KR102259126B1 - Appartus and method for generating customizing image - Google Patents

Abstract

A method and apparatus for creating a custom image are disclosed. According to an embodiment of the present invention, the method for generating a custom image includes the steps of receiving a 2D input image from a client, extracting feature points from the 2D input image, selecting a basic customized image based on the feature points from a database, and and generating a final 3D model based on the basic customized image and the feature point, and providing one or more preset images using the final 3D model.

Description

Apparatus and method for generating a custom image {APPARTUS AND METHOD FOR GENERATING CUSTOMIZING IMAGE}

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

A customization service capable of modifying the appearance of a character in a PC (Personal Computer) game, a video game, a wireless terminal game, and a network game is provided. In particular, in the case of a game in which multiple players connect at the same time, such as MMORPG, MMOSNG (Massively Multiplayer Online Social Network Game), and MMOSLG (Massively Multiplayer Online Simulation Game), the appearance of the character is used as a means of reflecting the individuality of the player, It acts as an important factor in attracting players to the game.

According to an embodiment of the present invention, the custom image generating apparatus may simply and accurately generate a 3D output custom image by using a 2D input image that a user can easily collect.

In addition, according to another embodiment of the present invention, the client starts the process of generating an output customization image based on the motion input received from the user, so that the system provides the user with an experience in which a 2D input image is converted into a 3D output customized image. and improve the user experience.

According to an embodiment of the present invention, a method for generating a custom image includes: receiving a 2D input image from a client; extracting feature points from the 2D input image; selecting a basic customized image from a database based on the feature point; and generating a final 3D model based on the basic customized image and the feature point, and providing one or more preset images using the final 3D model.

In addition, the selecting of the basic customized image may include selecting a basic customized image having a parameter similar to the characteristic point from among a plurality of basic customized images stored in the database.

In addition, the operation of providing the preset image may generate the final 3D model based on the parameters of the basic customized image.

In addition, the providing of the preset image may include inputting an input parameter of the final 3D model to a neural network and generating a final 3D model based on an output parameter.

In addition, the operation of receiving the 2D input image includes receiving one or more 2D input images corresponding to one or more body parts, and the operation of extracting the feature points includes extracting each feature point from the one or more 2D input images, The operation of providing the preset image may generate the final 3D model by combining feature points corresponding to the extracted one or more 2D input images.

The method for generating a customized image may further include learning a recommendation module using a final customized image selected and modified from among the one or more preset images by a user input.

According to an embodiment of the present invention, an apparatus for generating a customized image includes at least one processor; and an I/O interface in communication with a client, wherein the I/O interface receives a 2D input image from the client, the processor extracts a feature point from the 2D input image, and based on the feature point from a database may select a basic customized image, generate a final 3D model based on the basic customized image, and provide one or more preset images generated from the final 3D model.

The processor may select a basic customized image having a parameter similar to the feature point from among a plurality of basic customized images stored in the database.

The processor may generate the final 3D model based on the parameters of the basic customized image.

The processor may input the input parameter of the final 3D model to the neural network and generate the final 3D model based on the output parameter.

The I/O interface receives one or more 2D input images corresponding to one or more body parts, and the processor extracts each feature point from the one or more 2D input images, and adds the extracted one or more 2D input images. The final 3D model may be generated by combining corresponding feature points.

According to another embodiment of the present invention, at least one processor; a display comprising a first side and a second side; a folding sensor for detecting a folding state and an unfolding state of the first surface and the second surface; and an I/O interface communicating with a custom image generating device, wherein when the folding sensor detects an unfolded state, the first surface displays a 2D input image, and when the folding sensor detects a folded state , the I/O interface sends the 2D input image to the custom image generating device, the I/O interface receives one or more preset images from the custom image generating device, and customizes an output among the one or more preset images An image selection signal may be transmitted to the customized image generating apparatus, and the preset image may be generated from the 2D input image, the basic customized image stored in a database, and a 3D model generated based on a neural network.

According to another embodiment of the present invention, at least one processor; a display comprising a first side and a second side; a touch sensor for sensing a touch of the first surface and the second surface; and an I/O interface in communication with a custom image generating device, wherein the first side displays a 2D input image, and wherein the touch sensor moves from the first side toward the second side of the 2D input image. Upon detecting a drag operation, the I/O interface sends the 2D input image to the customizing image generating device, and the I/O interface receives one or more preset images from the customizing image generating device, and the one A selection signal of an output custom image among the preset images is transmitted to the custom image generating device, and the preset image may be generated from the 2D input image, the basic custom image stored in the database, and a 3D model generated based on a neural network. have.

According to an embodiment of the present invention, the custom image generating apparatus has an effect of simply and accurately generating a 3D output custom image by using a 2D input image that can be easily collected by a user.

According to an embodiment of the present invention, the system provides the user with an experience in which the 2D input image is converted into a 3D output customized image by the client starting the process of generating the output customizing image based on the motion input received from the user, It has the effect of improving the user experience.

1 is a diagram illustrating an overall configuration of a system for generating a customized image according to an embodiment.
2 is a flowchart illustrating an overall operation of a method for generating a customized image according to an exemplary embodiment.
3 is a flowchart illustrating a process of generating a customized image according to an exemplary embodiment.
4 is a diagram exemplarily illustrating a process of generating a custom image according to an embodiment.
5 is a diagram illustrating an overall configuration of an apparatus for generating a customized image according to an exemplary embodiment.
6 is a diagram illustrating an overall configuration of a client according to an embodiment.
7 is an example of generating a customized image of a client according to an embodiment.
8 is an example of generating a customized image of a client according to another 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 an overall configuration of a system for generating a customized image according to an embodiment.

According to an embodiment, the system may receive a 2D input image from a user and output an output customized image through the client. The system may analyze the 2D input image and output a 3D output customized image. Based on the analysis result of the 2D input image, the system may output the output customized image by synthesizing the accumulated data and the analysis result to generate the customized image. This allows the system to easily and accurately create 3D output customized images using 2D input images that users can easily collect.

Referring to FIG. 1 , a system may include a client 110 and a custom image generating apparatus 100 . The client 110 and the custom image generating apparatus 100 may be connected through a network. The custom image generating apparatus 100 may include a server. The custom image generating apparatus 100 may analyze the 2D input image received from the client 110 , synthesize the data accumulated for generating the custom image and the analysis result, and output the output customized image.

The custom image generating apparatus 100 may extract a feature point from the 2D input image. Here, the feature point may mean information for distinguishing one 2D input image from another 2D input image. The feature point may be referred to as a landmark, and is not limited to a point and may include a line or a surface. The feature point may occupy a physical location on the 2D input image, but may also include abstract information reflecting the characteristics of the 2D input image.

The custom image generating apparatus 100 may generate a 3D model using the feature points. The custom image generating apparatus 100 may generate a 3D model reflecting the feature points. Hereinafter, a 3D model initially generated by reflecting the feature points may be referred to as a basic 3D model. For example, the custom image generating apparatus 100 may generate a 3D model on a zero basis based on the feature point. As another example, the custom image generating apparatus 100 may generate a modified 3D model by reflecting feature points in an existing 3D model. The custom image generating apparatus 100 may select a suitable 3D model from among a plurality of existing 3D models as a basic model based on the feature point. The custom image generating apparatus 100 may generate the basic 3D model by changing the variable parameter of the basic model based on the feature point.

The custom image generating apparatus 100 may modify the basic 3D model using previously accumulated data. The customized image generating apparatus 100 may select a basic customized image to be used to correct the basic 3D model from among existing customized images. The customized image generating apparatus 100 may select a basic customized image based on the feature point. The customized image generating apparatus 100 may modify the basic 3D model by using the basic customized image. As such, the customized image generating apparatus 100 may reflect the user or a plurality of users' tendencies in the basic 3D model by modifying the basic 3D model using the existing customized image.

The custom image generating apparatus 100 may reflect previously accumulated data to the basic 3D model using a neural network. The neural network may be pre-trained by a deep learning method using training data. The training data means training data for training the neural network. The training data may include a 2D input image or a customized image. The training data may include vector information indicating the characteristics of the 2D input image or the customized image. The training data may include a training image pair. The training image pair includes a first training image pair and a second training image pair. The first training image pair may include a 2D input image and a customized image in which the 2D input image is well reflected. The second training image pair may include a 2D input image and a customized image in which the 2D input image is not well reflected.

Deep learning is a set of machine learning algorithms that attempt high-level abstraction (a task that summarizes core contents or functions in large amounts of data or complex data) through a combination of several nonlinear transformation methods. Defined. Deep learning can be viewed as a field of machine learning that teaches computers to think in a broad framework. Examples of deep learning techniques include Deep Neural Networks (DNN), Convolutional Deep Neural Networks (CNN), Reccurent Neural Network (RNN), and Deep Belief Networks (DBN). have. In addition, the custom image generating apparatus 100 may use a generative model such as Generative Adversarial Networks (GAN) or Variational Auto Encoder (VAE). A generative model is used to create a model capable of generating an image similar to a real image from sample images. For example, the generative model may be used to generate a photorealistic human face image from human face images.

The custom image generating apparatus 100 may modify the basic 3D model using a neural network. The custom image generating apparatus 100 may input input information representing the basic 3D model to the neural network. The input information may be converted into an output vector through each layer of the neural network in the form of a vector. The custom image generating apparatus 100 may modify the basic 3D model based on the output vector. As such, the custom image generating apparatus 100 may improve the quality of the basic 3D model by modifying the basic 3D model using the existing pre-trained neural network.

The user 120 may input a 2D input image to the client 110 . The user 120 may input a 2D input image to the client 110 and output a customized output image from the 2D input image through a simple operation. The client 110 may start a process of generating an output customizing image based on the motion input received from the user 120 . Here, the motion input may include an intuitive motion input based on the characteristics of the client 110 . Through this, the system may provide the user 120 with an experience in which a 2D input image is converted into a 3D output customized image, and the user experience may be improved.

For example, the user 120 may input an operation input for generating an output custom image to the client 110 through a folding operation of the foldable terminal. The client 110 may include a foldable terminal. The display of the client 110 may be formed of a foldable first surface and a second surface. The 2D input image input by the user 120 may be displayed on the first surface.

The user 120 may fold the first side and the second side of the client 110 . The client 110 may recognize the folding operation and start a process of generating a custom image for the 2D input image. The client 110 may transmit a 2D input image to the custom image generating apparatus 100 .

The custom image generating apparatus 100 may receive and analyze a 2D input image. The custom image generating apparatus 100 may extract a feature point from a 2D input image and output an output custom image using data accumulated for generating the custom image.

The customized image generating apparatus 100 may transmit the output customized image to the client 110 . The client 110 may display the output customization image on the second side of the display. As such, the 2D input image displayed on the first surface is displayed as an output customized image of the second surface through a folding operation, thereby providing an intuitive user experience to the user 120 .

As another example, the user 120 may input a motion input for generating an output customization image to the client 110 through a drag operation on the 2D input image displayed on the display. The client 110 may include a display. The display may be composed of a first side and a second side that are folded, or may be composed of a single side that is not folded.

A 2D input image input by a user may be displayed on the upper portion of the display. The user 120 may move the 2D input image to the lower end of the display through a drag operation. The client 110 may recognize the user 120's touch operation on the upper end of the display, and recognize the drag operation on the lower end of the display while the touch is maintained. The client 110 may start a process of generating a custom image for the 2D input image in response to the drag operation. The client 110 may transmit a 2D input image to the custom image generating apparatus 100 .

The custom image generating apparatus 100 may receive and analyze a 2D input image. The custom image generating apparatus 100 may extract a feature point from a 2D input image and output an output custom image using data accumulated for generating the custom image.

The customized image generating apparatus 100 may transmit the output customized image to the client 110 . The client 110 may display the output customization image at the bottom of the display. As such, the 2D input image displayed on the top of the display is converted through a drag operation, and an output customized image is displayed on the bottom of the display, thereby providing an intuitive user experience to the user 120 .

The client 110 refers to a terminal capable of transmitting and receiving various data using a communication intermediary device including an AP according to a user's key operation, and includes a tablet PC, a laptop, and a personal computer (PC). It may be any one of a computer), a smart phone, a personal digital assistant (PDA), and a mobile communication terminal. The client 110 may include a foldable smartphone. The client 110 is a terminal that performs voice or data communication using an AP and a communication network, and a memory for storing a program or protocol for communicating with an external device via the AP and a communication network, and operation and control by executing the program It means a terminal equipped with a microprocessor or the like for That is, the client 110 can be any terminal as long as communication is possible using an AP and a communication network, and is a broad concept including any communication computing device such as a notebook computer, a mobile communication terminal, and a PDA. On the other hand, the client 110 is preferably manufactured in a form equipped with a touch screen to perform input and output through a user's manipulation, and may drive a game program through an input according to a user's manipulation.

The client 110 preferably includes at least one of a video game terminal, an arcade terminal, a portable terminal, a PC terminal, and a wired or wireless terminal, and a predetermined key input means provided in the terminal includes a keyboard, a mouse, and a game device. It may include at least one manipulation unit, an analog unit, and a touch screen.

The client 110 is a user equipment, a mobile, a PC capable of wireless communication, a cell phone, a kiosk, a cellular phone, a cellular, a cellular terminal, a subscriber unit, a subscriber station, a mobile station, a terminal, a remote station, a PDA, a remote terminal, an access terminal, It may be referred to as, but not limited to, a user agent, a portable device having a wireless connection function, any device capable of using a wireless access mechanism, such as a wireless modem, and the like. The client 110 may also be referred to as a terminal or a peer.

The custom image generating apparatus 100 may include a game server together with a server for generating a custom image. The game server is connected to the client 110 , and may provide a service to the connected client 110 . In addition, the game server may store and manage various types of information of users who are registered as members by performing a membership registration procedure, and may provide purchase and payment functions for various game-related items. Also, the game server may relay execution data of a game application executed in each of a plurality of clients in real time so that a game match can be played between users.

In terms of hardware, such a server has the same configuration as a typical web server (Web Server) or WAP server (WAP Server). However, in terms of software, it includes program modules that perform various functions by being implemented through any language such as C, C++, Java, Visual Basic, and Visual C.

In addition, the server is generally connected to a large number of unspecified clients and/or other servers through an open computer network such as the Internet, and receives a request to perform work from a client or other server, derives the work result for it, and provides a computer system and It means computer software (server program) installed for him.

In addition, the server, in addition to the above-described server program, a series of application programs (Application Program) operating on the server and in some cases, various databases built inside or outside (DB: Database, hereinafter referred to as "DB") It should be understood as a broad concept that includes Accordingly, the server classifies member subscription information and various information and data about the game, stores and manages it in a DB, and this DB may be implemented inside or outside the server.

In addition, the server can be implemented using server programs that are provided in various ways according to operating systems such as DOS, Windows, Linux, UNIX, and Macintosh on general server hardware. As a representative example, a website used in a Windows environment, Internet Information Server (IIS), and CERN, NCSA, APPACH, etc. used in a Unix environment may be used.

The network refers to a connection structure capable of exchanging information between each node such as terminals and servers, or a network connecting the server and the clients 110 . Networks include the Internet, LAN (Local Area Network), Wireless LAN (Wireless Local Area Network), WAN (Wide Area Network), PAN (Personal Area Network), 3G, 4G, LTE, Wi-Fi, etc. It is not limited thereto. The network may be a closed type network such as a LAN or WAN, but is preferably an open type such as the Internet. The Internet is the TCP/IP protocol and several services that exist in the upper layer, namely HTTP (HyperText Transfer Protocol), Telnet, FTP (File Transfer Protocol), DNS (Domain Name System), SMTP (Simple Mail Transfer Protocol), SNMP ( It refers to a global open computer network structure that provides Simple Network Management Protocol), Network File Service (NFS), and Network Information Service (NIS). In addition, when the client 110 is a mobile communication terminal or a smart phone, the network may include a mobile communication network such as 3G, 4G, LTE, Wi-Fi, or the like.

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

2 is a flowchart illustrating an overall operation of a method for generating a customized image according to an exemplary embodiment.

According to an embodiment, in step 201 , the custom image generating apparatus 100 receives a 2D input image from a client. The user 120 may input a 2D input image to the client 110 and output a customized output image from the 2D input image through a simple operation. The client 110 may start a process of generating an output customizing image based on the motion input received from the user 120 . For example, the user 120 may input an operation input for generating an output custom image to the client 110 through a folding operation of the foldable terminal. As another example, the user 120 may input a motion input for generating an output customization image to the client 110 through a drag operation on the 2D input image displayed on the display.

In step 203 , the custom image generating apparatus 100 extracts a feature point from the 2D input image. The custom image generating apparatus 100 may extract information indicating characteristics of a person included in the 2D input image as a key point. A person included in the 2D input image may be distinguished from other persons through the feature point.

The customized image generating apparatus 100 may select a basic customized image from the database based on the feature points. The custom image generating apparatus 100 may generate a final 3D model based on the basic customized image and the feature point, and may provide one or more preset images using the final 3D model.

In operation 205 , the custom image generating apparatus 100 generates a basic 3D model using the feature points. For example, the custom image generating apparatus 100 may generate a 3D model on a zero basis based on the feature point. As another example, the custom image generating apparatus 100 may generate a modified 3D model by reflecting feature points in an existing 3D model.

In operation 207 , the customized image generating apparatus 100 selects a basic customized image from the database based on the feature points. The customized image generating apparatus 100 may select a basic customized image to be used to correct the basic 3D model from among existing customized images. The customized image generating apparatus 100 may select a basic customized image based on the feature point. The customized image generating apparatus 100 may select a basic customized image having a parameter similar to a feature point from among a plurality of basic customized images stored in the database.

In operation 209 , the customized image generating apparatus 100 modifies the basic 3D model based on the basic customized image to generate a final 3D model. The customized image generating apparatus 100 may generate a final 3D model by modifying the basic 3D model based on the parameters of the basic customized image. The custom image generating apparatus 100 may reflect the tendency of a user or a plurality of users in the basic 3D model by modifying the basic 3D model using the existing customized image. The custom image generating apparatus 100 may provide one or more preset images using the final 3D model.

In step 213, the client may select an output customization image from among one or more preset images generated from the final 3D model.

Thereafter, the user may select a desired preset image from one or more preset images. The user can create an output custom image by modifying the parameters of the selected preset image. The output customization image can be used for training of the recommendation module.

The 2D input image may include various parts of the body. For example, the 2D input image may include a person's face, torso, arms, or legs. The customized image generating apparatus 100 may generate a 3D output customized image by extracting feature points for each of various body parts of a person included in the 2D input image.

According to another embodiment, the customized image generating apparatus 100 may generate a 3D output customized image by synthesizing body parts of a person included in one or more 2D input images. The custom image generating apparatus 100 may receive one or more 2D input images corresponding to one or more body parts. The custom image generating apparatus 100 may extract each feature point from one or more 2D input images. The custom image generating apparatus 100 may generate a basic 3D model by combining feature points corresponding to one or more extracted 2D input images. Thereafter, the custom image generating apparatus 100 may modify the basic 3D model based on previously accumulated data and generate a plurality of preset images. The client 110 may select one preset image from among a plurality of preset images as the output customization image.

3 is a flowchart illustrating a process of generating a customized image according to an exemplary embodiment.

Referring to FIG. 3 , a system for generating an output custom image from a 2D input image includes a client 110 and a custom image generating apparatus 100 . The custom image generating apparatus 100 may include a feature point extracting unit 111 , a custom image generating unit 113 , a neural network 115 , and a database 117 .

The client 110 may transmit the 2D input image 121 to the custom image generating apparatus 100 . The client 110 may start a process of generating an output customizing image based on the motion input received from the user 120 . The feature point extraction unit 111 receives the 2D input image 121 from the client 110 .

The key point extractor 111 extracts a key point from the 2D input image 121 . The key point extractor 111 may extract information representing the characteristics of a person included in the 2D input image 121 as a key point. A person included in the 2D input image 121 may be distinguished from other persons through the feature point.

The feature point extractor 111 may transmit the extracted feature point to the custom image generator 113 and the database 117 . In step 123 , the keypoint extractor 111 may transmit the keypoint to the custom image generator 113 . In step 125 , the keypoint extractor 111 may transmit the keypoint to the database 117 . Steps 123 and 125 may be performed concurrently or concurrently.

The custom image generator 113 generates a basic 3D model using the feature points. For example, the custom image generator 113 may generate a 3D model from a zero base based on the feature point. As another example, the custom image generating apparatus 100 may generate a modified 3D model by reflecting feature points in an existing 3D model.

A basic customized image may be selected from the database 117 based on the feature points. A basic customized image 117 to be used for revising the basic 3D model may be selected from among the previously customized images from the database 117 . A basic customized image 117 having a parameter similar to a feature point may be selected from among a plurality of basic customized images stored in the database 117 .

The database 117 may transmit the basic customized image 117 to the customized image generator 113 . The customized image generating unit 113 generates a final 3D model by modifying the basic 3D model based on the basic customized image 117 . The custom image generator 113 may reflect the tendency of a user or a plurality of users in the basic 3D model by modifying the basic 3D model using the existing customized image 117 .

The custom image generator 113 may transmit the final 3D model 129 to the neural network 115 . The neural network 115 may generate the final 3D model 131 by inputting the final 3D model 129 . The neural network 115 may transmit the final 3D model 131 to the custom image generator 113 . The custom image generating apparatus 100 may improve the quality of the basic 3D model by modifying the basic 3D model using an existing pre-trained neural network.

The custom image generator 113 may generate one or more preset images from the final 3D model. The custom image generator 113 may transmit a plurality of preset images 133 to the client 110 .

4 is a diagram exemplarily illustrating a process of generating a custom image according to an embodiment.

In step 401, the client 110 may display the 2D input image on the display. For example, a 2D input image may be displayed on some areas of the display, and a basic 3D model reflecting default parameters may be displayed on other areas.

In step 403, the client 110 may receive a simple action input. The customizing process may be started by the client 110 receiving the motion input. For example, the client 110 may receive a folding operation or a drag operation.

In operation 405 , the client 110 may transmit the 2D input image to the custom image generating apparatus 100 . The custom image generating apparatus 100 may extract a feature point from the 2D input image.

In operation 407 , the custom image generating apparatus 100 may analyze a feature point for each element of a person's face included in the 2D input image, for example.

In operation 409 , the custom image generating apparatus 100 may convert the feature point into a parameter value for each element of the custom image. The custom image generating apparatus 100 may generate a basic 3D model by using parameter values for each element.

At step 413 , a base customizing image may be selected from a database. The basic customized image may be selected based on the feature point. A basic customizing image having parameters similar to the feature points may be selected.

In step 411, the base 3D model may be modified through the base custom image. The customized image generator 113 may correct the basic 3D model using the basic customized image and output the final 3D model.

In operation 417 , the custom image generating apparatus 100 may provide one or more preset images from the final 3D model and transmit the preset images to the client 110 . The user may select and modify a desired preset image among one or more preset images displayed on the client 110 to output an output customized image.

In step 415, the user's output customizing image may be used to learn the recommendation module.

5 is a diagram illustrating an overall configuration of an apparatus for generating a customized image according to an exemplary embodiment.

According to an embodiment, the custom image generating apparatus 100 includes at least one processor 501 and an I/O interface 503 . I/O interface 503 receives a 2D input image from a client. The processor 501 may extract a feature point from the 2D input image. The processor 501 may generate a basic 3D model using the feature points. The processor 501 may select a basic customized image based on the feature point from the database. The processor 501 may generate a final 3D model by modifying the basic 3D model based on the basic customized image. The processor 501 may generate the final 3D model by modifying the final 3D model using the neural network.

The I/O interface 503 may transmit one or more preset images generated from the final 3D model to the client. The I/O interface 503 may receive an output customizing image selected from among one or more preset images.

6 is a diagram illustrating an overall configuration of a client according to an embodiment.

According to one embodiment, the client 110 includes a processor 601 , a display 603 , and an I/O interface 605 . The display 603 may include a first surface 607 and a second surface 609 . The display 603 may include a foldable display. The I/O interface 605 may receive one or more preset images from the custom image generating apparatus 100 . The I/O interface 605 may transmit a selection signal of an output custom image among one or more preset images to the custom image generating device. The preset image may be generated by the custom image generating apparatus 100 from a 2D input image, a basic custom image stored in a database, and a 3D model generated based on a neural network.

7 is an example of generating a customized image of a client according to an embodiment.

According to an embodiment, in step 710 , the first side of the client 110 may display a symbol 713 for inputting a 2D input image. The client 110 may receive the 2D input image 711 . The first side of the client 110 may display a 2D input image. When the folding sensor detects an unfolded state, the first surface may display a 2D input image 721 . The second surface of the client 110 may display a default parameter 717 and a 3D shape 715 reflecting it.

In step 720 , the user may input an action input to the client 110 . For example, the motion input may include a folding/unfolding operation. The folding sensor may detect a folding state and an unfolding state of the first surface and the second surface. When the folding sensor detects the folding state, the I/O interface may send a 2D input image to the customizing image generating device.

In step 730, the I/O interface may receive one or more preset images from the custom image generating device. A 2D input image 721 may be displayed on the first surface. A plurality of preset images 731 , 733 , 735 , 737 and parameters 732 , 734 , 736 , and 738 may be displayed on the second surface. The user can select an output custom image from among one or more preset images. The I/O interface may send a selection signal to the custom image generating device.

8 is an example of generating a customized image of a client according to another embodiment.

According to an embodiment, in step 710 , the first side of the client 110 may display a symbol 713 for inputting a 2D input image. The client 110 may receive the 2D input image 711 . The first side of the client 110 may display a 2D input image. The second surface of the client 110 may display a default parameter 717 and a 3D shape 715 reflecting the default parameter 715 .

In step 720 , the user may input an action input to the client 110 . For example, the motion input may include a drag motion. When the touch sensor detects a drag operation of the 2D input image in a direction from the first surface to the second surface, the I/O interface may transmit the 2D input image to the customizing image generating device.

In step 730, the I/O interface may receive one or more preset images from the custom image generating device. A 2D input image 721 may be displayed on the first surface. A plurality of preset images 731 , 733 , 735 , and 737 and parameters 732 , 734 , 736 , and 738 may be displayed on the second surface. The user can select an output custom image from among one or more preset images. The I/O interface may send a selection signal to the custom image generating device.

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 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...Custom Image Generator 110...Client

Claims (13)

receiving a 2D input image from a client;
extracting feature points from the 2D input image;
selecting a basic customized image from a database based on the feature point; and
generating a final 3D model based on the basic customized image and the feature point, and providing one or more preset images using the final 3D model
Including,
the client,
modifying parameters of a preset image selected from among the one or more preset images based on a user's input,
The operation of selecting the basic customized image is,
selecting a basic customized image having parameters similar to the characteristic point from among a plurality of basic customized images stored in the database,
The operation of providing the preset image is,
A custom image generating method for generating a basic 3D model based on the feature points, and for generating the final 3D model by modifying the generated basic 3D model using the basic customized image. delete The method of claim 1,
The operation of providing the preset image is,
generating the final 3D model based on the parameters of the basic customized image,
How to create a custom image. The method of claim 1,
The operation of providing the preset image is,
Input parameters of the final 3D model to a neural network to generate a final 3D model based on the output parameters,
How to create a custom image. The method of claim 1,
The operation of receiving the 2D input image is,
receiving one or more 2D input images corresponding to one or more body parts;
The operation of extracting the feature point is
extracting each feature point from the one or more 2D input images,
The operation of providing the preset image is,
generating the final 3D model by combining feature points corresponding to the extracted one or more 2D input images,
How to create a custom image. The method of claim 1,
The method further comprising the operation of learning a recommendation module by using the final customized image selected and modified from the one or more preset images by a user input. At least one processor; and
comprising an I/O interface to communicate with the client;
The I/O interface receives a 2D input image from the client,
The processor,
extracting feature points from the 2D input image,
selecting a basic customized image based on the feature point from the database,
A final 3D model is generated based on the basic customized image and the feature point, and one or more preset images are provided using the final 3D model,
the client,
modifying parameters of a preset image selected from among the one or more preset images based on a user's input;
The processor,
selecting a basic customized image having parameters similar to the feature point from among a plurality of basic customized images stored in the database,
A customized image generating apparatus for generating a basic 3D model based on the feature points, and for generating the final 3D model by modifying the generated basic 3D model using the basic customized image. delete The method of claim 7,
The processor,
generating the final 3D model based on the parameters of the basic customized image,
Custom image generator. The method of claim 7,
The processor,
Input parameters of the final 3D model to a neural network to generate a final 3D model based on the output parameters,
Custom image generator. The method of claim 7,
The I/O interface is
receiving one or more 2D input images corresponding to one or more body parts;
The processor,
extracting each feature point from the one or more 2D input images,
generating the final 3D model by combining feature points corresponding to the extracted one or more 2D input images,
Custom image generator. At least one processor;
a display comprising a first side and a second side;
a folding sensor for detecting a folding state and an unfolding state of the first surface and the second surface; and
an I/O interface for communicating with a custom image generating device;
When the folding sensor detects an unfolded state, the first surface displays a 2D input image,
When the folding sensor detects the folding state, the I/O interface sends the 2D input image to the customizing image generating device,
The I/O interface receives one or more preset images from the customizing image generating device, and transmits a selection signal of an output customizing image among the one or more preset images to the customizing image generating device,
The preset image is generated from the 3D model generated based on the 2D input image, the basic customized image stored in the database, and the neural network,
The at least one processor,
Modifying a parameter of a preset image selected from among the one or more preset images based on a user's input to generate the output customized image,
The custom image generating device,
Selecting a basic customized image having parameters similar to feature points from among a plurality of basic customized images stored in the database,
A client that generates a basic 3D model based on the feature points, and modifies the generated basic 3D model using the basic customized image to generate a final 3D model. At least one processor;
a display comprising a first side and a second side;
a touch sensor for sensing a touch of the first surface and the second surface; and
an I/O interface for communicating with a custom image generating device;
The first side displays a 2D input image,
When the touch sensor detects a drag operation of the 2D input image in a direction from the first surface to the second surface, the I/O interface transmits the 2D input image to the custom image generating device,
The I/O interface receives one or more preset images from the customizing image generating device, and transmits a selection signal of an output customizing image among the one or more preset images to the customizing image generating device,
The preset image is generated from the 3D model generated based on the 2D input image, the basic customized image stored in the database, and the neural network,
The at least one processor,
Modifying a parameter of a preset image selected from among the one or more preset images based on a user's input to generate the output customized image,
The custom image generating device,
Selecting a basic customized image having parameters similar to feature points from among a plurality of basic customized images stored in the database,
A client that generates a basic 3D model based on the feature points, and modifies the generated basic 3D model using the basic customized image to generate a final 3D model.

Images