Method and apparatus for generating texture map

Abstract

The disclosure below relates to a method and an apparatus for generating a texture map, and the method may include the operations of: receiving input data related to a target fabric from a user; classifying the input data as mapping data for generating texture maps corresponding to information of the target fabric; inputting the mapping data to an artificial intelligence model to generate at least one of a normal map and a diffuse map; and generating a synthetic texture map corresponding to the target fabric on the basis of the normal map or the diffuse map.

Description

Method for generating texture maps and device therefor

[0001] The following embodiments relate to a method and device for generating a texture map.

[0002] To simulate virtual clothing naturally in a 3D virtual space, it may be necessary to naturally represent the fabric's characteristics and the patterns expressed on the fabric. Simply synthesizing color and shape onto the fabric to simulate patterns can be difficult to achieve a natural representation of the texture of the fabric itself. Therefore, when representing virtual clothing in a 3D virtual space, it is necessary to generate high-quality texture maps that capture the unique texture of the fabric.

[0003] A method for generating a texture map according to one embodiment may include an operation of receiving input data related to a target fabric from a user, an operation of classifying the input data into mapping data for generating texture maps corresponding to information of the target fabric, an operation of inputting the mapping data into an artificial neural network model to generate at least one of a normal map and a diffuse map, and an operation of generating a synthetic texture map corresponding to the target fabric based on the normal map or the diffuse map.

[0004] The operation of receiving the input data may include an operation of detecting information of the target fabric based on text data input by the user when the text data is received.

[0005] The operation of receiving the above input data may include an operation of detecting information of the target fabric included in the image data when receiving image data input by the user.

[0006] The operation of classifying with the above mapping data includes an operation of classifying information of the target fabric into normal map information, diffuse map information, or other information, wherein the normal map information includes information related to the type of the fabric, and the diffuse map information may include information related to the pattern expression of the fabric.

[0007] The above artificial neural network model may include a generative artificial neural network model that generates the texture maps as seamless maps.

[0008] The above artificial neural network model may include a machine learning model that learns the correlation between the type of fabric and the normal map to generate a preset.

[0009] The operation of classifying the above mapping data may include an operation of generating the mapping data based on the preset when the preset matches the input data.

[0010] The operation of classifying with the above mapping data may include an operation of outputting mapping data related to the target fabric based on the input data when the input data and the preset do not match.

[0011] The operation of outputting mapping data related to the target fabric may include an operation of outputting a candidate group of presets similar to the normal map information when the mapping data and the preset do not match and the mapping data is normal map information.

[0012] The operation of generating the texture map may include an operation of processing the normal map into a filter image, an operation of transforming the diffuse map using the filter image, and an operation of blending the transformed diffuse map and the filter image.

[0013] The above blending operation may include an operation of blending the filter image and the modified diffuse map through image compositing.

[0014] The method further includes an operation of inputting the above mapping data into an artificial neural network model to generate a miscellaneous map, and the operation of generating the synthetic texture map includes an operation of generating a final synthetic texture map corresponding to the target fabric based on at least one of the normal map, the diffuse map, and the miscellaneous map, and the miscellaneous map can be generated based on normal map information and other information.

[0015] The method may further include simulating the above-mentioned synthetic texture map on a three-dimensional virtual garment and displaying the three-dimensional virtual garment to which the above-mentioned synthetic texture map is applied through a user interface.

[0016] The method may further include outputting a feedback message to the user so that the user can make real-time adjustments to the input data.

[0017] An electronic device for simulating a virtual garment according to one embodiment includes a memory storing instructions and one or more processors, wherein the instructions, when executed by the one or more processors, cause the electronic device to receive input data related to a target fabric from a user, classify the input data into mapping data for generating texture maps corresponding to information of the target fabric, input the mapping data into an artificial neural network model to generate at least one of a normal map and a diffuse map, and generate a synthetic texture map corresponding to the target fabric based on the normal map or the diffuse map.

[0018] FIG. 1 is a flowchart illustrating a method for generating a texture map according to one embodiment.

[0019] FIG. 2 is a flowchart illustrating a method for generating a synthetic texture map according to one embodiment.

[0020] FIG. 3 is an exemplary drawing of a normal map, a diffuse map, and a composite texture map according to one embodiment.

[0021] FIG. 4 schematically illustrates an exemplary operation of an electronic device generating a synthetic texture map according to one embodiment.

[0022] FIG. 5 schematically illustrates an exemplary operation of an electronic device generating a synthetic texture map according to one embodiment.

[0023] Figure 6 schematically illustrates a process for an artificial neural network model to obtain a normal map according to one embodiment.

[0024] FIGS. 7 to 8d are schematic drawings illustrating a user interface according to one embodiment.

[0025] Figure 9 is a block diagram of an electronic device according to one embodiment.

[0026] Specific structural or functional descriptions of the embodiments are disclosed for illustrative purposes only and may be modified and implemented in various forms. Therefore, the actual implementation is not limited to the specific embodiments disclosed, and the scope of this specification includes modifications, equivalents, or alternatives within the technical concepts described in the embodiments.

[0027] Although terms such as "first" or "second" may be used to describe various components, these terms should be interpreted solely to distinguish one component from another. For example, a first component may be referred to as a second component, and similarly, a second component may also be referred to as a first component.

[0028] When it is said that a component is "connected" to another component, it should be understood that it may be directly connected or connected to that other component, but there may also be other components in between.

[0029] Singular expressions include plural expressions unless the context clearly dictates otherwise. In this specification, the terms "comprises" or "has" should be understood to indicate the presence of a described feature, number, step, operation, component, part, or combination thereof, but not to exclude the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof.

[0030] In this document, each of the phrases "A or B", "at least one of A and B", "at least one of A or B", "A, B, or C", "at least one of A, B, and C", and "at least one of A, B, or C" may include any one of the items listed together in that phrase, or all possible combinations thereof.

[0031] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as commonly understood by a person of ordinary skill in the art. Terms defined in commonly used dictionaries should be interpreted to have a meaning consistent with their meaning in the context of the relevant technology, and will not be interpreted in an idealized or overly formal sense unless explicitly defined herein.

[0032] Hereinafter, embodiments will be described in detail with reference to the attached drawings. In the description with reference to the attached drawings, identical components are assigned the same reference numerals regardless of the drawing numbers, and redundant descriptions thereof will be omitted.

[0033]

[0034] FIG. 1 is a flowchart illustrating a method for generating a texture map according to one embodiment.

[0035] For convenience of explanation, operations (110 to 140) are described as being performed using the electronic device (900) illustrated in FIG. 9. However, these operations (110 to 140) may be utilized via any other suitable electronic device and within any suitable system.

[0036] Furthermore, the operations of FIG. 1 may be performed in the order and manner illustrated, but the order of some operations may be changed or some operations may be omitted without departing from the spirit and scope of the illustrated embodiment. Multiple operations illustrated in FIG. 1 may be performed in parallel or simultaneously.

[0037] In one embodiment, an electronic device (900) may need to simultaneously display the type of fabric of the virtual garment and the pattern expressed on the fabric to simulate a three-dimensional virtual garment. Accordingly, the electronic device (900) may need to generate a texture map similar to reality for the fabric of the three-dimensional virtual garment and apply it to the simulation.

[0038] A texture map is a collection of images or data used to visually represent fabric surfaces, colors, and patterns in 3D modeling and computer graphics. Texture maps can provide visual effects for the appearance of objects in 3D simulations. Texture maps can come in various types, including diffuse maps, normal maps, displacement maps, specular maps, height maps, roughness maps, and ambient occlusion maps. Furthermore, multiple types of texture maps can be synthesized and used to achieve realistic surface representations.

[0039] An image that only expresses a pattern with color values ​​and resolution may perform a virtual clothing simulation that is far from reality when expressing fabric. In the described embodiment, a synthetic texture map may refer to a texture map that expresses both the pattern and material of the fabric. Accordingly, the synthetic texture map may be a high-quality texture map that simulates even the deformation of the pattern due to the unique material. The synthetic texture map is generated identically or similarly to the target fabric and provided to the user, so that the fabric to be applied to the clothing that the user actually wants to implement in reality can be generated according to the user's intention.

[0040] Hereinafter, a method for an electronic device (900) to generate a texture map using an artificial neural network model (e.g., the artificial neural network model (420) of FIG. 4, the artificial neural network model (520) of FIG. 5, or the artificial neural network model (602) of FIG. 6) is described.

[0041] In operation (110), the electronic device (900) may receive input data related to a target fabric from the user. Fabric may refer to a flexible material made by weaving, knitting, crocheting, or other methods using thread or fiber. Fabric may be used for various purposes, such as making clothing, household textiles (e.g., curtains, bedding), and industrial materials. The characteristics of the fabric may be determined by the type of fiber (e.g., cotton, wool, silk, polyester, etc.), the manufacturing method (e.g., weaving, knitting), and the processing method (e.g., dyeing, printing, coating). Due to this diversity, the fabric may have various textures, weights, strengths, elasticity, and transparency, and may be selected to suit a specific application. A target fabric is a fabric that reflects visual or physical characteristics that a user wishes to create using an electronic device (900), and may refer to a fabric to be applied to a virtual garment when the user creates a virtual garment in a virtual space. For example, if a user wishes to create a fabric with a floral pattern of the 'cotton' type, the target fabric may be 'flower pattern on cotton'.

[0042] The electronic device (900) may provide a user interface (e.g., the user interface (700) of FIG. 8) that allows the user to input input data, or may provide a service through a server so that the user can use the user interface with the user's terminal.

[0043] A terminal is a device capable of installing and executing server-related applications, and can provide an interface to the user. This interface may be provided by the terminal itself. For example, it may be provided by the terminal's operating system (OS) or by an application installed on the terminal. Furthermore, the interface may be provided by the server, or the terminal may simply receive and display the interface provided by the server.

[0044] The user interface may include an input area where a user can enter text or upload an image. Additionally, the user interface may include touchscreen operation, sliding menus, and the like to facilitate user input diversity and convenience. The electronic device (900) may vary its data processing method depending on the type and format of input data received through the user interface.

[0045] According to one embodiment, an electronic device (900) may, upon receiving text data (or a message) input by a user, detect information about a target fabric based on the text data. Information about the target fabric may refer to the type and details of data related to the fabric.

[0046] Information about the target fabric may include information such as the type of fabric, texture of the fabric, pattern of the fabric, color of the fabric, thickness and weight of the fabric, and elasticity of the fabric. The electronic device (900) can detect and output the type, composition, etc. of the fabric.

[0047] The types of fabric include, for example, 'Boucle', 'Canvas', 'Challis', 'Chambray' / Oxford', 'Chiffon', 'Clip jacquard', 'Corduroy', 'Crepe' / 'CDC', 'Crepe knit'. 'Crochet', 'Denim', 'Dewspo', 'Dobby', 'Dobby mesh', 'Double knit' / 'Interlock', 'Double weave', 'Eyelet', 'Flannel', 'Flatback rib', 'Fleece', 'French terry', 'Gauze' / 'Double gauze', 'Georgette', 'ITY' / 'Matte jersey', 'Jacquard' / 'Brocade', 'Jacquard knit', 'Jersey', 'Lace', 'Loop terry', 'Low gauge knit', 'Melton' / 'Boiled', 'Memory', 'Mesh' / 'Tulle', 'Neoprene' / 'Scuba', 'Organza', 'Ottoman', 'PVC', 'Pique', 'Plaid', 'Plain', 'Pointelle', It can correspond to 'Polar fleece', 'Ponte', 'Poplin', 'Quilted knit', 'Rib', 'Ripstop', 'Satin', 'Seersucker', 'Sherpa', 'TRS', 'Taffeta', 'Tricot', 'Tweed', 'Twill', 'Tyvek', 'Vegan fur', 'Vegan leather', 'Vegan suede', 'Velour', 'Velvet', 'Velvet' / 'Velveteen', 'Voile', 'Waffle', etc. However, the types of target fabrics listed are examples and are not necessarily limited to them.

[0048] The pattern of the fabric may be, for example, 'Argyle', 'Art Deco', 'Batik', 'Baroque', 'Brocade', 'Camouflage', 'Chevron', 'Chintz', 'Damask', 'Ditsy Print', 'Floral', 'Fleur de Lis', 'Gingham', 'Greek Key', 'Herringbone', 'Houndstooth', 'Ikat', 'Jacquard', 'Lattice', 'Leopard Print', 'Moroccan Tile', 'Ogee', 'Paisley', 'Plaid (Tartan)', 'Polka Dot', 'Quatrefoil', 'Seersucker', 'Stripes', 'Toile de Jouy', 'Zebra Print', etc. However, the patterns of the target fabrics listed are examples and are not necessarily limited thereto.

[0049] For example, assume that a user inputs "Fleur De Lis Patterns" and "challis type" through the user interface. At this time, the electronic device (900) can detect "Fleur De Lis," which is pattern information of the target fabric, and "challis," which is type information of the target fabric, from the text message.

[0050] As another example, a user may input a text message into the input area without explicitly specifying it as described above. Assume that the user inputs the text message "wool lily-shaped fabric" into the input area. At this time, the electronic device (900) can detect the pattern information of the target fabric, "lily-shaped," and the type information of the target fabric, "wool material," from the text message.

[0051] If the information of the exact target fabric cannot be detected from the text message entered by the user, or if the user wants to specify the target fabric, the electronic device (900) may prompt the user to input more text messages by inducing the use of a conversational user interface.

[0052] According to one embodiment, when an electronic device (900) receives image data input by a user, it can detect information about a target fabric included in the image data. For example, when a user inputs an image expressing the pattern of 'Fleur De Lis Patterns', the electronic device (900) can extract features from the image and detect pattern information about the target fabric.

[0053] The electronic device (900) may receive image data and text messages simultaneously or sequentially. For example, the electronic device (900) may simultaneously receive an image expressing the pattern of 'Fleur De Lis Patterns' and a text data message or preset input of 'challis type'. In this case, the electronic device (900) may detect the pattern information of the target fabric as 'Fleur De Lis Patterns' and the type information of the target fabric as 'challis type'. The preset input may include providing data that the electronic device (900) has learned and stored in a database through a user interface, and allowing the user to input the data by clicking or using a sliding menu. For example, the preset input may be a user input for a preset option on the user interface, such as the Fabric Type option (711), the Content Type option (712), and the Pattern Size option (713) provided in the first texture generation interface (710) of FIGS. 8A to 8C.

[0054] In operation (120), the electronic device (900) may classify input data into mapping data for generating texture maps corresponding to information of the target fabric. The mapping data may be classified data for analyzing the input data and generating a texture map corresponding to each type of input data according to the type of input data.

[0055] An electronic device (900) according to one embodiment may classify information about a target fabric into normal map information, diffuse map information, or other information. Normal map information may include information related to the type of fabric, and diffuse map information may include information related to the pattern expression of the fabric.

[0056] An artificial neural network model according to one embodiment may include a machine learning model that learns a correlation between a type of fabric and a normal map to generate a preset.

[0057] An electronic device (900) according to one embodiment can generate mapping data based on a preset when the preset matches the input data.

[0058] An electronic device (900) according to one embodiment can output mapping data related to a target fabric based on the input data when the input data and the preset do not match.

[0059] According to one embodiment, a user interface may provide a list of presets that have been learned and stored in a database, such as a sliding menu, so that a user can input presets. When a user inputs information related to a normal map through a conversational user interface, and the mapping data does not match the preset, the electronic device (900) may provide a group of normal map candidates related to the mapping data from the preset.

[0060] An electronic device (900) according to one embodiment may output a candidate group of presets similar to the normal map information when the mapping data and the preset do not match and the mapping data is normal map information. When an artificial neural network model generates and outputs a diffuse map and a normal map, whether a generative artificial neural network model or a machine learning model that has learned the presets is used may vary depending on the user's settings or the texture map generation environment.

[0061] A normal map can be a texture map that expresses the detailed texture and curvature of the fabric surface, depending on the fabric type. Therefore, normal map information related to the fabric type can correspond to the target fabric type information detected from the input data entered by the user.

[0062] A diffuse map can be a texture map that defines the surface color or pattern of a fabric. Therefore, diffuse map information related to the pattern expression of a fabric can correspond to the pattern information of the target fabric detected from the input data entered by the user. The pattern information of the target fabric can include not only the pattern information expressed on the target fabric, but also color information.

[0063] Other maps can include other maps besides normal map and diffuse map, such as displacement map, specular map, height map, roughness map, and ambient occlusion map.

[0064] A displacement map may be a texture map that reflects the geometric characteristics of the fabric surface to represent the actual physical depth and detail of the fabric. When the electronic device (900) applies the displacement map to a 3D virtual garment, it can simulate a realistic sense of depth of the fabric and provide it to the user interface.

[0065] A specular map may be a texture map that expresses the reflectivity and glossiness of a fabric surface by reflecting the reflection of light on the fabric surface. When the electronic device (900) applies the specular map to a 3D virtual garment, specific parts of the fabric applied to the 3D virtual garment may appear glossy or matte, thereby simulating various textures of the fabric and providing them to the user interface.

[0066] A height map may be a texture map that expresses the depth and curvature of a texture by representing the elevation of a fabric surface as a grayscale image. When an electronic device (900) applies a height map to a three-dimensional virtual garment, it can simulate realistic textures of detailed elevation changes of the fabric surface and provide them to a user interface.

[0067] A roughness map may be a texture map that expresses the degree of roughness of a fabric surface. When the electronic device (900) applies the roughness map to a three-dimensional virtual garment, it can realistically express changes in the roughness of the fabric surface, thereby simulating the characteristics of various fabric materials and providing them to the user interface.

[0068] An ambient occlusion map may be a texture map that expresses shadows in areas of the surrounding environment where light does not reach. When the electronic device (900) applies an ambient occlusion map to a 3D virtual garment, it can simulate soft shadows in gaps, corners, and small holes of the 3D virtual garment, thereby providing a sense of depth and realism through the user interface.

[0069] Texture maps other than the normal map and diffuse map are not limited to the above-described embodiments, and may include texture maps that can be employed by a conventional technician, such as an opacity map.

[0070] An electronic device (900) according to one embodiment may receive input data related to a target fabric, analyze the same, and then generate another map.

[0071] An electronic device (900) according to one embodiment can analyze input data to classify mapping data corresponding to information about a target fabric. The remaining information, excluding normal map information and diffuse map information, can be classified as "other information." The "other information" can include information about physical properties or composition related to the target fabric, such as fabric thickness and weight, fabric roughness, and fabric elasticity.

[0072] According to one embodiment, the electronic device (900) can input mapping data into an artificial neural network model to generate a normal map. The electronic device (900) can generate the normal map based on normal map information or other information. According to one embodiment, the electronic device (900) can input mapping data into an artificial neural network model to generate other maps. The electronic device (900) can generate other maps based on normal map information or other information.

[0073] In one embodiment, other information may be input into an artificial neural network model and utilized to generate other maps, such as a displacement map, a specular map, a height map, a roughness map, an ambient occlusion map, etc. For example, if the other information relates to the degree of roughness (e.g., smoothness), the artificial neural network model may generate a roughness map that expresses the smoothness of the fabric surface.

[0074] In operation (130), the electronic device (900) may input mapping data into an artificial neural network model to generate at least one of a normal map and a diffuse map of the target fabric. The artificial neural network model may include at least one of a generative adversarial network (GAN) and a machine learning model.

[0075] An artificial neural network model according to one embodiment may include a generative artificial neural network model that generates a texture map as a seamless map. A seamless map may mean that the texture map appears to be smoothly connected without any noticeable borders or seams on the surface of the fabric. A seamless map may be designed so that the texture can be infinitely repeated in any direction in the texture map. The generative artificial neural network model may generate normal maps, diffuse maps, and other maps as seamless maps.

[0076] For example, if a user inputs "Fleur De Lis Patterns" as a text message, the electronic device (900) can generate a diffuse map for "Fleur De Lis Patterns" as a seamless map using an artificial neural network model. As another example, if a user inputs a non-seamless image expressing "Fleur De Lis Patterns," the electronic device (900) can generate a diffuse map for "Fleur De Lis Patterns" as a seamless map using an artificial neural network model. This also applies to normal maps and other maps, so that the artificial neural network model can generate each map as a seamless map depending on the input.

[0077] Since a normal map is a texture map generated using fabric type information, if an artificial neural network model has learned the correlation between the normal map and the fabric type, a normal map can be generated from the fabric type information. As mentioned above, a preset can refer to data that pairs the fabric with various information that can define the fabric, such as the fabric type, fabric pattern, fabric texture, and fabric thickness. An artificial neural network model that has learned the correlation between presets and the characteristics of actual fabrics can compare the user's input with the mapping data and output a normal map.

[0078] If the mapping data corresponds to a preset learned by the artificial neural network model, the artificial neural network model can output a texture map according to the learned correlation. For example, if the artificial neural network model has previously learned 'Fleur De Lis Patterns' and 'challis type' and saved them as presets, when a user inputs "Fleur De Lis Patterns + challis type" through the user interface, the artificial neural network model can search for a preset corresponding to 'Fleur De Lis Patterns' and 'challis type' and output a diffuse map and normal map that are the same as or similar to the preset. In addition, since the artificial neural network model includes a generative artificial neural network model, it can also newly generate a diffuse map for 'Fleur De Lis Patterns'.

[0079] On the other hand, if the mapping data does not correspond to the presets learned by the artificial neural network model or does not exist, the artificial neural network model can analyze the mapping data to generate appropriate texture maps.

[0080] For example, assume that a user inputs "a fabric with a pattern of blue dots and a curve" into the input area. At this time, the electronic device (900) can detect the "pattern of blue dots" as mapping data related to the pattern of the target fabric, and detect "curvature" as mapping data related to the type of the target fabric. Here, the artificial neural network model can receive the detected mapping data as input. Since the artificial neural network model includes a generative artificial neural network model, a new diffuse map for the "pattern of blue dots" can be generated, or a new diffuse map can be generated based on a pattern such as "Blue coin dot" that has been previously learned from a preset.

[0081] An electronic device (900) according to one embodiment can generate a normal map by directly reflecting the "curvature" input by the user. For example, the electronic device (900) can select a normal map expressing "curvature" from among various normal map examples generated and provided through an artificial neural network model, and this can be a result generated directly without relying on an existing preset. When the electronic device (900) selects a normal map expressing "curvature," it can request additional information from the user by providing various examples of normal maps generated by the artificial neural network model on the user interface, or request additional information required for selecting a normal map from the user through the user interface, thereby generating a final normal map.

[0082] In one embodiment, an artificial neural network model can learn correlations between inputs describing fabric characteristics, such as "curvature," "roughness," and "smoothness," and normal maps. For example, the artificial neural network model can learn correlations that enable it to generate a "corduroy normal map" based on inputs describing the fabric characteristic "curvature."

[0083] Additionally, the artificial neural network model can learn correlations between inputs describing fabric characteristics, such as "curvature," "rough texture," and "smooth," and fabric types, and generate normal maps corresponding to the fabric characteristics based on the fabric types detected in this learning. For example, the artificial neural network model can learn correlations that enable it to detect a fabric type called "coduroy" based on inputs describing fabric characteristics, and then generate normal maps corresponding to "coduroy" based on the detected fabric types.

[0084] Typically, diffuse maps can be implemented in reality by generating virtual patterns based on user input. However, normal maps can utilize an interactive user interface to create target fabrics by inputting the properties or characteristics of a fabric type interactively, rather than using presets provided in the user interface or by entering the fabric type name.

[0085] Referring to FIG. 8d, even if the user does not provide an input related to the type of target fabric that matches the name of a preset learned by the user, the artificial neural network model can compare the user's input with the preset and output a candidate set of presets similar to the user's input, and the electronic device (900) can provide the candidate set of presets to the user interface to induce the user's selection of a normal map.

[0086] The electronic device (900) can display the generated diffuse map and normal map as intermediate results through a user interface. If the user inputs into the user interface that the intermediate results are not the intended results, the electronic device (900) can generate other intermediate results through an artificial neural network model. In addition, the electronic device (900) can induce additional information about the input data in a conversational format with the user through the user interface to generate a texture map that meets the user's intention. However, the user interface of the electronic device (900) is not limited to a conversational format, and a method of inducing input by the user's selection, such as by displaying multiple intermediate results and clicking on a sliding menu or a candidate group, can be used.

[0087] In operation (140), the electronic device (900) may generate a synthetic texture map corresponding to the target fabric based on a normal map or a diffuse map. The synthetic texture map may include a texture map generated by synthesizing texture maps (e.g., a normal map, a diffuse map, and other maps).

[0088] According to one embodiment, the electronic device (900) can generate a final composite texture map corresponding to the target fabric based on at least one of a normal map, a diffuse map, and other maps.

[0089] According to one embodiment, the electronic device (900) may generate an ND composite texture map based on a normal map (N) and a diffuse map (D), and then additionally apply other maps to the ND composite texture map to generate a final composite texture map. In this case, the electronic device (900) may generate an ND composite texture map by synthesizing the normal map (N) and the diffuse map (D) according to a user input, and then apply other maps, such as a specular map or a roughness map, to generate a final composite texture map to which additional characteristics, such as gloss or roughness of the surface, are imparted. Through this, the user may add physical properties or characteristics on top of the basic texture, color, and pattern of the target fabric to be generated.

[0090] According to one embodiment, the electronic device (900) may generate an NE composite texture map based on a normal map (N) and other maps (E), and additionally apply a diffuse map to the NE composite texture map to generate a final composite texture map. For example, the electronic device (900) may generate an NE composite texture map reflecting a basic texture by synthesizing the generated normal map and roughness map, and then additionally apply a diffuse map according to a user input to generate a final texture map having the same texture but with various colors and patterns, and provide the final texture map to a user interface. Through this, the user may add various color and pattern expressions on top of the same texture and material properties of the target fabric to be generated.

[0091] According to one embodiment, the electronic device (900) can generate a final synthetic texture map by simultaneously reflecting the generated normal map, diffuse map, and / or other maps. Alternatively, the electronic device (900) can simultaneously input normal map information, diffuse map information, and other map information included in data input from a user into an artificial neural network model to generate a final synthetic texture map that integrates the information of each map. Through this, the user can obtain a final synthetic texture map that comprehensively reflects the texture of the normal map, the color and pattern of the diffuse map, and additional material properties or characteristics of the other maps for the target fabric to be generated.

[0092] The described embodiments illustrate some of the methods for generating a final composite texture map, and the method and order of generating the final composite texture are not limited to the described embodiments. The electronic device (900) can generate the final composite texture map by reflecting the normal map, diffuse map, and other maps in various combinations and orders according to user needs and input data.

[0093]

[0094] FIG. 2 is a flowchart illustrating a method for generating a synthetic texture map according to one embodiment.

[0095] The description referring to Fig. 1 can be equally applied to Fig. 2, and overlapping content can be omitted.

[0096] In operation (140), the electronic device (900) can perform operations (141 to 143) to generate a synthetic texture map corresponding to the target fabric.

[0097] In operation (141), the electronic device (900) can adjust the sizes of the normal map and the diffuse map. For example, it is assumed that the diffuse map to be used has a size of 512x512 and the normal map has a size of 1822x602. In order to synthesize the normal map and the diffuse map, the electronic device (900) can adjust the size of the diffuse map to 3584x512 and the size of the normal map to 3644x602 so that the two maps have similar sizes. However, adjusting the width and height of the maps in this way to an integer ratio is for the convenience of explanation, and the sizes of the normal map and the diffuse map can be adjusted to be similar or the same.

[0098] In operation (142), the electronic device (900) may process the normal map into a filter image and transform the diffuse map using the filter image. For example, the electronic device (900) may change the normal map into a 1-channel grayscale image so that the pattern of the fabric expressed in the normal map stands out. Additionally, the gamma value of the normal map image may be adjusted to emphasize patterns such as the curvature of the fabric.

[0099] A 1-channel grayscale image represents an image's color information as a single channel. These images contain only brightness information, not color intensity, and pixel values ​​typically range from 0 (black) to 255 (white). Each pixel's value represents its brightness, and intermediate values ​​can represent various shades of gray.

[0100] A 1-channel grayscale image can be used when color information is not required, or to reduce complexity in texture analysis, pattern recognition, and image processing. Furthermore, it can save storage space and processing time, thereby increasing data processing efficiency. Since the electronic device (900) requires information such as the curvature of the fabric expressed in the normal map, it can convert the normal map into a 1-channel grayscale image. However, the 1-channel grayscale method is only one example of a method by which the electronic device (900) can express the curvature of the normal map prominently, and other methods capable of expressing the curvature prominently may be used.

[0101] An electronic device (900) according to one embodiment can transform a diffuse map using a filter image to apply an expression, such as the curvature of a fabric expressed in a normal map, to the diffuse map.

[0102] For example, the size of the normal map and the size of the diffuse map are the same width. , height Assume that the 3D RGB color values ​​of each pixel that makes up the diffuse map (here, the RGB color value range is assumed to be 0 to 255) (Here, The range is 0 or more under, The range is 0 or more (assumed to be less than 0) is assumed. The 1-channel gray scale color value of the filter image (here, the range of the gray scale color value is assumed to be 0 to 255) is At this time, the 3D RGB color values ​​of each pixel that constitutes the transformed diffuse map, which is the output result image by applying the filter image to the diffuse map, are can be done with The formula for obtaining can be expressed as the following mathematical formula 1.

[0103]

[0104] In mathematical expression 1, silver It means the integer part of, is the intensity of distortion, the larger the value, the greater the distortion of the diffuse map. Here, is the degree of distortion in the horizontal direction of the image, refers to the intensity of distortion in the vertical direction of the image. Each factor has no range limitation, and a negative value (-) may mean that the distortion appears to occur in the opposite direction. At this time, the electronic device (900) may use a fixed specific value, but may also change the value as needed.

[0105] In operation (143), the electronic device (900) can blend the transformed diffuse map and the filter image. Referring to FIGS. 3 to 5 described below, the electronic device (900) can blend the transformed diffuse map and the filter image to generate a final result, a synthetic texture map intended by the user. The electronic device (900) can blend the filter image and the transformed diffuse map through image compositing. As a result, the electronic device (900) can generate a synthetic texture map.

[0106] Image compositing (or image compositing mode) refers to the process of combining multiple images or elements in image editing software to create a new image. Through image compositing, the electronic device (900) can appropriately arrange each element and use a blending mode to combine different elements to achieve the final desired visual effect. In image compositing mode, the electronic device (900) can primarily perform tasks such as image combining, layer adjustment, masking, and color tone and lighting adjustment.

[0107] Image stitching is the process of combining multiple images to create a new image. This can be used to composite people or objects onto a background image, or to create a scene composed of multiple photographs.

[0108] Layer manipulation involves stacking individual images or elements into layers and then adjusting the transparency or blending mode of each layer to achieve the desired effect. This allows you to achieve natural harmony between image elements or emphasize specific areas.

[0109] Masking is the process of using a mask to hide or reveal parts of an image, revealing only the desired portion. This allows you to manipulate parts of an image or emphasize specific areas. Color and lighting adjustments allow you to adjust the hue or contrast of an image, or add lighting effects to enhance the overall visual effect.

[0110] For example, the electronic device (900) can blend a filter image and a modified diffuse map using a soft light method among image compositing methods.

[0111] Soft light can refer to a method of compositing two images based on one image. Using the color values ​​of one image, the contrast of the other can be enhanced or diminished. Soft light can be used to combine two images to create a new image. Blending using soft light can simulate lighting effects and add soft illumination to an image.

[0112] The Soft Light method can be determined primarily by two factors: the brightness of the blending target and the brightness of the blending color. The Soft Light method can process bright and dark areas differently. In bright areas, the color of the layer being blended with the Soft Light method can be adjusted to a lighter hue, while in dark areas, the color can be adjusted to a darker hue. Images blended with the Soft Light method can produce a soft lighting effect.

[0113] The Soft Light method can be useful for adjusting the color, contrast, and lighting of an image. Blending images using the Soft Light method can achieve a variety of effects, including brightening poorly lit photos and adding soft shadows to images.

[0114] That is, the electronic device (900) can apply the soft light method so that the relationship between each pixel of the modified diffuse map becomes clear based on the filter image by using the soft light method. If it is determined that the brightness of the modified diffuse map is not sufficient and the texture or curves of the fabric expressed in the normal map are not well expressed, the electronic device (900) can repeatedly apply the soft light method. However, the image compositing method of the described embodiment is not limited to the soft light method, and an image compositing method applicable by a person of ordinary skill in the art (e.g., methods such as Normal, Overlay, and Hard Light) can be used.

[0115] An electronic device (900) according to one embodiment may input other information related to the target fabric included in the input data into an artificial neural network model to generate a other map. The electronic device (900) may synthesize the other map into the synthetic texture map to update the synthetic texture map.

[0116] For example, if the other information relates to the elasticity of the target fabric, an artificial neural network model can generate a texture map, a guitar map, related to elasticity. However, it may be difficult to express a guitar map related to elasticity in a 2D user interface. However, if the user actually applies a synthetic texture map with a guitar map applied to a 3D virtual garment, elasticity data can be displayed in the simulation results, or elasticity can be simulated by applying physical forces to the 3D virtual garment during simulation.

[0117]

[0118] FIG. 3 is an exemplary drawing of a normal map, a diffuse map, and a composite texture map according to one embodiment.

[0119] The description with reference to FIGS. 1 and 2 can be equally applied to FIG. 3, and overlapping content can be omitted.

[0120] Referring to FIG. 3, an electronic device (900) according to an embodiment may generate a composite texture map (350) by synthesizing a diffuse map (310) and a normal map (330). The diffuse map (310) may be an exemplary diffuse map generated by the electronic device (900) by receiving 'Fleur De Lis' as text or an image. The normal map (330) may be an exemplary normal map generated by the electronic device (900) by receiving 'Quilted knit (Polyester 88%, Viscose rayon 9%, Spandex / Elastane 3%)' as text or an image. The electronic device (900) may generate a composite texture (350) map in which a 'Fleur De Lis' pattern is expressed on a 'Quilted Knit' material by synthesizing the diffuse map (310) and the normal map (330). The electronic device (900) can display the generated synthetic texture map (350) through a user interface.

[0121]

[0122] FIG. 4 schematically illustrates an exemplary operation of an electronic device generating a synthetic texture map according to one embodiment.

[0123] The description referring to FIGS. 1 and 2 can be equally applied to FIG. 4, and overlapping content can be omitted.

[0124] Referring to FIG. 4, the electronic device (900) can receive input related to a target fabric. The user can input normal maps and diffuse maps differently depending on the form of the user interface. For example, if the user interface is a texture generation interface (e.g., the first texture generation interface (710) of FIG. 8a) in which information related to the normal map is input through a preset-based input area and information related to the diffuse map is input through an input area in the form of a prompt, the user can select 'Canvas' from the preset and input 'Blue coin dot' from the prompt. As another example, if the user interface is a user interface (e.g., the second texture generation interface (810) of FIG. 8d) having a dialogue-based input area in which both the normal map and the diffuse map are input through a prompt, the user's input can be "Blue coin dot on Canvas."

[0125] The electronic device (900) can classify the received input (410) to detect mapping data. For example, since the preset is 'Canvas', the electronic device (900) can detect 'Canvas' as information related to the type of the target fabric, and 'Blue coin dot' as information related to the pattern of the target fabric from the user input 'Blue coin dot' entered in the input area. For another example, the electronic device (900) can detect 'Canvas' as information related to the type of the target fabric from the user input 'Blue coin dot on Canvas', and 'Blue coin dot' as information related to the pattern of the target fabric. Accordingly, the electronic device (900) can classify 'Blue coin dot' as information related to the diffuse map, and 'Canvas' as information related to the normal map. Accordingly, the electronic device (900) can classify 'Blue Coin dot, Canvas' as mapping data (410).

[0126] The electronic device (900) can input mapping data (410), which is information related to the classified diffuse map and information related to the normal map, into an artificial neural network model (420). For example, the artificial neural network model (420) can generate a 'Blue coin dot diffuse map (431)' from information related to the diffuse map and a 'Canvas normal map (432)' from information related to the normal map using a generative artificial neural network model and a machine learning model. At this time, the artificial neural network model can generate the 'Blue coin dot diffuse map (431)' and the 'Canvas normal map (432)' as a seamless map, so that even when each map is connected and expanded, the shape of the pattern or type can be continuously connected without a break.

[0127] The electronic device (900) can generate a synthetic texture map using a synthesis algorithm (440) for the diffuse map and normal map generated by the artificial neural network model. For example, the artificial neural network model can generate a 'Blue coin dot on canvas texture map (450)' using the 'Blue coin dot diffuse map (431)' and the 'Canvas normal map (432)' using the synthesis algorithm (440). The synthesis algorithm (440) may be an algorithm that utilizes the method referred to in FIG. 2 described above.

[0128] The electronic device (900) can display the 'Blue coin dot on canvas texture map (450)' and the 'Canvas normal map (432)' generated in response to the user's input together in the result window (460) of the user interface.

[0129]

[0130] FIG. 5 schematically illustrates an exemplary operation of an electronic device generating a synthetic texture map according to one embodiment.

[0131] The description referring to FIGS. 1 and 2 can be equally applied to FIG. 5, and overlapping content can be omitted.

[0132] Referring to FIG. 5, the electronic device (900) can receive input related to the target fabric. For example, it is assumed that the user uploads a target image (511) related to the pattern of the target fabric he or she wants to use, and inputs 'Canvas (512)' as the type of the target fabric. The electronic device (900) can receive the target image (511) and 'Canvas (512)' as inputs related to the target fabric. At this time, the 'Canvas (512)', which is information related to the normal map, may be information selected by the user from a preset provided in the user interface, or may be mapping data related to the normal map detected based on information input by the user in the input area.

[0133] The electronic device (900) can classify the received input and detect it as mapping data. For example, the electronic device (900) can detect information related to the pattern of the target fabric from the target image (511), which is the user's input. At this time, the electronic device (900) can extract the features of the target image (511) and detect that it is a 'Blue coin dot' pattern. 'Canvas (512)' may be information related to the type of the target fabric. Accordingly, the electronic device (900) can classify the target image (511) as information related to the diffuse map and classify 'Canvas (512)' as information related to the normal map.

[0134] The electronic device (900) can input mapping data (510), which is information related to the classified diffuse map and information related to the normal map, into the artificial neural network model (520). For example, the artificial neural network model (520) can generate a 'Blue coin dot diffuse map (531)' from information related to the diffuse map and a 'Canvas normal map (532)' from information related to the normal map using a generative artificial neural network model and a machine learning model. At this time, the artificial neural network model can generate the 'Blue coin dot diffuse map (531)' and the 'Canvas normal map (532)' as a seamless map, so that even if each map is extended by connecting them, the shape of the pattern or type can be continuously connected without a break. Here, since the artificial neural network model (520) receives the target image (511) from the user, it can only perform the task of generating it as a seamless map without generating it separately.

[0135] The electronic device (900) may utilize inpainting technology when generating a seamless map. Inpainting may refer to a technology that naturally restores and reattaches missing or damaged areas when processing an image. Inpainting can fill in missing areas based on surrounding data of the missing areas to make the image appear complete. When generating a seamless map, if inpainting technology is used, the edges of the texture can be adjusted to be well connected to the center, thereby preventing the formation of a seam when the image is repeated. In other words, even if the texture map is assumed to extend infinitely, the same pattern can be repeated without forming a seam.

[0136] The electronic device (900) can generate a composite texture map using a synthesis algorithm (540) for the diffuse map and normal map generated by the artificial neural network model. For example, the artificial neural network model can generate a 'Blue coin dot on canvas texture map (550)' using the 'Blue coin dot diffuse map (531)' and the 'Canvas normal map (532)' using the synthesis algorithm (540). The synthesis algorithm (540) may be an algorithm that utilizes the method referred to in FIG. 2 described above.

[0137] The electronic device (900) can display the 'Blue coin dot on canvas texture map (550)' and the 'Canvas normal map (532)' generated in response to the user's input together in the result window (560) of the user interface.

[0138]

[0139] Figure 6 schematically illustrates a process for an artificial neural network model to obtain a normal map according to one embodiment.

[0140] The description referring to FIGS. 1 to 5 can be equally applied to FIG. 6, and overlapping content can be omitted.

[0141] An electronic device (900) according to one embodiment may include an artificial neural network model capable of generating a normal map. The artificial neural network model (602) may be trained by a machine learning device (601) to generate a normal map based on learning data (610). The machine learning device (601) may include a first encoder (621) capable of outputting a first latent vector (622) from learning data (610) (e.g., which may refer to a preset in the above descriptions), and a first decoder (623) capable of outputting a learning result (640) from the latent first vector (622).

[0142] For example, referring to FIG. 6, the machine learning device (601) may receive training data (610) including data in which the names of multiple fabric types (e.g., Canvas (611), Corduroy (612), French terry (613), etc.) are paired (or labeled) with normal maps of the fabric types.

[0143] The machine learning device (601) can extract a first latent vector (622) from the received learning data (610). The first latent vector (622) is a concept used in machine learning and deep learning models, and can refer to a vector representing the characteristics of data. The first latent vector (622) can compress or summarize the characteristics of the original data, and can simply express a complex data structure. Accordingly, the first encoder (621) can extract important information from data such as images, and thereby extract the first latent vector (622) so that the first decoder (623) of the machine learning device can learn the characteristics of the data.

[0144] The machine learning device (601) can extract learning results (640) from the first latent vector (622) using the first decoder (623). The first decoder (623) can be trained (or updated) by the second decoder (634) to output a normal map suitable for input of the user's fabric type based on the first latent vector (622), the learning result (640), and the learning data (610). In this process, the first latent vector (622) can also be trained (or updated) by the second latent vector (633) to include more appropriate image features.

[0145] The artificial neural network model (602) may be modeled to include a second decoder (634) and a second latent vector (633) learned from a machine learning device (601). The artificial neural network model (602) may include a second encoder (631) capable of performing type one hot encoding, a multi-layer perceptron (MLP) (632) that outputs a second latent vector (633), and a second decoder (634) that outputs a normal map corresponding to a user's input from the second latent vector (633).

[0146] Type one hot encoding may be a data preprocessing technique used to input categorical data into a machine learning model. For example, assuming that the categories representing the fabric types, such as training data (610), are ['Canvas (611)', 'Corduroy (612)', 'French Terry (613)'], 'Canvas (611)' can be encoded as [1, 0, 0], 'Corduroy (612)' as [0, 1, 0], and 'French Terry (613)' as [0, 0, 1] to digitize the categorical data. In this case, if the user inputs 'French Terry', the encoder (631) can use type one hot encoding to encode the user's input as [0, 0, 1].

[0147] A multilayer perceptron (632) can receive an encoded user input and extract a latent vector (633). For example, the encoded user input [0, 0, 1] can be received and extracted as a latent vector (633) for 'French Terry (613)'.

[0148] The decoder (634) can output the 'French Terry (613) normal map', which is a learning result, using the latent vector (633) for 'French Terry (613)'. Consequently, the electronic device (900) can display the 'French Terry (613) normal map' generated by the artificial neural network model (602) on the user interface.

[0149]

[0150] FIGS. 7 to 8d are schematic drawings illustrating a user interface according to one embodiment.

[0151] The description with reference to FIGS. 1 to 6 can be equally applied to FIGS. 7 to 8d, and overlapping content can be omitted.

[0152] Referring to FIG. 7, in operation (150), the electronic device (900) can simulate a synthetic texture map on a three-dimensional virtual garment and display the three-dimensional virtual garment with the synthetic texture map applied through the user interface (700).

[0153] A user interface (700) according to one embodiment may include an avatar (720), a simulation area (701) which is a virtual clothing display area, and a texture generator interface (a first texture generator interface (710) or a second texture generator interface (810)). A user may command an electronic device (900) to simulate fabric through the texture generator interface (710).

[0154] Referring to FIGS. 8A to 8C, the first texture generation interface (710) may include a Fabric Type option (711), a Content Type option (712), a Pattern Size option (713), a prompt (714), and a composite texture map area (715) that displays a composite texture map.

[0155] The fabric type option (711) provides preset options for the normal map learned by the electronic device (900), and the user can select a desired fabric type from the fabric type option (711). Accordingly, the electronic device (900) can generate a normal map based on the correlation between the learned preset and the normal map.

[0156] The options selected in the content type options (712) may be options related to other information to be added to the composite texture map, excluding information about the diffuse map and information about the normal map.

[0157] The option selected in the pattern size option (713) may refer to an option regarding the size of patterns to be displayed in the composite texture map.

[0158] Prompt (714) may be an area where a user inputs information about the pattern of the fabric. In the described embodiment, text is input for convenience of explanation, but as described above, the user may also input an image.

[0159] As in FIG. 8b, when a user specifies the fabric type option (711) as 'Canvas', inputs the text 'blue coin dot' in the prompt (714), and clicks the Generate button, the electronic device (900) can receive the user inputs of the first texture generation interface (710) and generate a composite texture map of 'Blue coin dot on Canvas' in the composite texture map area (715).

[0160] Referring to FIG. 8c, when a user clicks the 'Apply' button in the first texture creation interface (710), the electronic device (900) can receive the user input and simulate a virtual costume (721) with a synthetic texture map applied to the avatar (720). At this time, the user can modify the properties of the synthetic texture map through the property window interface (730).

[0161] Referring to FIG. 8d, the second texture generation interface (810) may be implemented as an interactive interface. The user may input text to the second texture generation interface (810) to generate (811) a texture map of a canvas material in which 'blue coin dot' is expressed. Alternatively, the user may input something like 'rough and rough texture' instead of inputting the name 'canvas'.

[0162] The electronic device (900) that receives the user's input may output the composite texture map (814) without going through an intermediate step, but may also first display the normal map (812) of the canvas according to the user's settings or commands. In this case, the electronic device (900) may output a feedback message to the user (e.g., whether the desired target fabric type is correct, or output a candidate group of presets to guide a selection) so that the user can perform real-time adjustments to the input data. The electronic device (900) that receives the user's input for confirming (813) the normal map of the canvas may display the composite texture map (814), which is a 'blue coin dot canvas', on the second texture generation interface (810).

[0163] However, even in this case, if the pattern is not what the user wants and a different synthetic texture map is desired, or if the user wants a synthetic texture map with more information added by inputting additional information (e.g., information related to the material properties of the fabric, such as elasticity and softness), additional work can be performed to generate the final synthetic texture map.

[0164] The electronic device (900) can receive a command or message to apply (815) a user's synthetic texture map (814), and simulate a virtual costume (721) of an avatar (720) in which the synthetic texture map (814) is expressed. Similarly to FIGS. 8A to 8C, a property window interface (730) for the applied synthetic texture map (814) can be displayed to allow the user to modify the properties of the synthetic texture map.

[0165] The embodiments described in FIGS. 8A to 8D express an avatar (720) wearing a three-dimensional virtual costume (721), but are not limited thereto, and two-dimensional pattern pieces may be expressed to express synthetic texture maps.

[0166]

[0167] Figure 9 is a block diagram of an electronic device according to one embodiment.

[0168] Referring to FIG. 9, an electronic device (900) according to one embodiment may include a processor (930), a memory (950), and an output device (970) (e.g., a display). The processor (930), the memory (950), and the output device (970) may be connected to each other via a communication bus (905). In the above-described process, for convenience of explanation, the electronic device (900) may include a processor (930) for performing at least one method or an algorithm corresponding to at least one method described above.

[0169] The output device (970) may display a user interface (700) that receives user input regarding status and operations related to texture map generation provided by the processor (930).

[0170] The memory (950) can store data acquired from the texture map generation algorithm, sensors, and communication unit performed by the processor (930). In addition, the memory (950) can store various pieces of information generated during the processing of the processor (930) described above. In addition, the memory (950) can store various types of data and programs. The memory (950) can include volatile memory or non-volatile memory. The memory (950) can store various types of data by including a large-capacity storage medium such as a hard disk.

[0171] In addition, the processor (930) can perform at least one method or an algorithm corresponding to at least one method described above with reference to FIGS. 1 to 8. The processor (930) may be a data processing device implemented as hardware having a circuit having a physical structure for executing desired operations. For example, the desired operations may include code or instructions included in a program. The processor (930) may be configured as, for example, a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), or an NPU (Neural Network Processing Unit). For example, the electronic device (900) implemented as hardware may include a microprocessor, a central processing unit, a processor core, a multi-core processor, a multiprocessor, an ASIC (Application-Specific Integrated Circuit), and an FPGA (Field Programmable Gate Array).

[0172] The processor (930) can execute a program and control the electronic device (900). The program code executed by the processor (930) can be stored in the memory (950).

[0173] The processor (930) receives input data related to a target fabric from a user, classifies the input data into mapping data for generating texture maps corresponding to information of the target fabric, inputs the mapping data into an artificial neural network model to generate at least one of a normal map and a diffuse map, and can generate a synthetic texture map corresponding to the target fabric based on the normal map or the diffuse map.

[0174]

[0175] The learning process and operation algorithms described above can be performed on a server including the electronic device (900) described above and communicated with a user terminal so that they can be simulated through a user interface (700) displayed by the user terminal.

[0176] For example, a server may operate a service platform that provides a simulation of virtual clothing, and a user may utilize the service platform to utilize a service that generates a synthetic texture map. Accordingly, a user may utilize a user interface of the service platform via a user terminal to transmit input data, and the server may transmit, through the server and the service platform, the simulation results of the synthetic texture map generated based on the user's input received via the aforementioned electronic device (900), thereby enabling the user terminal to display the virtual clothing.

[0177] Alternatively, the user terminal may be an electronic device (900) that performs the aforementioned operation algorithm and method.

[0178]

[0179] The embodiments described above may be implemented using hardware components, software components, and / or a combination of hardware components and software components. For example, the devices, methods, and components described in the embodiments may be implemented using a general-purpose computer or a special-purpose computer, such as, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate array (FPGA), a programmable logic unit (PLU), a microprocessor, or any other device capable of executing instructions and responding to them. The processing device may execute an operating system (OS) and software applications running on the operating system. Furthermore, the processing device may access, store, manipulate, process, and generate data in response to the execution of the software. For ease of understanding, the processing device is sometimes described as being used alone; however, one of ordinary skill in the art will recognize that the processing device may include multiple processing elements and / or multiple types of processing elements. For example, a processing unit may include multiple processors, or a processor and a controller. Other processing configurations, such as parallel processors, are also possible.

[0180] Software may include a computer program, code, instructions, or a combination of one or more of these, which may configure a processing device to perform a desired operation or may, independently or collectively, command the processing device. The software and / or data may be permanently or temporarily embodied in any type of machine, component, physical device, virtual equipment, computer storage medium or device, or transmitted signal wave, for interpretation by the processing device or for providing instructions or data to the processing device. The software may also be distributed over networked computer systems and stored or executed in a distributed manner. The software and data may be stored on a computer-readable recording medium.

[0181] The method according to the embodiment may be implemented in the form of program commands that can be executed through various computer means and recorded on a computer-readable medium. The computer-readable medium may include program commands, data files, data structures, etc., alone or in combination, and the program commands recorded on the medium may be those specially designed and configured for the embodiment or may be known and available to those skilled in the art of computer software. Examples of the computer-readable recording medium include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, magneto-optical media such as floptical disks, and hardware devices specially configured to store and execute program commands such as ROMs, RAMs, and flash memories. Examples of program commands include not only machine language codes such as those generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter, etc.

[0182] The hardware device described above may be configured to operate as one or more software modules to perform the operations of the embodiment, and vice versa.

[0183] Although the embodiments described above have been described with limited drawings, those skilled in the art will appreciate that various technical modifications and variations can be applied based on the described embodiments. For example, appropriate results can still be achieved even if the described techniques are performed in a different order than described, and / or components of the described systems, structures, devices, circuits, etc. are combined or combined in a different manner than described, or are replaced or substituted with other components or equivalents.

[0184] Therefore, other implementations, other embodiments, and equivalents to the claims also fall within the scope of the claims described below.

Claims

1. An action to receive input data related to a target fabric from a user; An operation of classifying the above input data into mapping data for generating texture maps corresponding to information of the target fabric; An operation of inputting the above mapping data into an artificial neural network model to generate at least one of a normal map and a diffuse map; and An operation of generating a synthetic texture map corresponding to the target fabric based on the normal map or the diffuse map. A method for generating a texture map, comprising:

2. In paragraph 1, The action of receiving the above input data is When receiving text data entered by the user, an operation of detecting information of the target fabric based on the text data A method for generating a texture map, comprising:

3. In paragraph 1, The operation of receiving the above input data is an operation of detecting information of the target fabric included in the image data when receiving the image data input by the user. A method for generating a texture map, comprising:

4. In paragraph 1, The action of classifying with the above mapping data is An operation to classify the information of the above target fabric into normal map information, diffuse map information, or other information. Including, A method for generating a texture map, wherein the normal map information includes information related to the type of fabric, and the diffuse map information includes information related to the pattern expression of the fabric.

5. In paragraph 1, The above artificial neural network model A method for generating a texture map, comprising a generative artificial neural network model that generates the above texture maps as seamless maps.

6. In paragraph 1, The above artificial neural network model A method for generating a texture map, comprising a machine learning model that learns a correlation between a type of fabric and the normal map to generate a preset.

7. In paragraph 6, The action of classifying with the above mapping data is An operation of generating the mapping data based on the preset when the preset matches the input data. A method for generating a texture map, comprising:

8. In paragraph 6, The action of classifying with the above mapping data is If the above input data and the above preset do not match, an operation of outputting mapping data related to the target fabric based on the above input data. A method for generating a texture map, comprising:

9. In paragraph 8, The action of outputting mapping data related to the above target fabric is An operation of outputting a candidate group of presets similar to the normal map information when the mapping data and the preset do not match and the mapping data is normal map information. A method for generating a texture map, comprising:

10. In paragraph 1, The action of generating the above texture map is An operation of processing the above normal map into a filter image and transforming the diffuse map using the filter image; and An operation of blending the above-described modified diffuse map and the above-described filter image. A method for generating a texture map, comprising:

11. In paragraph 9, The above blending action is An operation of blending the filter image and the transformed diffuse map through image compositing. A method for generating a texture map, comprising:

12. In paragraph 1, Further comprising an action of inputting the above mapping data into an artificial neural network model to generate another map, The operation of generating the above synthetic texture map is: An operation of generating a final composite texture map corresponding to the target fabric based on at least one of the normal map, the diffuse map, and the other map, A method for generating a texture map, wherein the above other maps are generated based on normal map information and other information.

13. In paragraph 1, An operation of simulating the above-mentioned synthetic texture map on a 3D virtual garment and displaying the 3D virtual garment to which the above-mentioned synthetic texture map is applied through a user interface. A method for generating a texture map, further comprising:

14. In paragraph 1, An action of outputting a feedback message to the user so that the user can make real-time adjustments to the input data. A method for generating a texture map, further comprising:

15. In an electronic device simulating a virtual costume, Memory that stores instructions; and One or more processors Including, The above instructions, when executed by the one or more processors, cause the electronic device to: Receive input data related to the target fabric from the user, Classify the above input data into mapping data for generating texture maps corresponding to information of the target fabric, By inputting the above mapping data into an artificial neural network model, at least one of a normal map and a diffuse map is generated, An electronic device that generates a synthetic texture map corresponding to the target fabric based on the normal map or the diffuse map.

Images