). In this case, the preset parameter may be a parameter necessary to generate a third image, but the preset parameter is not limited thereto.

According to an embodiment, when the third image is a head part of a game character, the processor 110 may select at least one parameter that contributes to an external change of the face of the game character. In this case, the at least one parameter may be a parameter related to an eye size, an eyebrow shape, a chin angle, and the like, but the at least one parameter is not limited thereto.

According to an embodiment, the processor 110 may extract a preset parameter from the generated second image (eg, a face image) using a convolutional neural network (CNN).

According to an embodiment, an encoder (e.g., a deep neural network (DNN)) used by the processor 110 to extract a preset parameter from a second image (e.g., a face image) is a face recognition database ( VGGFace2) may be a retraining (fine-tuning) of an encoder (eg, a deep neural network (DNN)) having a VGG-16 structure that has been learned in advance.

The processor 110 may change the preset parameter based on the first image.

According to an embodiment, when the first image is a picture including a face of a real person, the processor 110 may extract a parameter related to the shape of the face or elements constituting the face from the first image.

Iii) Creating a third image

The processor 110 may generate a third image based on the extracted parameter.

According to an embodiment, when the third image is the head of a game character, the processor 110 may use the game engine to generate (render) the third image based on the extracted parameter.

The processor 110 may generate a third image based on the extracted parameter using a decoder (eg, a deep neural network (DNN)). In this case, the decoder may be a game engine, but the decoder is not limited thereto.

The processor 110 may output the generated third image through the input/output interface module 120.

The processor 110 may match the generated third image with the parameter that generated the third image and store it in the memory 130 or a face image database (not shown).

According to an embodiment, when the third image is a head part of a game character, the processor 110 may generate a thumbnail by extracting a face part from the generated third image.

The processor 110 may match the generated thumbnail with the parameter that generated the third image and store it in the memory 130 or a face image database (not shown).

The processor 110 may modify the generated third image based on a user input acquired through the input/output interface module 120.

The processor 110 may modify the generated third image based on the previously generated third image stored in the memory 130 and the generated parameter.

The processor 110 may modify the generated third image based on a thumbnail generated based on a third image generated in the past stored in the memory 130 and a parameter that generated the image.

The processor 110 may modify the generated third image based on an Asian face image stored in a face image database (not shown).

The processor 110 may use a decoder (eg, a deep neural network (DNN)) to modify the generated third image.

The processor 110 may add a face image database (not shown) in which Asian face images are stored in order to learn an encoder or a decoder (eg, a deep neural network (DNN)).

The processor 110 matches the third image and the parameter generated by the third image to learn an encoder or a decoder (eg, a deep neural network (DNN)), and Can be saved.

The processor 110 matches a thumbnail generated based on a third image and a parameter generated by the third image in order to learn an encoder or a decoder (eg, a deep neural network (DNN)) to provide facial image data. It can be stored in a base (not shown).

The processor 110 may acquire a large amount of face images generated by random model parameters from an external device (eg, a game client) in order to learn an encoder or a decoder (eg, a deep neural network (DNN)). .

The input/output interface module 120 may be connected to an external device (eg, a game client) through a network.

The input/output interface module 120 may obtain data from an external device.

The input/output interface module 120 may acquire an image from an external device.

The input/output interface module 120 may obtain a parameter from an external device.

The input/output interface module 120 may acquire learning data from an external device.

The input/output interface module 120 may obtain a user's input.

The input/output interface module 120 may output an image generated by the processor 110.

The input/output interface module 120 may be provided integrally with the image automatic generating device 100.

The input/output interface module 120 may be provided separately from the image automatic generating device 100.

The input/output interface module 120 may be a separate device to be communicatively connected with the image automatic generating device 100.

The input/output interface module 120 may include a port (eg, a USB port) for connecting to an external device.

The input/output interface module 120 may include a monitor, a touch screen, a mouse, an electronic pen, a microphone, a keyboard, a speaker, an earphone, a headphone, or a touch pad.

The memory 130 may store data acquired through the input/output interface module 120.

The memory 130 may store an image acquired through the input/output interface module 120.

The memory 130 may store images generated by the processor 110.

The memory 130 may store facial landmarks acquired by the processor 110.

The memory 130 may store the reference area extracted by the processor 110.

The memory 130 may store the domain characteristics acquired by the processor 110.

The memory 130 may store a feature map generated by the processor 110.

The memory 130 may store a weight assigned by the processor 110.

The memory 130 may store a loss function for an attention feature map generated by the processor 110.

The memory 130 may store an average and a deviation of colors constituting an image detected by the processor 110.

The memory 130 may store parameters extracted by the processor 110.

The memory 130 may store a parameter selected by the processor 110.

The memory 130 may store a thumbnail generated by the processor 110.

The memory 130 may store an image modified by the processor 110.

The term'device or module' used herein denotes a logical structural unit, and it is obvious to those skilled in the art that the present invention is not necessarily a physically classified component.

2 is a flowchart illustrating a method of automatically generating an image according to an exemplary embodiment.

Referring to FIG. 2, the apparatus for automatically generating an image detects a preset area in the first image (200).

In this case, the preset area may vary according to an object included in the first image.

In addition, when the first image is a photograph including a face of a real person, the automatic image generation device may detect a region including the face of the real person in the first image as a preset region.

The automatic image generating apparatus extracts a reference area including facial landmarks from the detected area (210).

In this case, the apparatus for automatically generating an image may generate facial landmarks based on a landmark point determined using a detector (eg, a facial landmark detector).

In addition, the apparatus for automatically generating an image may extract a reference area having a smaller range than the detected area including facial landmarks acquired from the detected area.

The apparatus for generating an image automatically detects differences between features of a domain to which the first image belongs and features of a domain to which the second image belongs (220).

In this case, the apparatus for automatically generating an image may acquire first domain features from a domain (eg, a photo) to which the first image belongs.

In addition, the apparatus for automatically generating an image may acquire second domain features from a domain (eg, a game style) to which the second image belongs.

In addition, the apparatus for automatically generating an image detects a difference between the acquired first domain features and the acquired second domain features, and provides a feature map for the detected difference. ) Can be created.

The automatic image generation apparatus assigns a weight to the detected difference (230).

In this case, the automatic image generation device generates a loss function for an attention feature map, which is a feature map to which the weight is assigned, in order to generate a more natural second image based on the first image. can do.

The automatic image generation device detects an average and a deviation of colors constituting the reference region (240).

The automatic image generating device generates a second image (250).

In this case, the automatic image generation apparatus generates a second image based on the reference region, a loss function for the generated attention feature map, and the detected average and deviation of colors constituting the reference region. Can be generated.

The automatic image generation device extracts a preset parameter from the generated second image (260).

In this case, the preset parameter may be a parameter necessary to generate a third image.

In addition, the apparatus for automatically generating an image may extract at least one parameter that contributes to an external change of the object expressed in the third image.

An automatic image generating device generates a third image based on the extracted parameter (270).

In this case, the apparatus for automatically generating an image may use a game engine to generate the third image.

Also, the apparatus for automatically generating an image may modify the generated third image based on a user's input.

The image generating device outputs the generated third image (280).

In this case, the apparatus for automatically generating an image may output the generated third image to a monitor.

3 is a diagram illustrating a state in which a face region is detected from a photo including a face of a real person and a reference region is extracted from the detected face region, according to an exemplary embodiment.

Referring to FIG. 3, FIG. 3 (a) is a diagram illustrating a state of detecting a face region, and FIG. 3 (b) is a diagram illustrating a state of extracting a reference region.

Referring to FIG. 3A, the apparatus for automatically generating an image may detect only the face area 300 using a detector (eg, a facial landmark detector) from a photograph including a face of a real person.

Referring to FIG. 3(b), the apparatus for automatically generating an image may acquire facial feature point information 321, 322, 323, 324, 325 from an area 310 detected using a detector (eg, a facial landmark detector). have.

The automatic image generation device can determine a plurality of landmark points 321, 322, 323, 324, 325, which are points that can be easily identified even if the shape, size, and location of the object included in the detected area 310 change. In addition, facial landmarks may be generated based on the determined plurality of feature points 321, 322, 323, 324, and 325.

The apparatus for automatically generating an image may extract a reference area 320 including facial landmarks acquired from the detected area 310.

The apparatus for automatically generating an image may extract the reference region 320 so as to have a predetermined interval 311 within the detected region 310.

4 is a diagram illustrating an example of extracting a parameter from a face image according to an exemplary embodiment.

Referring to FIG. 4, the apparatus for automatically generating an image may extract a parameter 420 from the face image 400 and 410.

The apparatus for automatically generating an image may use an encoder (eg, a deep neural network (DNN)) to extract a parameter.

The apparatus for automatically generating an image may extract 420 at least one parameter contributing to an external change of a face included in the partial face images 400 and 410.

At least one parameter contributing to an external change of the included face extracted by the automatic image generating device may be an eye size, a chin angle, and eyebrow shapes 430 and 440.

The parameters extracted by the automatic image generating apparatus may be displayed as numbers 431 and 441, and when the displayed numbers 431 and 441 are different, they may contribute differently to an external change of the face.

5 is a diagram illustrating a state in which an image is generated based on an extracted parameter according to an exemplary embodiment.

Referring to FIG. 5, the apparatus for automatically generating an image may generate the head image 550 and 560 of a game character based on the extracted parameter 500.

The image automatic generating apparatus may generate the head image 550 and 560 of the game character corresponding to each of the parameters 510 and 520 having different external characteristics.

The parameters 510 and 520 extracted by the automatic image generating apparatus may be displayed as numbers 511 and 521, and when the displayed numbers 511 and 521 are different, they may contribute differently to the appearance change of the face.

Each of the parameters 510 and 520 extracted by the automatic image generating apparatus may be displayed on one interface 530.

The apparatus for automatically generating an image may use the game engine 540 to generate the head image 550 and 560 of the game character corresponding to the extracted parameters 510 and 520, respectively. In this case, the game engine 540 renders the head image 550 and 560 of the game character.

6 is a diagram illustrating a state in which a plurality of game characters are generated from a photo including a face of a real person according to an exemplary embodiment.

Referring to FIG. 6, the apparatus for automatically generating an image may acquire a photo 611 including a face of a real person, and automatically generate 610 a plurality of game characters 620 based on the acquired photo.

The apparatus for automatically generating an image may acquire a photo 611 including a face of a real person, and generate a game style image based on the acquired photo.

The automatic image generation device uses an encoder (e.g., Deep Neural Network (DNN)) to contribute to the appearance change of the face in the generated game style image, eye size, chin angle, eyebrow shape, beard, and face. Parameters such as wounds, hairstyles, and hair color can be extracted.

The apparatus for automatically generating an image may automatically generate 610 a plurality of game characters 620 based on the extracted parameters using a decoder (eg, a deep neural network (DNN)).

The automatic image generation device can create a plurality of game characters 621, 622, 623, 624 having different eye sizes, chin angles, eyebrow shapes, beards, facial wounds, hairstyles, hair colors, etc. according to the underlying parameters. have.

In the above, even if all the constituent elements constituting the embodiments of the present invention are described as being combined into one or operating in combination, the present invention is not necessarily limited to these embodiments. That is, within the scope of the object of the present invention, all the constituent elements may be selectively combined into at least one and operated.

In addition, although all of the components may be implemented as one independent hardware, a program module that performs some or all functions combined in one or more hardware by selectively combining some or all of the components. It may be implemented as a computer program having Codes and code segments constituting the computer program may be easily inferred by those skilled in the art of the present invention.

Such a computer program is stored in a computer-readable storage medium, and is read and executed by a computer, thereby implementing an embodiment of the present invention. The storage medium of the computer program may include a magnetic recording medium, an optical recording medium, and the like.

In addition, the terms such as "include", "consist of" or "have" described above mean that the corresponding component may be included unless otherwise stated, excluding other components. It should not be construed as being able to further include other components.

All terms, including technical or scientific terms, unless otherwise defined, have the same meaning as commonly understood by one of ordinary skill in the art to which the present invention belongs. Terms generally used, such as terms defined in the dictionary, should be interpreted as being consistent with the meaning in the context of the related technology, and are not interpreted as ideal or excessively formal meanings unless explicitly defined in the present invention.

The methods disclosed in the present invention comprise one or more actions or steps for achieving the above-described method. Method actions and/or steps may be interchanged with each other without departing from the scope of the claims. In other words, unless a specific order for the actions or steps is specified, the order and/or use of specific actions and/or steps may be modified without departing from the scope of the claims.

As used herein, a phrase referring to “at least one of” in a list of items refers to any combination of these items, including single members. As an example, “at least one of a, b, or c:” means a, b, c, ab, ac, bc, and abc, as well as any combination with multiples of the same element (e.g., aa , aaa, aab, aac, abb, acc, bb, bbb, bbc, cc, and ccc or any other ordering of a, b, and c).

As used herein, the term "determining" encompasses a wide variety of actions. For example, “determining” may include computing, computing, processing, deriving, examining, looking up (eg, looking up in a table, database, or other data structure), identifying, and the like. . Further, “determining” may include receiving (eg, receiving information), accessing (accessing data in a memory), and the like. Also, “determining” may include resolving, choosing, choosing, establishing, and the like.

The above description is merely illustrative of the technical idea of the present invention, and those of ordinary skill in the art to which the present invention pertains will be able to make various modifications and variations without departing from the essential characteristics of the present invention.

Accordingly, the embodiments disclosed in the present invention are not intended to limit the technical idea of the present invention, but to explain the technical idea, and the scope of the technical idea of the present invention is not limited by these embodiments. The scope of protection of the present invention should be construed by the following claims, and all technical ideas within the scope equivalent thereto should be construed as being included in the scope of the present invention.

100... Automatic image generation device

Claims (10)

In the image automatic generating device,
Including at least one processor,
The at least one processor,
Creating a second image of a different domain based on the first image,
Extracting a preset parameter from the generated second image
An image automatic generation device that generates a third image based on the extracted parameters. The method of claim 1,
The at least one processor,
Detecting a preset area in the first image,
Automatic image generation apparatus for extracting a reference area including facial landmarks from the detected area. The method of claim 1,
The at least one processor,
Detecting differences between features of the domain to which the first image belongs and features of the domain to which the second image belongs,
An image automatic generation device that assigns a weight to the detected difference. The method of claim 1,
The at least one processor,
An image automatic generation device that detects an average and a deviation of colors constituting the first image. The method of claim 1,
The preset parameters are,
An apparatus for automatically generating an image that is changed based on the first image. Generating a second image of a different domain based on the first image;
Extracting a preset parameter from the generated second image; And
Generating a third image based on the extracted parameters
Automatic image generation method comprising a. The method of claim 7,
The operation of generating the second image,
Detecting a preset area in the first image; And
Extracting a reference area including facial landmarks from the detected area
Automatic image generation method comprising a. The method of claim 7,
The operation of generating the second image,
Detecting differences between features of a domain to which the first image belongs and features of a domain to which the second image belongs; And
Assigning a weight to the detected difference
Automatic image generation method comprising a. The method of claim 7,
The operation of generating the second image,
The operation of detecting the average and deviation of colors constituting the first image
Automatic image generation method comprising a. The method of claim 7,
The preset parameters are,
A method of automatically generating an image that is changed based on the first image.

Images