Method and system for generating texture by artificial intelligence

Abstract

Disclosed is a method for generating texture by artificial intelligence. The method for generating texture by artificial intelligence according to an embodiment of the present disclosure may comprise the steps of: transmitting a prompt input by a user to a server; inputting, by the server, the prompt into an artificial neural network, generating a first image corresponding to the prompt on the basis of an output of the artificial neural network for the input of the prompt, and transmitting the first image to a computing system; generating a first texture corresponding to the first image; and in response to a first user input related to a first object existing in a first virtual world and a second user input related to the first image, applying the first texture to the first object while performing the operation of displaying at least a part of the first virtual world on a screen.

Description

Method and system for generating textures using artificial intelligence

[0001] A method for generating a texture using artificial intelligence and a system applying the method are disclosed. More specifically, the present disclosure relates to a method for generating a texture associated with a prompt entered by a user using artificial intelligence and a computing system applying the method.

[0002] Recently, in the video game industry, the 'customizing' function, which allows gamers to modify the world they play in to their own taste, has become popular in order to increase gamers' immersion in the game.

[0003] When a game features customization, gamers can customize the appearance of characters and the shape of props within the game world to their personal taste. When modifying the shape of a specific prop, gamers often modify its color or texture. In other words, when modifying the shape of a specific prop, gamers must change the texture applied to that prop.

[0004] Conventional methods for changing textures include a method where the gamer selects one of several presets prepared in advance by the game developer, a method where the gamer creates the texture himself and applies it to each prop, and a method where the gamer applies a third-party asset.

[0005] However, having gamers select one of multiple presets prepared by the game developer can be difficult to immerse players in the game if no presets match their preferences. Furthermore, creating textures and applying them to individual props requires gamers to possess expert-level skills in the relevant field. Furthermore, applying third-party assets poses the problem of difficulty in enhancing player immersion if no assets that match the player's preferences exist within the platform that provides these assets.

[0006] In addition, the conventional texture change methods described above had the inconvenience of requiring the gamer to quit and restart the game after applying a new texture to the prop, other than selecting one of the presets prepared in advance by the game developer.

[0007] In order to resolve the inconvenience of the above-mentioned users, a method of applying a new texture to an object by reloading only the loaded virtual world without restarting the game program for the game has been provided in the past, but the method of reloading the virtual world also causes a waste of time cost corresponding to the method of restarting the game.

[0008] Therefore, there is a need for a method that automatically generates a texture to be applied to a prop that the gamer wants to customize according to the gamer's command and immediately applies the texture to the prop without going through the process of restarting the game, but the conventional texture change method does not provide such technology.

[0009] A technical problem to be achieved through some embodiments of the present disclosure is to provide a method for generating a texture based on an image obtained based on a prompt input by a user.

[0010] Another technical problem to be achieved through some embodiments of the present disclosure is to provide a method for optimizing client terminal storage space when storing textures generated based on user-entered prompts.

[0011] Another technical problem to be achieved through some embodiments of the present disclosure is to provide a method for changing the texture of an object displayed on a screen during runtime of an application.

[0012] Another technical challenge to be achieved through some embodiments of the present disclosure is to provide a method for changing the texture of an object without requiring reloading the virtual world.

[0013] Another technical problem to be achieved through some embodiments of the present disclosure is to provide a method for obtaining a normal map of a texture based on an albedo map of the texture.

[0014] The technical problems of the present disclosure are not limited to the technical problems mentioned above, and other technical problems not mentioned will be clearly understood by those skilled in the art from the description below.

[0015] In order to solve the above technical problem, a computing system according to an embodiment of the present disclosure may include one or more processors; and a memory storing a computer program executed by the one or more processors. The computer program may include instructions that cause the processor to perform an operation of transmitting a prompt input by a user to a server, an operation of receiving a first image corresponding to the prompt from the server, an operation of generating a first texture corresponding to the first image, and an operation of applying the first texture to a first object while performing an operation of displaying at least a portion of the first virtual world on a screen in response to a first user input related to a first object existing in the first virtual world and a second user input related to the first image. The first image may be generated by the server inputting the prompt to an artificial neural network and based on an output of the artificial neural network in response to the prompt input.

[0016] In some embodiments, the operation of applying the first texture to the first object may include an operation of changing a parameter value corresponding to an albedo map of the first object to a value corresponding to the albedo map of the first texture and an operation of changing a parameter value corresponding to a normal map of the first object to a value corresponding to the normal map of the first texture.

[0017] In some embodiments, the prompt may include a negative prompt. The artificial neural network may not output a texture associated with the negative prompt.

[0018] In some embodiments, the first texture may be an asset that does not exist at the time the first virtual world is loaded into the memory.

[0019] In some embodiments, the operation of generating a first texture corresponding to the first image may include the operation of generating an albedo map associated with the first image and the operation of generating a normal map corresponding to the albedo map associated with the first image based on the albedo map associated with the first image.

[0020] In some embodiments, the operation of generating a first texture corresponding to the first image may further include the operation of storing only an albedo map associated with the first image among parameters related to the first texture corresponding to the generated first image.

[0021] In some embodiments, the operation of generating a normal map corresponding to the albedo map associated with the first image based on the albedo map associated with the first image may include the operation of generating the normal map based on a value of each pixel included in the albedo map associated with the first image and a value of an adjacent pixel of each of the pixels.

[0022] In some embodiments, the computer program may further include instructions that cause the processor to perform the following actions: displaying on a screen an interface including thumbnails of images corresponding to each of the generated plurality of textures; deleting a thumbnail corresponding to the user input from the interface in response to a user input for any one of the thumbnails; and changing the texture of one or more objects to which the texture corresponding to the thumbnail is applied to a default texture.

[0023] According to another embodiment of the present disclosure for solving the above-described technical problem, a computing system may include one or more processors and a memory storing a computer program executed by the one or more processors. The computer program may include instructions that cause the processor to perform an operation of identifying a third image based on a third user input associated with the third image, an operation of generating a second texture associated with the third image, and an operation of applying the second texture to a second object while performing an operation of displaying at least a portion of the second virtual world on a screen in response to the third user input and a fourth user input associated with a second object existing in the second virtual world. The third image may be an image previously stored in the computing system.

[0024] In some embodiments, the operation of generating the second texture associated with the third image may include the operation of generating an albedo map associated with the third image, and the operation of generating a normal map corresponding to the albedo map associated with the third image based on the albedo map associated with the third image.

[0025] In some embodiments, the act of applying the second texture to the second object while performing the act of displaying at least a portion of the second virtual world on the screen may include the act of changing a parameter value corresponding to an albedo map of the second object to a value corresponding to the albedo map of the second texture and the act of changing a parameter value corresponding to a normal map of the second object to a value corresponding to the normal map of the second texture.

[0026] In some embodiments, the operation of generating a normal map corresponding to the albedo map associated with the third image based on the albedo map associated with the third image may include the operation of generating the normal map based on a value of each pixel included in the albedo map associated with the third image and a value of an adjacent pixel of each of the pixels.

[0027] In some embodiments, the second texture may be an asset that does not exist at the time the second virtual world is loaded into the memory.

[0028] According to another embodiment of the present disclosure for solving the above-described technical problem, a method for generating a texture by artificial intelligence may include the steps of: transmitting a prompt input by a user to a server; the server inputting the prompt to an artificial neural network, generating a first image corresponding to the prompt based on an output of the artificial neural network in response to the prompt input, and transmitting the first image to the computing system; generating a first texture corresponding to the first image; and applying the first texture to the first object while performing an operation of displaying at least a part of the first virtual world on a screen in response to a first user input related to a first object existing in the first virtual world and a second user input related to the first image.

[0029] In some embodiments, the step of applying the first texture to the first object may include the step of changing a parameter value corresponding to an albedo map of the first object to a value corresponding to the albedo map of the first texture and the step of changing a parameter value corresponding to a normal map of the first object to a value corresponding to the normal map of the first texture.

[0030] In some embodiments, the first texture may be an asset that does not exist at the time the first virtual world is loaded into the memory.

[0031] In some embodiments, the step of generating a first texture corresponding to the first image may include the step of generating an albedo map associated with the first image and, based on the albedo map associated with the first image, generating a normal map corresponding to the albedo map associated with the first image.

[0032] In some embodiments, the step of generating a first texture corresponding to the first image may further include the step of storing only an albedo map associated with the first image among parameters related to the first texture corresponding to the generated first image.

[0033] In some embodiments, the step of generating a normal map corresponding to the albedo map associated with the first image based on the albedo map associated with the first image may include the step of generating the normal map based on a value of each pixel included in the albedo map associated with the first image and a value of adjacent pixels of each of the pixels.

[0034] In some embodiments, the method may further include identifying a third image based on a third user input associated with the third image, generating a second texture associated with the third image, and applying the second texture to a second object while performing an operation of displaying at least a portion of the second virtual world on a screen in response to the third user input and a fourth user input associated with a second object existing in the second virtual world. The third image may be an image previously stored in the computing system. The second texture may be an asset that does not exist at the time of loading the second virtual world into the memory.

[0035] According to the various embodiments of the present invention described above, the user inconvenience caused by application restarts or world reloading required to create and apply new textures can be alleviated. According to the various embodiments of the present invention described above, a method for solving the storage capacity issue of a computing system that stores textures can be provided.

[0036] According to various embodiments of the present invention described above, the user can obtain the effect of increasing immersion in the virtual world by applying a desired texture to a specific object without exiting the virtual world.

[0037] According to the various embodiments of the present invention described above, the effect of eliminating the inconvenience of a user having to individually reset an object with a texture applied that the user no longer wishes to exist in the virtual world can be achieved.

[0038] The effects of the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the description below.

[0039] FIG. 1 is a diagram illustrating an exemplary environment to which a client terminal according to one embodiment of the present disclosure can be applied.

[0040] FIG. 2 is a flowchart of a method for generating texture using artificial intelligence according to another embodiment of the present disclosure.

[0041] FIG. 3 is a diagram illustrating a step of transmitting a user-entered prompt to an image generation system that may be performed in some embodiments of the present disclosure.

[0042] FIG. 4 is a diagram illustrating a step of receiving an image from an image generation system that may be performed in some embodiments of the present disclosure.

[0043] FIG. 5 is a diagram illustrating a step of generating a texture associated with a received image that may be performed in some embodiments of the present disclosure.

[0044] FIG. 6 is a diagram illustrating a step of receiving a selection input for a specific texture that may be performed in some embodiments of the present disclosure.

[0045] FIG. 7 is a diagram illustrating a step of receiving a selection input for a specific object that may be performed in some embodiments of the present disclosure.

[0046] FIG. 8 is a diagram illustrating an object with a new texture applied that can be displayed on a screen as a result of performing some embodiments of the present disclosure.

[0047] FIG. 9 is a diagram illustrating a step of changing the texture of an object to which a specific texture is applied to a default texture in response to an input for deleting a specific texture, which may be performed in some embodiments of the present disclosure.

[0048] FIG. 10 is a flowchart of a method for generating texture using artificial intelligence according to another embodiment of the present disclosure.

[0049] FIG. 11 is a diagram illustrating a step of identifying a third image that may be performed in some embodiments of the present disclosure.

[0050] FIG. 12 is a hardware configuration diagram of a computing system according to another embodiment of the present disclosure.

[0051] Hereinafter, preferred embodiments of the present disclosure will be described in detail with reference to the attached drawings. The advantages and features of the present invention, and methods for achieving them, will become clearer with reference to the embodiments described in detail below together with the attached drawings. However, the technical spirit of the present invention is not limited to the following embodiments and may be implemented in various different forms. The following embodiments are provided only to complete the technical spirit of the present invention and to fully inform those skilled in the art of the present invention of the scope of the present invention, and the technical spirit of the present invention is defined only by the scope of the claims.

[0052] In describing the present disclosure, if it is determined that a detailed description of a related known configuration or function may obscure the gist of the present invention, the detailed description will be omitted.

[0053] Unless otherwise defined, the terms (including technical and scientific terms) used in the following examples may be used with meanings commonly understood by those of ordinary skill in the art to which this disclosure pertains; however, this may vary depending on the intentions of engineers working in the relevant field, precedents, the emergence of new technologies, etc. The terminology used in this disclosure is for the purpose of describing the embodiments and is not intended to limit the scope of this disclosure.

[0054] In the following examples, singular expressions include plural concepts unless the context clearly specifies that they are singular. Furthermore, plural expressions include singular concepts unless the context clearly specifies that they are plural.

[0055] In addition, terms such as first, second, A, B, (a), (b), etc. used in the following embodiments are only used to distinguish certain components from other components, and the nature, order, or sequence of the components are not limited by the terms.

[0056] Before describing various embodiments of the present disclosure, the terms used in the following embodiments will be clarified.

[0057] In the following embodiments, 'prompt' may mean text related to a characteristic of a specific texture that a user inputs into the system to obtain the specific texture.

[0058] In the following examples, the 'albedo map' may be understood as one of the components of a texture in an object rendering method. More specifically, the 'albedo map' may refer to a map containing color value information of a specific texture. In the relevant technical field, the 'albedo map' may be used interchangeably with terms such as 'base color'.

[0059] In the following embodiments, a 'normal map' may be understood as one of the components of a texture in an object rendering method. More specifically, a 'normal map' may refer to a map that stores the degree to which each pixel is influenced by a light source as a value based on the normal vector of each pixel included in the texture. In the relevant technical field, the 'normal map' may be used interchangeably with terms such as 'normal map'.

[0060] Hereinafter, some embodiments of the present disclosure will be described with reference to the drawings.

[0061] FIG. 1 is a diagram illustrating an exemplary environment to which a client terminal (200) according to one embodiment of the present disclosure can be applied.

[0062] Each component illustrated in Figure 1 may represent software or hardware such as a Field Programmable Gate Array (FPGA) or an Application-Specific Integrated Circuit (ASIC). However, the components are not limited to software or hardware, and may be configured to reside on an addressable storage medium or configured to execute one or more processors. The functions provided within the components may be implemented by more detailed components, or multiple components may be combined to form a single component that performs a specific function.

[0063] The client terminal (200) illustrated in FIG. 1 may be a notebook, desktop, laptop, smartphone, tablet, etc., but is not limited thereto and may include all types of devices equipped with computing functions.

[0064] In some embodiments, the image generation system (100) illustrated in FIG. 1 may communicate with other components via a network. The network may be implemented as any type of wired / wireless network, such as a Local Area Network (LAN), a Wide Area Network (WAN), a mobile radio communication network, or Wibro (Wireless Broadband Internet).

[0065] The image generation system (100) and client terminal (200) may be understood to operate according to a server-client model. However, in some embodiments, the system may be configured in a stand-alone client manner, without the need for a server. In this case, the operations performed by the image generation system (100) may be understood to be performed on the client terminal (200).

[0066] A client terminal (200) according to one embodiment of the present disclosure can transmit a prompt input by a user to an image generation system (100).

[0067] In some embodiments of the present disclosure, the image generation system (100) may receive a prompt from a client terminal (200).

[0068] In some other embodiments of the present disclosure, the prompt may include a negative prompt.

[0069] In some other embodiments of the present disclosure, the image generation system (100) can input the prompt to an artificial neural network (ANN) and obtain an image based on an output of the artificial neural network for the prompt input.

[0070] According to one embodiment, an artificial neural network can be subjected to supervised, semi-supervised, or unsupervised learning based on multiple learning data (i.e., a training set). For example, a training data set including multiple positive prompts, multiple negative prompts, and multiple correct answer data can be applied to an artificial neural network, so that an image (or feature data) output from the artificial neural network based on the positive / negative prompts becomes similar / identical to a texture (or feature data) included in the correct answer data. When the artificial neural network is repeatedly trained, the weights of each node included in the artificial neural network can converge to an optimal value.

[0071] A client terminal (200) according to another embodiment of the present disclosure can receive the image from the image generation system (100) and generate a texture associated with the received image.

[0072] In some embodiments of the present disclosure, the texture may include at least one of an albedo map, a normal map, a height map, an occlusion map, a bump map, a displacement map, a roughness map, a metalness map, a specular map, an opacity map, and a refraction map.

[0073] In some other embodiments of the present disclosure, the client terminal (200) can perform the operation of generating the texture while running the texture generation application (300).

[0074] In some further embodiments of the present disclosure, the texture generation application (300) may be a video game program.

[0075] In some further embodiments of the present disclosure, the client terminal (200) may receive a selection input for a specific object from the user. Here, the specific object may refer to an object included in a virtual world loaded as a result of the client terminal (200) running the texture generation application (300).

[0076] In some further embodiments of the present disclosure, the client terminal (200) may generate an albedo map associated with the received image.

[0077] In some other embodiments of the present disclosure, the client terminal (200) may generate a normal map associated with the image based on an albedo map associated with the received image, as will be described in detail later.

[0078] Hereinafter, to help understand some embodiments of the present disclosure, a dataset including information of an albedo map associated with the image and information of a normal map associated with the image may be exemplified as a texture corresponding to the image. However, this is merely an example to help understand some embodiments of the present disclosure, and it should be understood that a texture according to some embodiments of the present disclosure, as exemplified in the above-described embodiments, includes at least one of an albedo map, a normal map, a height map, an occlusion map, a bump map, a displacement map, a roughness map, a metalness map, a specular map, an opacity map, and a refraction map.

[0079] A client terminal (200) according to another embodiment of the present disclosure can receive a selection input for a specific image from a user.

[0080] According to another embodiment of the present disclosure, a client terminal (200) may apply a texture related to the specific image to the specific object while performing an operation of displaying at least a portion of the virtual world on a screen in response to a user input related to the specific object existing in the virtual world and a user input related to the specific image.

[0081] In some embodiments of the present disclosure, the client terminal (200) can change a parameter value corresponding to an albedo map of the specific object to a value corresponding to an albedo map of a texture associated with the specific image.

[0082] In some other embodiments of the present disclosure, the client terminal (200) can change a parameter value corresponding to a normal map of the specific object to a value corresponding to a normal map of a texture associated with the specific image.

[0083] In some further embodiments of the present disclosure, the texture of the particular object may be understood as parameterized.

[0084] In some further embodiments of the present disclosure, the texture of the specific object may be understood as being stored as an object. That is, the albedo map and normal map of the texture of the specific object may each be configured as separate instances. Accordingly, the shape of the specific object appearing in the virtual world may be changed simply by changing the instance properties of the albedo map and normal map corresponding to the texture of the specific object.

[0085] In the case of a conventional texture change method, a specific texture file is composed of a single data in which an albedo map, a normal map, a height map, etc. are configured as parameters, so that the texture applied to an object cannot be changed by changing a specific parameter value included in the texture. For example, the world had to be reloaded with the file of the first texture replaced with the file of the second texture so that the texture desired by the user could be applied to the object. According to the present embodiment, by instantiating some properties of the texture, the effect of resolving the user's inconvenience of restarting the application or reloading the world required for creating and applying a new texture can be achieved.

[0086] So far, with reference to FIG. 1, the components included in an exemplary environment to which a client terminal (200) can be applied and the operations that these components can perform have been described. It should be understood that the embodiments described above are exemplary in all respects and not limiting. Furthermore, the configuration and operation of the client terminal (200) according to this embodiment may be supplemented by several embodiments described below.

[0087] Hereinafter, a method for generating a texture using artificial intelligence according to another embodiment of the present disclosure will be described with reference to FIGS. 2 to 9. It should be understood that the steps described in the flowcharts below are performed by the client terminal (200) described with reference to FIG. 1 unless otherwise stated. Furthermore, it should be understood that the technical concepts understood in the above-described embodiment with reference to FIG. 1 can be readily applied to the method for generating a texture using artificial intelligence according to the present embodiment.

[0088] FIG. 2 is a flowchart of a method for generating texture using artificial intelligence according to another embodiment of the present disclosure.

[0089] In step S100, the client terminal (200) can transmit a prompt entered by the user to the image generation system (100).

[0090] In some embodiments related to step S100, referring to FIG. 3, the image generation system (100) may receive a prompt (31) and a negative prompt (31-1) input by a user into a user interface (30) for prompt input displayed on the screen of the client terminal (200) from the client terminal (200).

[0091] In some other embodiments related to step S100, the prompt (31) shown in FIG. 3 may be text related to a feature of an image that the user wishes to acquire, and the negative prompt (31-1) may be text related to a feature of an image that the user wishes not to include in the image to be acquired.

[0092] In step S200, the client terminal (200) can receive an image associated with the prompt transmitted by the client terminal (200) to the image generation system (100) in step S100 from the image generation system (100).

[0093] In some embodiments related to step S200, the image generation system (100) may input the prompt received in step S100 into an artificial neural network, and generate an image based on the output of the artificial neural network in response to the prompt input. In this case, referring to FIG. 5, the artificial neural network may be configured to output an image (51) reflecting the features of the prompt, or output feature data related to the image (51).

[0094] Meanwhile, if the prompt includes a negative prompt, the artificial neural network can be configured to output a texture that reflects the characteristics of the positive prompt but does not reflect the characteristics of the negative prompt. Here, the positive prompt may be a prompt excluding the negative prompt among all prompts.

[0095] In some other embodiments related to step S200, referring to FIGS. 3 and 4, the client terminal (200) may receive from the image generation system (100) a first image (41) generated by the image generation system (100) based on the prompt (31) and the negative prompt (31-1) received from the client terminal (200).

[0096] In some other embodiments related to step S200, referring to FIG. 4, when the client terminal (200) receives an input for a save button (42) from a user, the first image (41) may be stored in the client terminal (200). Here, the first image (41) may be stored in a directory corresponding to the texture generation application (300) described with reference to FIG. 1.

[0097] In some other embodiments related to step S200, the image generation system (100) can transmit the generated image to the client terminal (200).

[0098] According to one embodiment, an artificial neural network can be subjected to supervised, semi-supervised, or unsupervised learning based on multiple learning data (i.e., a training set). For example, a training data set including multiple positive prompts, multiple negative prompts, and multiple correct answer data can be applied to an artificial neural network, so that the artificial neural network can be subjected to supervised learning such that the texture (or feature data) output from the artificial neural network based on the positive / negative prompts becomes similar / identical to the texture (or feature data) included in the correct answer data. When the artificial neural network is repeatedly trained, the weights of each node included in the artificial neural network can converge to an optimal value.

[0099] In some embodiments, the artificial neural network may include a language model for extracting features for the prompt. For example, the artificial neural network may include a language model such as a Generative Pre-training Transformer (GPT) and a Bidirectional Encoder Representations from Transformers (BERT). As another example, the artificial neural network may include a model for generating images, such as a Generative Adversarial Network (GAN) or a Variational Autoencoder (VAE).

[0100] In step S200-1, the client terminal (200) can generate a texture related to the image received from the image generation system (100) in step S200.

[0101] In some embodiments related to step S200-1, the client terminal (200) can generate a texture related to the image received from the image generation system (100) in step S200.

[0102] In some embodiments related to step S200-1, the texture generated by the client terminal (200) may mean a texture file in a texture file format that includes all texture-related components. Here, the texture-related components may be clearly understood by referring to some embodiments described above with reference to FIG. 1. In addition, the texture file format is not limited to any file format that encodes and compresses conventional texture information.

[0103] In some other embodiments related to step S200-1, the texture generated by the client terminal (200) may be generated after the client terminal (200) compiles the first virtual world and loads it into the memory of the client terminal (200). That is, the texture generated by the client terminal (200) may be an asset that does not exist at the time the client terminal (200) loads the first virtual world into the memory.

[0104] In some other embodiments related to step S200-1, referring to FIG. 5, the client terminal (200) can generate an albedo map (41-1) of a first texture corresponding to the first image (41) based on color information of each pixel of the first image (41) received from the image generation system (100) in step S200.

[0105] In some other embodiments related to step S200-1, referring to FIG. 5, the client terminal (200) can obtain a normal map (41-2) of the first texture based on the albedo map (41-1) of the user's first texture.

[0106] In some other embodiments related to step S200-1, the client terminal (200) may generate a normal map corresponding to the albedo map based on a specific directional color change of the albedo map.

[0107] For example, referring to FIG. 5, the client terminal (200) can determine the RGB matrix value of the second pixel (41-1b) on the normal map (41-2) of the first texture associated with the albedo map (41-1) of the first texture based on the pixel value of the second pixel (41-1b) on the albedo map (41-1) of the first texture and the pixel values ​​of the first pixel (41-1a) and the third pixel (41-1c) which are adjacent pixels of the second pixel (41-1b).

[0108] For another example, referring to FIG. 5, it can be understood that the color of the pixels becomes closer to black as one proceeds from the first pixel (41-1a) to the third pixel (41-1c). In this case, the client terminal (200) can determine that the third pixel (41-1c) is an area that receives less light than the first pixel (41-1a) on the normal map (41-2) of the first texture. In addition, it can be understood that the second pixel (41-1b) receives less light than the first pixel (41-1a), but more than the third pixel (41-1c).

[0109] In some other embodiments related to step S200-1, referring to FIG. 5, the client terminal (200) may generate a first texture corresponding to a first image (41), but store only an albedo map (41-1) of the first texture.

[0110] A texture file is a file with a large capacity, including color information (albedo map), shading information (normal map), height information (height map), and light reception information (normal map) for each pixel of the texture. When the client terminal (200) stores the above-mentioned texture file, there may be an inconvenience that the storage space of the client terminal (200) is wasted. According to the present embodiment, the client terminal (200) may achieve the effect of saving the storage space of the client terminal (200) by storing only the albedo map of the texture.

[0111] In addition, according to the above-described embodiment, it can be understood that the client terminal (200) according to some embodiments of the present disclosure generates a normal map of the specific texture based on the albedo map of the specific texture whenever it receives a user's selection input for applying the specific texture to a specific object.

[0112] In some other embodiments related to step S200-1, referring to FIG. 6, the client terminal (200) may display on the screen a texture selection interface (61) including thumbnails of each of a plurality of images generated based on a user's prompt input.

[0113] Here, the texture selection interface (61) is exemplified as displaying a thumbnail of an image, but referring to the above-described embodiments, it can be understood that an albedo map of a texture corresponding to each image included in the texture selection interface (61) is pre-stored in the client terminal (200).

[0114] In some other embodiments related to step S200-1, referring to FIGS. 4 and 6, when the client terminal (200) receives a user input for the save button (42) illustrated in FIG. 4, the client terminal (200) may refresh the texture selection interface (61) so that the thumbnail (62) of the first image (41) is included in the texture selection interface (61).

[0115] In some other embodiments related to step S200-1, referring to FIGS. 5 and 6, when the client terminal (200) receives a user's selection input for a thumbnail (62) of a first image (41) included in a texture selection interface (61), the client terminal (200) may perform an operation of generating an albedo map (41-1) of a first texture corresponding to the first image (41) and a normal map (41-2) of the first texture.

[0116] In some other embodiments related to step S200-1, referring to FIGS. 5 and 6, when an albedo map (41-1) of a first texture corresponding to a first image (41) is previously stored in a client terminal (200), the client terminal (200) can generate a normal map (41-2) of a first texture corresponding to the first image (41) based on the albedo map of the first texture (41-1) in response to a user's selection input for a thumbnail (62) of the first image (41) included in a texture selection interface (61).

[0117] Hereinafter, the explanation will continue with reference to Fig. 2.

[0118] In step S300, the client terminal (200) may apply a texture related to the specific image to the first object while performing an operation of displaying at least a portion of the first virtual world on the screen in response to a first user input related to a first object existing in the first virtual world and a second user input related to a specific image.

[0119] In some embodiments related to step S300, when a client terminal (200) receives a user input while running a texture generation application (300), the client terminal (200) can load a first virtual world included in the texture generation application (300) into memory and display the first virtual world on the screen.

[0120] In some other embodiments related to step S300, referring to a part of the first virtual world illustrated in FIG. 7, the client terminal (200) can receive a selection input from the user for a chair body (71) among a plurality of objects included in the first virtual world.

[0121] In some further embodiments related to step S300, the plurality of objects included in the first virtual world may include characters, special effects, cameras, light sources, and items. Furthermore, the plurality of objects may include entities with specific textures, specific hit boxes, and predefined animations applied.

[0122] In some other embodiments related to step S300, referring to a part of the first virtual world illustrated in FIGS. 5 and 8, the client terminal (200) displays a predefined affordance on a screen on which the first virtual world is displayed in response to receiving a selection input for a chair body (71) from a user, and in response to receiving a selection input for a thumbnail (62) of a first image (41) in a texture selection interface (61), generates an albedo map (41-1) of a first texture corresponding to the first image (41), generates a normal map (41-2) of the first texture based on the albedo map (41-1) of the first texture, and changes a parameter value corresponding to the albedo map of the default texture of the chair body (71) to a value corresponding to the albedo map (41-1) of the first texture, and changes the parameter value corresponding to the albedo map of the default texture of the chair body (71) to a value corresponding to the albedo map (41-1) of the first texture. By changing the parameter value corresponding to the normal map of the texture to a value corresponding to the normal map (41-2) of the first texture, the chair body (71-1) to which the first texture is applied can be displayed on the screen.

[0123] Here, it can be understood that the operation of the client terminal (200) according to the above-described embodiment to change the texture of the chair body (71) to the first texture is performed in a state where the client terminal (200) does not re-compile the first virtual world.

[0124] In the case of a conventional method of changing the texture of an object included in a virtual world, there was a problem in that after a user sets a desired texture to be applied to a specific object in a specific virtual world, the specific virtual world must be reloaded in order for the desired texture to be applied to the specific object. According to the present embodiment, since the user can apply a desired texture to a specific object without exiting the virtual world, the user can obtain the effect of increasing immersion in the virtual world.

[0125] According to this embodiment, a texture with a shading effect can be applied to an object using only an albedo map stored in the client terminal (200). In other words, the user can enjoy the effect of minimizing the waste of storage space in the client terminal (200) when customizing the user's own world, while also obtaining the effect of increasing the user's immersion in the world by creating and applying various textures that suit the user's taste.

[0126] In some other embodiments related to step S300, the predefined affordance may mean an effect of displaying at least a part of an area corresponding to a chair body (71) selected by a user differently from a chair body (71) in a state in which the user has not made a selection input. However, this is merely an example to help understanding of the present disclosure, and in some other embodiments related to step S300, the predefined affordance is not limited to any one of an effect or screen component that induces a user to make a selection input for any one of the thumbnails of images displayed on the texture selection interface (61).

[0127] In some other embodiments related to step S300, referring to FIGS. 7 and 8, the default texture of the chair body (71) may mean the texture applied to the chair body (71) before the client terminal (200) performs a texture change operation on the chair body (71), as shown in FIG. 7.

[0128] In some other embodiments related to step S300, referring to FIGS. 7 and 8, the default texture of the chair body (71) may mean a specific texture that is defined to be applied to the chair body (71) in the setting information related to the first virtual world when the client terminal (200) runs the texture generation application (300) and loads the first virtual world, as shown in FIG. 7.

[0129] In some other embodiments related to step S300, referring to FIG. 9, when the client terminal (200) receives a deletion input (62-1) for a thumbnail (62) of a first image displayed on a texture selection interface (61), the texture selection interface (61) can be refreshed and displayed so as not to display the thumbnail (62) of the first image deleted by the user.

[0130] In some other embodiments related to step S300, referring to FIG. 9, when the client terminal (200) receives a deletion input (62-1) for a thumbnail (62) of a first image displayed on a texture selection interface (61), the client terminal (200) can change the first texture applied to the chair body (71-1) to which the first texture corresponding to the first image is applied back to the default texture of the chair body (71), and display the chair body (71) to which the default texture is applied on the screen.

[0131] According to this embodiment, the client terminal (200) can also achieve the effect of preventing the user's inconvenient experience of individually resetting objects with textures applied that are no longer desired to exist in the virtual world.

[0132] So far, with reference to FIGS. 2 through 9, a method for generating textures using artificial intelligence according to another embodiment of the present disclosure has been described. It should be understood that the embodiments described above are exemplary in all respects and are not limiting.

[0133] Hereinafter, a method for generating a texture using artificial intelligence according to another embodiment of the present disclosure will be described with reference to FIGS. 10 and 11. It should be understood that the steps described in the flowcharts below are performed by the client terminal (200) described with reference to FIG. 1 unless otherwise stated. Furthermore, it should be understood that the technical concepts understood in the above-described embodiment with reference to FIG. 1 can be readily applied to the method for generating a texture using artificial intelligence according to the present embodiment.

[0134] In step S1000, the client terminal (200) can identify the third image based on a third user input associated with the third image.

[0135] In some embodiments related to step S1000, the third image may be an image pre-stored in the client terminal (200). That is, the user may enjoy the effect of applying any of the preferred images pre-stored in the personal computer (PC) as a texture of the virtual world object.

[0136] In some other embodiments related to step S1000, referring to FIG. 11, the client terminal (200) can identify a second image (112) input by the user into the image input interface (111) displayed on the screen of the client terminal (200). Here, the client terminal (200) can, in response to the user's input of the second image (112), transform the image input interface (111) displayed on the screen of the client terminal (200) into an image input interface (111-1) including the second image (112) input by the user and display it.

[0137] In some other embodiments related to step S1000, the third image may be obtained as a result of the client terminal (200) transmitting a prompt received from the user to an external device. Here, the external device may refer to the image generation system (100) described with reference to FIG. 1.

[0138] In step S200-1, the client terminal (200) can generate a second texture associated with the third image identified in step S1000.

[0139] In some embodiments related to step S200-1, referring to FIG. 11, the client terminal (200) may generate a texture associated with the second image (112) based on input information including the received second image (112). According to some embodiments, the input information input to the client terminal (200) through the image input interface (111) may include data related to an image effect applied to the texture in addition to the second image (112). For example, the input information may include at least one of first image effect data related to overdrawing, second image effect data related to input image resizing, third image effect data related to overdrawing blur processing, fourth image effect data related to an area (i.e., a portion) to which overdrawing is applied, fifth image effect data related to pixels included around the overdrawing, and sixth image effect data related to a denoising intensity, as illustrated in FIG. 11.

[0140] In this case, a second image (112) and one or more image effect data are input to a client terminal (200), and the client terminal (200) can be configured to generate a texture to which features of the second image (112) and features of one or more image effects are applied, or to output feature data related to the texture.

[0141] In some other embodiments related to step S200-1, the second texture may be an asset that does not exist at the time the client terminal (200) loads the second virtual world into the memory of the client terminal (200).

[0142] Below, some embodiments related to step S200-1 may be clearly understood by referring to some embodiments related to step S200-1 described with reference to FIG. 2.

[0143] In step S300, the client terminal (200) may apply the second texture to the second object while performing an operation of displaying at least a part of the second virtual world on the screen in response to a third user input associated with a third image and a fourth user input associated with a second object existing in the second virtual world.

[0144] In some embodiments related to step S300, when a client terminal (200) receives a user input while running a texture generation application (300), the client terminal (200) can load a second virtual world included in the texture generation application (300) into memory and display the second virtual world on the screen.

[0145] In some further embodiments related to step S300, the plurality of objects included in the second virtual world may include characters, special effects, cameras, light sources, and items. Furthermore, the plurality of objects may include entities with specific textures, specific hit boxes, and predefined animations applied.

[0146] Below, some embodiments related to step S300 may be clearly understood by referring to some embodiments related to step S300 described with reference to FIG. 2.

[0147] A method for generating textures using artificial intelligence according to another embodiment of the present disclosure has been described with reference to FIGS. 10 and 11. It should be understood that the embodiments described above are exemplary in all respects and are not limiting.

[0148] FIG. 12 is a hardware configuration diagram of a computing system (1000) according to some embodiments of the present disclosure. The computing system (1000) of FIG. 12 may refer to, for example, the client terminal (200) described with reference to FIG. 1. As another example, the computing system (1000) of FIG. 12 may refer to the image generation system (100) described with reference to FIG. 1. The computing system (1000) may include one or more processors (1100), a system bus (1600), a communication interface (1200), a memory (1400) for loading a computer program (1500) executed by the processor (1100), and a storage (1300) for storing the computer program (1500).

[0149] The processor (1100) controls the overall operation of each component of the computing system (1000). The processor (1100) can perform operations on at least one application or program for executing methods / operations according to various embodiments of the present disclosure. The memory (1400) stores various data, commands, and / or information. The memory (1400) can load one or more computer programs (1500) from the storage (1300) to execute methods / operations according to various embodiments of the present disclosure. The bus (1600) provides a communication function between components of the computing system (1000). The communication interface (1200) supports Internet communication of the computing system (1000). The storage (1300) can non-temporarily store one or more computer programs (1500). The computer program (1500) may include one or more instructions implementing methods / operations according to various embodiments of the present disclosure. When the computer program (1500) is loaded into the memory (1400), the processor (1100) may execute the one or more instructions to perform the methods / operations according to various embodiments of the present disclosure.

[0150] In some embodiments, the computing system (1000) described with reference to FIG. 12 may be configured using one or more physical servers included in a server farm based on cloud technologies such as virtual machines. In this case, at least some of the components illustrated in FIG. 12, such as the processor (1100), memory (1400), and storage (1300), may be virtual hardware, and the communication interface (1200) may also be configured as a virtualized networking element such as a virtual switch.

[0151] A computer program (1500) according to some embodiments of the present disclosure may include instructions related to performing an operation of transmitting a prompt input by a user to a server, an operation of receiving a first image corresponding to the prompt from the server, an operation of generating a first texture corresponding to the first image, and an operation of applying the first texture to a first object while performing an operation of displaying at least a portion of the first virtual world on a screen in response to a first user input related to a first object existing in the first virtual world and a second user input related to the first image.

[0152] A computer program (1500) according to some other embodiments of the present disclosure may include instructions related to an operation of inputting a prompt received from another computing system into an artificial neural network of a computing system (1000), generating the first image based on an output of the artificial neural network for the prompt input, and transmitting the first image to the other computing system.

[0153] Various embodiments of the present disclosure and effects according to the embodiments have been described with reference to FIGS. 1 through 12. The effects according to the technical concept of the present disclosure are not limited to the effects described above, and other effects not mentioned will be clearly understood by those skilled in the art from the description below.

[0154] The technical concepts of the present disclosure described so far can be implemented as computer-readable code on a computer-readable medium. The computer program recorded on the computer-readable recording medium can be transmitted to another computing device via a network such as the Internet, installed on the other computing device, and thus used on the other computing device.

[0155] Although the operations are depicted in a specific order in the drawings, it should not be understood that the operations must be performed in the specific order depicted, or in a sequential order, or that all depicted operations must be performed to achieve the desired results. In certain circumstances, multitasking and parallel processing may be advantageous. Although the embodiments of the present disclosure have been described above with reference to the attached drawings, those skilled in the art to which the present disclosure pertains will understand that the present invention can be implemented in other specific forms without changing the technical spirit or essential characteristics thereof. Therefore, it should be understood that the embodiments described above are illustrative in all respects and not restrictive. The scope of protection of the present invention should be interpreted by the claims below, and all technical ideas within a scope equivalent thereto should be interpreted as being included in the scope of the technical ideas defined by the present disclosure.

Claims

1. One or more processors; and A memory storing a computer program executed by one or more processors, The above computer program causes the processor to: The act of sending a prompt entered by the user to the server; An action of receiving a first image corresponding to the above prompt from the server, wherein the first image is generated by the server by inputting the prompt into an artificial neural network and based on an output of the artificial neural network for the prompt input; An operation of generating a first texture corresponding to the first image; and An instruction comprising: an instruction for performing an operation of applying a first texture to a first object while performing an operation of displaying at least a portion of the first virtual world on a screen in response to a first user input related to a first object existing in a first virtual world and a second user input related to the first image; Computing system.

2. In paragraph 1, The action of applying the first texture to the first object is: An operation of changing a parameter value corresponding to the albedo map of the first object to a value corresponding to the albedo map of the first texture; and An operation including changing a parameter value corresponding to a normal map of the first object to a value corresponding to a normal map of the first texture, Computing system.

3. In paragraph 1, The above first texture is, An asset that does not exist at the time of loading the first virtual world into the memory, Computing system.

4. In paragraph 1, The above first texture is, At least one of an albedo map of the first texture, a normal map of the first texture, a height map of the first texture, and an occlusion map of the first texture, Computing system.

5. In paragraph 1, The operation of generating a first texture corresponding to the first image is: An operation of generating an albedo map associated with the first image; and An operation of generating a normal map corresponding to the albedo map associated with the first image based on the albedo map associated with the first image, Computing system.

6. In paragraph 5, The operation of generating a first texture corresponding to the first image is: Further comprising an operation of storing only an albedo map associated with the first image among parameters related to the first texture corresponding to the first image generated above. Computing system.

7. In paragraph 5, An operation of generating a normal map corresponding to the albedo map associated with the first image based on the albedo map associated with the first image is as follows: An operation of generating the normal map based on the value of each pixel included in the albedo map associated with the first image and the adjacent pixel value of each pixel, Computing system.

8. In paragraph 1, The above computer program causes the processor to: An action of displaying on the screen an interface including a thumbnail of an image corresponding to each of the plurality of previously generated textures; In response to a user input for any one of the above thumbnails, an action of deleting a thumbnail corresponding to the user input from the interface; Further comprising an instruction that causes an operation to change the texture of one or more objects to which a texture corresponding to the thumbnail is applied to a default texture and apply the same. Computing system.

9. One or more processors; and A memory storing a computer program executed by one or more processors, The above computer program causes the processor to: An action of identifying a third image based on a third user input associated with the third image, wherein the third image is an image previously stored in a computing system; An operation of generating a second texture associated with the third image; and In response to the third user input and the fourth user input associated with the second object existing in the second virtual world, an instruction is included to perform an operation of applying the second texture to the second object while performing an operation of displaying at least a portion of the second virtual world on the screen. Computing system.

10. In paragraph 9, The operation of generating a second texture associated with the third image is: An operation of generating an albedo map associated with the third image; and An operation of generating a normal map corresponding to the albedo map associated with the third image based on the albedo map associated with the third image, Computing system.

11. In paragraph 10, An action of applying the second texture to the second object while performing an action of displaying at least a part of the second virtual world on the screen, An operation of changing a parameter value corresponding to the albedo map of the second object to a value corresponding to the albedo map of the second texture; and An operation including changing a parameter value corresponding to a normal map of the second object to a value corresponding to a normal map of the second texture, Computing system.

12. In paragraph 10, An operation of generating a normal map corresponding to the albedo map associated with the third image based on the albedo map associated with the third image is as follows: An operation of generating the normal map based on the value of each pixel included in the albedo map associated with the third image and the adjacent pixel value of each pixel, Computing system.

13. In paragraph 9, The above second texture is, An asset that does not exist at the time of loading the second virtual world into the memory, Computing system.

14. In a method performed by a computing system, A step of sending a prompt entered by a user to the server; The server inputs the prompt into an artificial neural network, and generates a first image corresponding to the prompt based on an output of the artificial neural network for the prompt input, and transmits the first image to the computing system; A step of generating a first texture corresponding to the first image; and In response to a first user input related to a first object existing in a first virtual world and a second user input related to the first image, a step of applying the first texture to the first object while performing an operation of displaying at least a portion of the first virtual world on a screen, A method for generating textures using artificial intelligence.

15. In paragraph 14, The step of applying the first texture to the first object is: A step of changing a parameter value corresponding to the albedo map of the first object to a value corresponding to the albedo map of the first texture; and A step of changing a parameter value corresponding to a normal map of the first object to a value corresponding to a normal map of the first texture, A method for generating textures using artificial intelligence.

16. In paragraph 14, The above first texture is, An asset that does not exist at the time of loading the first virtual world into the memory of the computing system, A method for generating textures using artificial intelligence.

17. In paragraph 14, The step of generating a first texture corresponding to the first image is: generating an albedo map associated with the first image; and A step of generating a normal map corresponding to the albedo map associated with the first image based on the albedo map associated with the first image, A method for generating textures using artificial intelligence.

18. In paragraph 17, The step of generating a first texture corresponding to the first image is: Further comprising a step of storing only the albedo map associated with the first image among the parameters related to the first texture corresponding to the first image generated above. A method for generating textures using artificial intelligence.

19. In paragraph 17, The step of generating a normal map corresponding to the albedo map associated with the first image based on the albedo map associated with the first image is as follows: A step of generating the normal map based on the value of each pixel included in the albedo map associated with the first image and the adjacent pixel value of each pixel, A method for generating textures using artificial intelligence.

20. In paragraph 14, A step of identifying a third image based on a third user input associated with the third image, wherein the third image is an image previously stored in the computing system; generating a second texture associated with the third image; and Further comprising a step of applying the second texture to the second object while performing an operation of displaying at least a portion of the second virtual world on the screen in response to the third user input and the fourth user input associated with the second object existing in the second virtual world, The above second texture is, An asset that does not exist at the time of loading the second virtual world into the memory, A method for generating textures using artificial intelligence.

Images