System and method for providing customized image

Abstract

Proposed are a system and method for providing a customized image, the system for providing a customized image comprises an image generation device that generates an image according to a request of an operator terminal of a shopping mall operator, and the image generation device comprises: a control unit that extracts unique data and variation data from shopping mall information corresponding to a shopping mall and product information about one or more products sold in the shopping mall, generates a plurality of prompts by selectively combining the variation data with the unique data, and generates a plurality of images corresponding to the respective generated prompts; and a memory storing the plurality of images generated by the control unit.

Description

System and method for providing customized images to customers

[0001] The embodiments disclosed in this specification relate to a system and method for providing a customer-tailored image, and more particularly, to a system and method for providing a customer-tailored image that generates multiple images of a product based on shopping mall information and product information, and displays the customized images to a customer who has accessed an online shopping mall.

[0002] Customers want online shopping malls to offer them personalized products, but the product images available are limited. Online shopping mall operators hire models for each product, take photos of them wearing the item, and create a single, representative image and detail page for each product. Each time a new product is released, operators must repeat this process of hiring models, taking photos of them wearing the item, creating a representative image, and creating a detail page.

[0003] Operators face the problem that the time and cost of marketing activities increase linearly with the number of products sold. Consequently, the need for technology that efficiently generates multiple images of a product, regardless of the number of products sold, has emerged.

[0004] In this regard, prior art document, Korean Patent Publication No. 10-2023-0029156, describes a device for automatically generating product detailed images that applies target platform information to a detailed image learning model to output product detailed image information with an optimized promotional effect corresponding to the target platform information.

[0005] However, even with this, it fails to satisfy both the needs of customers who visit online shopping malls and those of operators who want to save time and money on marketing activities by efficiently creating product images.

[0006] Therefore, a technology to solve the above-mentioned problems became necessary.

[0007] Meanwhile, the background technology described above is technical information that the inventor possessed for the purpose of deriving the present invention or acquired during the process of deriving the present invention, and cannot necessarily be said to be publicly known technology disclosed to the general public prior to the application for the present invention.

[0008] The embodiments disclosed in this specification aim to present a system and method for providing a customer-tailored image that can provide a product image tailored to each customer based on shopping mall information and product information.

[0009] The embodiments disclosed herein aim to present a system and method for providing customer-tailored images that can automatically generate multiple product promotional images matching one product.

[0010] The embodiments disclosed in this specification aim to present a system and method for providing customer-tailored images that can improve marketing performance by saving time and costs for marketing activities of online shopping mall operators.

[0011] The embodiments disclosed in this specification are intended to provide a system and method for providing customer-tailored images that can increase the immersion and interest of online shopping for customers.

[0012] The embodiments disclosed in this specification aim to present a system and method for providing customer-tailored images that can improve work productivity by reducing the cost of producing promotional images even when the operator's products are updated or added, and increase advertising performance relative to the investment.

[0013] As a technical means for achieving the above-described technical task, according to one embodiment, a customer-customized image providing system including an image generating device that generates an image according to a request from an operator terminal of a shopping mall operator, wherein the image generating device may include a control unit that extracts unique data and variable data from shopping mall information corresponding to the shopping mall and product information on one or more products sold in the shopping mall, selectively combines the variable data with the unique data to generate a plurality of prompts, and generates a plurality of images corresponding to each of the plurality of generated prompts; and a memory that stores the plurality of images generated by the control unit.

[0014] According to another embodiment, a method for providing a customized image, performed by a customized image providing system including an image generating device that generates an image according to a request from an operator terminal of a shopping mall operator, may include a step in which the image generating device extracts unique data and variable data from shopping mall information corresponding to the shopping mall and product information on one or more products sold in the shopping mall; a step in which the image generating device selectively combines the variable data with the unique data to generate a plurality of prompts; and a step in which the image generating device generates a plurality of images corresponding to each of the plurality of generated prompts.

[0015] According to another embodiment, a computer-readable recording medium having recorded thereon a program for performing a method for providing a customized image, the method being performed by a customized image providing system including an image generating device that generates an image according to a request from an operator terminal of a shopping mall operator, the method for providing a customized image may include a step in which the image generating device extracts unique data and variable data from shopping mall information corresponding to the shopping mall and product information on one or more products sold in the shopping mall; a step in which the image generating device selectively combines the variable data with the unique data to generate a plurality of prompts; and a step in which the image generating device generates a plurality of images corresponding to each of the plurality of generated prompts.

[0016] Furthermore, according to another embodiment, a method for providing a customized image is performed by a customized image providing system including an image generating device that generates an image according to a request from an operator terminal of a shopping mall operator. The method for providing a customized image may include a step in which the image generating device extracts unique data and variable data from shopping mall information corresponding to the shopping mall and product information on one or more products sold in the shopping mall; a step in which the image generating device selectively combines the variable data with the unique data to generate a plurality of prompts; and a step in which the image generating device generates a plurality of images corresponding to each of the plurality of generated prompts.

[0017] According to any one of the aforementioned problem solving methods, a system and method for providing a customer-tailored image can be proposed that can provide a product image tailored to each customer based on shopping mall information and product information.

[0018] According to any one of the aforementioned problem solving means, a system and method for providing customer-tailored images that can automatically generate multiple product promotional images matching one product can be proposed.

[0019] According to any one of the aforementioned problem solving methods, a system and method for providing customer-tailored images can be proposed that can improve marketing performance by saving time and cost for marketing activities of online shopping mall operators.

[0020] According to any one of the aforementioned problem solving methods, a system and method for providing customer-tailored images that can increase customer immersion and interest in online shopping can be proposed.

[0021] According to any of the aforementioned problem solving methods, a system and method for providing customer-tailored images can be proposed that can reduce the cost of producing promotional images even when the operator's products are updated or added, thereby improving work productivity and increasing advertising performance relative to the investment.

[0022] The effects that can be obtained from the disclosed embodiments are not limited to the effects mentioned above, and other effects that are not mentioned can be clearly understood by a person having ordinary skill in the art to which the disclosed embodiments belong from the description below.

[0023] FIG. 1 is a diagram illustrating a network configuration of a customer-tailored image provision system according to one embodiment.

[0024] FIG. 2 is a block diagram illustrating an image generation device and a shopping mall server according to one embodiment.

[0025] FIGS. 3 and 4 are exemplary diagrams for explaining a method for an image generation device to generate an image in a customer-tailored image provision system according to one embodiment.

[0026] Figures 5 and 6 are flowcharts for explaining a method for providing a customer-customized image according to one embodiment.

[0027] Below, various embodiments are described in detail with reference to the attached drawings. The embodiments described below may be modified and implemented in various different forms. To more clearly explain the features of the embodiments, detailed descriptions of matters widely known to those skilled in the art to which the embodiments pertain below have been omitted. In addition, parts of the drawings that are not related to the description of the embodiments have been omitted, and similar parts have been designated with similar drawing reference numerals throughout the specification.

[0028] Throughout the specification, when a component is said to be "connected" to another component, this includes not only the "direct connection" but also the "connection with other components in between." Furthermore, when a component is said to "include" another component, this does not exclude other components, but rather implies that other components may be included, unless otherwise specifically stated.

[0029] The embodiments will be described in detail with reference to the attached drawings below.

[0030] FIG. 1 is a schematic diagram illustrating a customer-tailored image provision system according to one embodiment.

[0031] The customer-customized image provision system (10) illustrated in FIG. 1 is a system that automatically generates multiple images for products in an online shopping mall according to a customer-customized image provision method and exposes at least one of the generated images to a user who accesses the online shopping mall.

[0032] This customer-tailored image provision system (10) may include an image generation device (100), a shopping mall server (200), an operator terminal (300) of a shopping mall operator, and a user terminal (400).

[0033] An image generation device (100) is a device that provides a service to a shopping mall operator that automatically generates images of products sold in a shopping mall or images for the promotion of a shopping mall.

[0034] Such an image generation device (100) can be implemented as a server and can interact with multiple operators running an online shopping mall. The image generation device (100) can be implemented as a computer capable of communicating over a network with an operator terminal (300), which is an electronic terminal equipped with an application or web browser for interaction with the operator, or can be implemented as a cloud computing server. The image generation device (100) can include a storage device capable of storing data, or can store data through a third-party server.

[0035] In addition, the shopping mall server (200) is a device that provides a web page for selling products and provides a service for selling products through the web page to users. It is a device that displays product images or promotional images to users by selecting product images or promotional images for selling products and configuring a web page according to the user who accessed the shopping mall.

[0036] A shopping mall server (200) such as this can access an online shopping mall and interact with users who wish to purchase a product among multiple products registered within the online shopping mall and the shopping mall operator. The shopping mall server (200) can be implemented as a computer capable of communicating via a network with an operator terminal (300) and a user terminal (400), which is an electronic terminal equipped with an application or web browser for user interaction, or can be implemented as a cloud computing server. In addition, the shopping mall server (200) can include a storage device capable of storing data, or can store data via a third-party server.

[0037] The operator terminal (300) is a device for an operator of an online shopping mall to interact with an image generation device (100) to request image generation, or to interact with a shopping mall server (200) to set images to be displayed in the shopping mall. In addition, the user terminal (400) is a device for interacting with the shopping mall server (200) to selectively view product images and detailed information posted on the online shopping mall, purchase products, save product information in a list of products of interest or a shopping cart list, or share product information.

[0038] Here, the operator terminal (300) and user terminal (400) can be implemented as a computer, portable terminal, television, wearable device, etc. that can access a remote server through a network (N) or connect to other terminals and servers. Here, the computer may include, for example, a notebook, desktop, or laptop equipped with a web browser, and the portable terminal may include, for example, all kinds of handheld-based wireless communication devices such as a PCS (Personal Communication System), PDC (Personal Digital Cellular), PHS (Personal Handyphone System), PDA (Personal Digital Assistant), GSM (Global System for Mobile communications), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), Wibro (Wireless Broadband Internet), Smart Phone, and Mobile WiMAX (Mobile Worldwide Interoperability for Microwave Access), which are wireless communication devices that ensure portability and mobility. In addition, the television may include IPTV (Internet Protocol Television), Internet TV, terrestrial TV, cable TV, etc. Furthermore, wearable devices are information processing devices that can be directly worn on the human body, such as watches, glasses, accessories, clothing, and shoes, and can be connected to a remote server or other terminals via a network, either directly or through another information processing device.

[0039] FIG. 2 is a block diagram illustrating an image generation device (100) and a shopping mall server (200) according to one embodiment. Referring to FIG. 2, the image generation device (100) according to one embodiment may include a memory (110), a communication unit (120), and a control unit (130).

[0040] The memory (110) may store and install various types of data, such as files, applications, and programs. Specifically, the memory (110) may store shopping mall information and product information. At this time, "shopping mall information" refers to information about the characteristics of the shopping mall, and may include background information and image atmosphere information, which serve as the basis for the shopping mall operator to create multiple images to be included in the shopping mall webpage. At this time, "background information" refers to information about the time or space in which the products sold in the shopping mall are used, and "image atmosphere information" refers to information about the brightness, saturation, and color settings of the image. The shopping mall information is set by the shopping mall operator and may also include information about the shopping mall's main target customer group. "Information about the shopping mall's main target customer group" refers to information about the user group that primarily uses the products being sold among multiple user groups classified according to predetermined classification criteria, such as women in their 30s and men in their 40s. Furthermore, the shopping mall information may include information about users who have accessed the shopping mall over a certain period of time, i.e., incoming customers. This 'information on incoming customers' can be set by the operator and collected from the shopping mall server (200). Meanwhile, the above-described shopping mall information may optionally include image data in addition to text data, and may be stored in the memory (110).

[0041] "Product Information" provides detailed information about the main product categories offered by the shopping mall. It may include product classification, purpose, size, material, color, user information about the intended wearer or user of the product, and search terms associated with the product. Optionally, "Product Information" may also include product image data, such as product images or design drawings. This "shopping mall information" and "product information" can be extracted as unique data and variable data, as described below.

[0042] Additionally, a plurality of images generated by a control unit to be described later can be stored in the memory (110).

[0043] The communication unit (120) can perform wired or wireless communication with other devices or networks. To this end, the communication unit (120) may include a communication module that supports at least one of various wired or wireless communication methods. For example, the communication module may be implemented in the form of a chipset.

[0044] The wireless communication supported by the communication unit (120) may be, for example, Wi-Fi (Wireless Fidelity), Wi-Fi Direct, Bluetooth, UWB (Ultra-Wide Band), or NFC (Near Field Communication). In addition, the wired communication supported by the communication unit (120) may be, for example, USB or HDMI (High Definition Multimedia Interface).

[0045] The communication unit (120) can perform wired and wireless communication with the operator's operator terminal (300) through a network.

[0046] The control unit (130) controls the overall operation of the image generation device (100) and may include a processor such as a CPU, GPU, etc. When the control unit (130) receives an image generation request from the operator terminal (300), it may control other components included in the image generation device (100) to perform a method for providing a customer-tailored image.

[0047] For example, the control unit (130) may execute a program stored in the memory (110), read a file stored in the memory (110), or store a new file in the memory (110).

[0048] According to one embodiment, the control unit (130) may receive an image generation request including data included in shopping mall information and product information from the operator terminal (300). To this end, the control unit (130) may receive shopping mall information and product information from the operator terminal (300) via the communication unit (120) and store the received shopping mall information and product information in the memory (110). The control unit (130) may transmit a user interface for receiving data included in shopping mall information and product information from the operator to the operator terminal (300).

[0049] In another embodiment, the control unit (130) may crawl a shopping mall's webpage to collect at least one of shopping mall information and product information. Specifically, the control unit (130) may crawl the shopping mall's webpage based on the shopping mall's URL information. In another embodiment, the control unit (130) may communicate with the shopping mall server (200) to collect at least one of shopping mall information and product information from the storage of the shopping mall server (200).

[0050] The control unit (130) can extract unique data and variable data from shopping mall information corresponding to the shopping mall and product information for one or more products sold in the shopping mall.

[0051] In this case, "unique data" refers to data that remains unchanged from shopping mall and product information. For example, information on the location or time period where products sold at a shopping mall are primarily used can be extracted as unique data.

[0052] Additionally, "variable data" refers to data that selectively changes from shopping mall and product information. For example, information about product options, such as the size and color of a product sold in a shopping mall, can be extracted as variable data.

[0053] The control unit (130) can generate multiple prompts by selectively combining variable data with unique data. In this case, 'prompt' means a string generated by selectively combining unique data and variable data.

[0054] Specifically, the control unit (130) can generate multiple prompts by inputting unique data and variable data into a prompt generation model. At this time, the 'prompt generation model' is an artificial intelligence model that extracts information and creates a string, and may be, for example, a Generative Pre-trained Transformer (GPT), a Text-To-Text Transfer Transformer (T5), or a Bidirectional Encoder Representations from Transformers (BERT).

[0055] The control unit (130) can generate multiple images corresponding to each of the generated multiple prompts. Specifically, the control unit (130) can generate multiple images by inputting each of the multiple prompts into an image generation model. At this time, the 'image generation model' is an artificial intelligence model that generates new images using strings and image data, and may be, for example, Generative Adversarial Networks (GANs), Variational Autoencoders (VAEs), DALL-E, and Contrastive Language-Image Pretraining (CLIP).

[0056] The control unit (130) can generate one or more images in response to each of a plurality of prompts, and can store one or more images in the memory (110) by associating corresponding variation data with the generated one or more images. The control unit (130) can assign identification information to one or more images and store them in the memory (110).

[0057] The method by which the control unit (130) generates an image based on shopping mall information and product information will be described below with reference to FIGS. 3 and 4. FIGS. 3 and 4 are exemplary diagrams for explaining the method by which an image generation device generates an image in a customer-tailored image provision system according to one embodiment.

[0058] Referring to FIG. 3, the control unit (130) may extract text data to be included in a plurality of categories constituting a prompt based on shopping mall information and product information. According to one embodiment, the plurality of categories constituting the prompt may include a background category (301), a product category (302), a model motion category (307), and a model category (308). Since the control unit (130) can generate a more detailed image as the information included in the prompt becomes more specific, the plurality of categories constituting the prompt is not limited thereto and may be added or deleted. According to one embodiment, some of the categories constituting the prompt may be deleted or added depending on the purpose or settings of the image to be generated.

[0059] The control unit (130) can set text data to be input into the background category (301) based on the shopping mall information set by the operator. For example, if the operator runs a shopping mall that sells golf wear and golf-related items, and the time when the golf wear and golf-related items are used is 'fall' and the place is 'golf course', the control unit (130) can extract text data such as 'fall golf course' or 'fall golf course' as unique data from the online shopping mall information. The control unit (130) can set the extracted unique data as text data to be input into the background category (302). In addition, if the time and place of use of products sold by the operator through the shopping mall are two or more, the control unit (130) can extract a plurality of variable data, for example, text data such as 'winter camping site' and 'fall golf course', from the online shopping mall information, and set text data to be input into the background category (301) by selecting one of the extracted variable data.

[0060] The control unit (130) can set text data to be entered into the product category (302) based on product information of products sold by the operator through the shopping mall. Specifically, the category (302) for the product can include, but is not limited to, a material category (303), a size category (304), a color category (305), and a product identification information category (306). For example, if the product sold by the operator through the online shopping mall is 'B Golf Gloves', the control unit (130) can extract text data titled 'B Golf Gloves' or an identification number assigned to 'B Golf Gloves' as unique data from the product information. The control unit (130) can set the extracted unique data as text data to be entered into the product identification information category (306).

[0061] In addition, if the material of 'B golf gloves' is a single option of material A, the control unit (130) can extract text data such as 'material A' or 'of material A' from the product information as unique data. The control unit (130) can set the extracted unique data as text data to be entered into the material category (303). If the material of 'B golf gloves' changes depending on the option, the control unit (130) can extract multiple text data, for example, text data such as 'material A' and 'material B', from the product information as variable data. The control unit (130) can set text data to be entered into the material category (303) by selecting one of the extracted variable data.

[0062] Furthermore, if 'B Golf Gloves' are manufactured and sold as 'XL size', 'L size', 'M size', and 'S size' depending on the option, the control unit (130) can extract text data of 'XL size', 'L size', 'M size', and 'S size' respectively as variable data from the product information. The control unit (130) can select one of the extracted variable data and set the text data to be entered into the size category (304). If a product is manufactured and sold in only a single size, the control unit (130) can extract text data for one size, for example, 'FREE size', from the product information as unique data, and set the extracted unique data as text data to be entered into the size category (304).

[0063] Furthermore, if 'B Golf Gloves' are manufactured and sold in 'white color' or 'black color' depending on the option, the control unit (130) can extract text data 'white color' and 'black color' from the product information as variable data. The control unit (130) can select one of the extracted variable data and set the text data to be entered into the color category (305). If a product is manufactured and sold only in a single red color, the control unit (130) can extract the text data 'red color' from the product information as unique data and set the extracted unique data as the text data to be entered into the color category (305).

[0064] The control unit (130) can set text data to be input into the model's action category (307) based on the shopping mall information and product information set by the operator. Specifically, the control unit (130) can select one or more actions from among multiple actions that the model takes when using the product based on the shopping mall information and the product information. If there is only one selected action, the control unit (130) can extract text corresponding to the selected action as unique data and set the extracted unique data as text data to be input into the model's action category (307). If there are multiple selected actions, the control unit (130) can extract each text corresponding to the selected action as variable data and set text data to be input into the model's action category (307) by selecting one of the extracted variable data.

[0065] For example, if the product sold by the operator is 'golf gloves', the control unit (130) can select one of the multiple motions performed when wearing the golf gloves, specifically, the motion of putting, and based on the selected motion, the control unit (130) can extract the text data 'putting' as unique data. The control unit (130) can set the extracted unique data as text data to be input into the motion category (307) of the model. If the control unit (130) selects the motion of putting and the motion of cheering among the multiple motions performed when wearing the golf gloves, the control unit (130) can extract the text data 'putting' and 'cheering' as variation data, respectively. The control unit (130) can select one of the extracted variation data 'putting' and 'cheering' and set the text data to be input into the motion category (307) of the model.

[0066] The control unit (130) can set text data to be entered into the model category (308) from shopping mall information. Specifically, the control unit (130) can set text data to be entered into the model category (308) based on information about the main target customer group of the product being sold, set by the operator. The model category (308) can include an age category (309) and a gender category (310).

[0067] For example, if the main target customer groups of a product sold by an operator are women in their 30s and women in their 40s, the control unit (130) can extract the text data '30s' and '40s' as variable data from the information on the main target customer groups of the product, and can extract the text data 'female' as unique data. The control unit (130) can select one of the extracted variable data '30s' and '40s' and set the text data to be input into the age category (309), and set the extracted unique data 'female' as the text data to be input into the gender category (310) of the model.

[0068] If the main target customer group of the product sold by the operator is men and women in their 20s, the control unit (130) can extract the text data '20s' as unique data from the information on the main target customer group of the product, and extract the text data 'male' and 'female' respectively as variable data. The control unit (130) can set the extracted text data '20s' as text data to be input into the age category (309) of the model, and can set the text data to be input into the gender category (310) of the model by selecting either 'male' or 'female' extracted as variable data.

[0069] The control unit (130) can extract text data corresponding to a plurality of categories forming a prompt as described above, corresponding to each of a plurality of products sold by the operator through an online shopping mall, as unique data or variable data.

[0070] The control unit (130) can generate a prompt by combining extracted unique data and variable data. Specifically, the control unit (130) can generate a prompt by setting text to be entered in all categories using the extracted unique data and variable data for each of the multiple products sold by the operator in the online shopping mall.

[0071] The control unit (130) can generate multiple prompts by combining unique data and variable data of all options for one product among multiple products sold by the operator in the online shopping mall. For example, as illustrated in FIG. 3, the control unit (130) can input unique data and variable data extracted for 'B golf gloves' into each category to generate multiple prompts such as 'A woman in her 40s putting on a golf course in the fall while wearing a B golf glove, made of material A, size XL, and in white' and 'A woman in her 30s putting on a golf course in the fall while wearing a B golf glove, made of material A, size L, and in black'.

[0072] According to one embodiment, the control unit (130) can generate multiple images corresponding to each of a plurality of prompts. To this end, as illustrated in (a) of FIG. 4, the control unit (130) can input collected shopping mall information and product information into the prompt generation model (101) to generate prompts. The control unit (130) can input the generated prompts and product images into the image generation model (102) to generate multiple images associated with the products.

[0073] That is, as shown in (b) of FIG. 4, the control unit (130) can input a prompt (401) generated from the prompt generation model (101) and an image (402) of a product into the image generation model (102) to obtain an image (403) associated with the product as output.

[0074] In another embodiment, the control unit (130) can extract an image from at least one of shopping mall information and product information, input the extracted image and a prompt generated from the prompt generation model (101) into the image generation model (102), and obtain an image for the product as output.

[0075] The control unit (130) can generate different images depending on the purpose of use of the image. Specifically, when generating an image to be inserted into a product detail page, the image can be generated by inputting the product image and a prompt into the image generation model (102). Meanwhile, when generating a thumbnail image, banner image, or advertisement image of a shopping mall, the image can be generated by inputting only the prompt generated by the prompt generation model (101) into the image generation model (102).

[0076] Referring again to FIG. 2, a shopping mall server (200) according to one embodiment may include a storage (210), a communication unit (220), and a controller (230).

[0077] Various types of data, such as files, applications, and programs, can be installed and stored in the storage (210). Specifically, a plurality of images generated by the image generation device (100) can be stored in the storage (210).

[0078] The communication unit (220) can perform wired or wireless communication with other devices or networks. To this end, the communication unit (220) may include a communication module that supports at least one of various wired or wireless communication methods. For example, the communication module may be implemented in the form of a chipset.

[0079] The wireless communication supported by the communication unit (220) may be, for example, Wi-Fi (Wireless Fidelity), Wi-Fi Direct, Bluetooth, UWB (Ultra-Wide Band), or NFC (Near Field Communication). In addition, the wired communication supported by the communication unit (220) may be, for example, USB or HDMI (High Definition Multimedia Interface).

[0080] The communication unit (220) can perform wired and wireless communication with the operator's operator terminal (300) through a network.

[0081] The controller (230) controls the overall operation of the shopping mall server (200) and may include a processor such as a CPU, GPU, etc. For example, the controller (230) may execute a program stored in the storage (210), read a file stored in the storage (210), or store a new file in the storage (210).

[0082] The controller (230) can store multiple images generated by the image generation device (100) in the storage (210). To this end, the image generation device (100) can transmit the multiple images generated to the operator terminal (300), and the communication unit (220) can receive the multiple images from the operator terminal (300).

[0083] The controller (230) can configure a webpage to be provided to the user terminal (400) by selecting an image corresponding to at least one of the user information of the user corresponding to the user terminal (400) accessing the shopping mall and the search word entered by the user. To this end, the controller (230) can receive user information from the user terminal (400) accessing the shopping mall.

[0084] At this time, the controller (230) can select an image based on the variable data stored in association with each of the plurality of images, user information, and search words searched by the user.

[0085] In addition, the controller (230) may collect user feedback data on images constituting a web page provided to the user terminal (400) and provide the data to the control unit (130). At this time, the 'feedback data' may include information on user actions, including clicking on a product image included in a web page provided to a user who has accessed a shopping mall, saving product information of a product corresponding to a product image to a list of products of interest or a shopping cart list, paying for a product corresponding to a product image, and sharing product information of a product corresponding to a product image. That is, the controller (230) may collect feedback data on images and store them in the storage (210). The controller (230) may transmit the collected feedback data to the control unit (130) or to the operator terminal (300).

[0086] The control unit (130) can additionally train at least one of the prompt generation model (101) and the image generation model (102) based on feedback data about the image. To this end, the control unit (130) can receive feedback data from the shopping mall server (200) or the operator terminal (300) via the communication unit (120). For example, the control unit (130) can additionally train the image generation model (102) by inputting images on which a user action has occurred a predetermined number or more among the images exposed to the user into the image generation model (102). In addition, the control unit (130) can additionally train the prompt generation model by inputting a combination of unique data and variable data used to generate images on which a user action has occurred a predetermined number or more among the images exposed to the user into the prompt generation model (101).

[0087] The controller (230) can classify users of a shopping mall into multiple groups based on preset criteria. For example, the controller (230) can classify users of a shopping mall into multiple groups such as 'female in her 20s', 'male in her 20s', 'female in her 30s', 'male in her 30s', 'female in her 40s', and 'male in her 40s' based on age and gender. The control unit (130) can further train the prompt generation model (101) and the image generation model (102) based on feedback information of the groups. Specifically, the control unit (130) can change at least one of the unique data and the variable data used to generate the image provided to a group having less than a predetermined number of actions to the unique data and the variable data used to generate the image provided to a group having more than a predetermined number of actions, thereby further training the prompt generation model (101). The control unit (130) can additionally train the image generation model (101) by inputting a prompt generated from the additionally learned prompt generation model (101) into the image generation model (102).

[0088] The controller (230) may select related product images stored in the storage (210) based on feedback information of a group to which a user corresponding to a user terminal (400) accessing the shopping mall belongs. For example, when a user corresponding to a user terminal (400) accessing the shopping mall is a 33-year-old woman and is currently viewing a webpage on which images of women's golf wear are displayed, the controller (230) may select, as a related product image, an image of "B Golf Gloves," which is another product, among images of products other than golf wear, in the group "Women in their 30s" to which the user corresponding to the user terminal (400) belongs, and which has a preset number of clicks, a preset number of times saved in a list of products of interest or a shopping cart, a preset number of times purchased, or a preset number of times shared.

[0089] The controller (230) can configure a web page to be provided to a user terminal (400) using the selected related product image.

[0090] According to another embodiment, the control unit (130) may receive a request for creating an advertisement image for a shopping mall from the operator terminal (300). The control unit (130) may input shopping mall information and product information into the prompt generation model (101) to generate a prompt. The control unit (130) may input the generated prompt into the image generation model (102) to generate an advertisement image as output. The control unit (130) may transmit the generated advertisement image to the operator terminal (300), or may receive information about a third-party server (not shown) providing an advertisement service from the operator terminal (300) and transmit it to the third-party server. Through this configuration, an advertisement image for a shopping mall can be generated even if the operator does not access the shopping mall to generate the advertisement image.

[0091] Through this configuration, when a user accesses a third-party server providing an advertising service, at least one advertising image generated based on user information can be selected, and a webpage to be provided to the user terminal (400) can be configured using the selected advertising image. In this case, the advertising image can include shopping mall information, for example, information regarding the URL of the shopping mall, so that the user can access the shopping mall operated by the operator.

[0092] FIG. 5 is a flowchart for explaining a method for providing a customer-tailored image according to one embodiment, and FIG. 6 is a flowchart for explaining a method for providing a customer-tailored image performed between an image generation device (100), a shopping mall server (200), an operator terminal (300), and a user terminal (400) according to one embodiment.

[0093] The method for providing a customized image according to the embodiment illustrated in FIGS. 5 and 6 includes steps that are processed in time series in the customized image providing system (10) illustrated in FIGS. 1 to 4. Therefore, even if the content is omitted below, the content described above regarding the customized image providing system (10) illustrated in FIGS. 1 to 4 can also be applied to the customized image providing method according to the embodiment illustrated in FIGS. 5 and 6.

[0094] Referring to FIG. 5, the image generation device (100) can extract unique data and variable data from shopping mall information corresponding to a shopping mall and product information regarding one or more products sold at the shopping mall (S510). To this end, the image generation device (100) can receive an image generation request including data included in the shopping mall information and product information from the operator terminal (300). At this time, the product information may optionally include product image data. According to one embodiment, the image generation device (100) can transmit a user interface for receiving data included in the shopping mall information and product information from the operator to the operator terminal (300).

[0095] In another embodiment, the image generation device (100) may crawl a shopping mall's webpage to collect at least one of shopping mall information and product information. Specifically, the image generation device (100) may crawl the shopping mall's webpage based on the shopping mall's URL information. In another embodiment, the image generation device (100) may communicate with the shopping mall server (200) to collect at least one of shopping mall information and product information from the storage (210) of the shopping mall server (200).

[0096] The image generation device (100) can generate multiple prompts by selectively combining variable data with unique data (S520). Specifically, the image generation device (100) can generate multiple prompts by inputting unique data and variable data into a prompt generation model (101).

[0097] The image generation device (100) can generate multiple images corresponding to each of the plurality of prompts (S530). Specifically, the image generation device (100) can generate one or more images corresponding to each of the plurality of prompts. According to one embodiment, the image generation device (100) can generate multiple images by inputting multiple prompts into the image generation model. In this case, the multiple images generated by the image generation device (100) can be used as advertising images.

[0098] According to another embodiment, the image generation device (100) can input a prompt generated from a prompt generation model (101) and a product image into the image generation model (102) to generate multiple images associated with a product. In this case, the images associated with the product generated by the image generation device (100) can be used as thumbnail images, banner images, or product images of a shopping mall.

[0099] The image generation device (100) can store a plurality of images by associating corresponding fluctuation data for the plurality of images (S540). The image generation device (100) can transmit the generated plurality of images to the operator terminal (300). According to one embodiment, the image generation device (100) can generate one or more images corresponding to each of the plurality of prompts, and the image generation device (100) can store one or more images by associating corresponding fluctuation data for one or more images.

[0100] The shopping mall server (200) can store multiple images. To this end, the shopping mall server (200) can receive multiple images from the operator terminal (300) or from the image generation device (100).

[0101] The shopping mall server (200) can configure a web page to be provided to the user terminal (400) by selecting user information corresponding to the user terminal (400) accessing the shopping mall and an image corresponding to at least one of the search terms entered by the user. To this end, the shopping mall server (200) can receive user information from the user terminal (400) accessing the shopping mall.

[0102] At this time, the shopping mall server (200) can select an image based on the variable data stored in association with each of the multiple images, user information, and search words searched by the user.

[0103] Additionally, the shopping mall server (200) can collect user feedback data on images that make up a web page provided to a user terminal (400) and provide the data to an image generation device (100).

[0104] The shopping mall server (200) can collect feedback data about images. The shopping mall server (200) can transmit the collected feedback data to the image generation device (100) or to the operator terminal (300).

[0105] The image generation device (100) can additionally train at least one of the prompt generation model (101) and the image generation model (102) based on feedback data about the image. To this end, the image generation device (100) can receive feedback data from the shopping mall server (200) or the operator terminal (300). For example, the image generation device (100) can additionally train the image generation model (102) by inputting images exposed to the user, on which a user action has occurred a predetermined number of times or more, into the image generation model (102). In addition, the image generation device (100) can additionally train the prompt generation model by inputting a combination of unique data and variable data used to generate images exposed to the user, on which a user action has occurred a predetermined number of times or more, into the prompt generation model (101).

[0106] The shopping mall server (200) can classify users of the shopping mall into a plurality of groups according to preset criteria. The image generation device (100) can further train the prompt generation model (101) and the image generation model (102) based on feedback information of the groups. Specifically, the image generation device (100) can further train the prompt generation model (101) by changing at least one of the unique data and the variable data used to generate the image provided to the group having less than a predetermined number of actions into the unique data and the variable data used to generate the image provided to the group having more than a predetermined number of actions. The image generation device (100) can further train the image generation model (101) by inputting the prompt generated from the additionally trained prompt generation model (101) into the image generation model (102).

[0107] The shopping mall server (200) can select related product images stored in the shopping mall server (200) based on feedback information of a group to which a user corresponding to a user terminal (400) accessing the shopping mall belongs. The shopping mall server (200) can configure a web page to be provided to the user terminal (400) using the selected related product images.

[0108] According to another embodiment, the image generation device (100) can receive a request for producing an advertisement image of a shopping mall from an operator terminal (300). The image generation device (100) can input shopping mall information and product information into a prompt generation model (101) to generate a prompt. The image generation device (100) can input the generated prompt into an image generation model (102) to generate an advertisement image as an output. The image generation device (100) can transmit the generated advertisement image to the operator terminal (300), or receive information about a third-party server (not shown) providing an advertisement service from the operator terminal (300) and transmit it to the third-party server.

[0109] FIG. 6 is a flowchart for explaining a method for providing a customer-customized image performed between an image generation device (100), a shopping mall server (200), an operator terminal (300), and a user terminal (400) according to one embodiment.

[0110] As illustrated in Figure 6, the operator terminal (300) can request the image generation device (100) to generate an image (S601). The image generation device (100) can generate multiple images of a product (S602). At this time, when receiving a request for the generation of an advertising image from the operator terminal (300), in addition to multiple images of the product, an image of the shopping mall can also be generated and output.

[0111] The image generation device (100) can transmit the generated image to the operator terminal (300) (S603). The operator terminal (300) can upload the generated image to the shopping mall server (200) (S604). The shopping mall server (200) can store multiple images received from the operator terminal (300) (S605).

[0112] A user terminal (400) accesses an online shopping mall (S606), and the shopping mall server (200) can confirm the user terminal's (400) access to the shopping mall webpage. The shopping mall server (200) can select an image corresponding to user information and configure a webpage (S607). The user terminal (400) can input a search term into the webpage provided by the shopping mall server (200) (S609).

[0113] The shopping mall server (200) can configure a web page by selecting an image corresponding to user information and a search term (S610). The shopping mall server (200) can provide the web page to the user terminal (400) (S611).

[0114] The shopping mall server (200) can collect feedback information from users (S612). The shopping mall server (200) can transmit the feedback information to the image generation device (100) (S613). According to another embodiment, the shopping mall server (200) can transmit the feedback information to the operator terminal (300), and the operator terminal (300) can transmit the feedback information to the image generation device (100).

[0115] The image generation device (100) can additionally train at least one of the prompt generation model (101) and the image generation model (102).

[0116] The shopping mall server (200) can select a related product image based on feedback information of a group to which a user corresponding to a user terminal accessing the shopping mall belongs, and can provide a web page including the related product image (S615).

[0117] The term '~ unit' used in the above embodiments means a software or hardware component such as an FPGA (field programmable gate array) or an ASIC, and the '~ unit' performs certain roles. However, the '~ unit' is not limited to software or hardware. The '~ unit' may be configured to be on an addressable storage medium or may be configured to play one or more processors. Thus, as an example, the '~ unit' includes components such as software components, object-oriented software components, class components, and task components, processes, functions, attributes, procedures, subroutines, segments of program code, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays, and variables.

[0118] The functionality provided within the components and '~sub-components' may be combined into a smaller number of components and '~sub-components' or separated into additional components and '~sub-components'.

[0119] Additionally, components and '~parts' may be implemented to regenerate one or more CPUs within a device or secure multimedia card.

[0120] The method for providing a customized image according to the embodiments described through FIGS. 5 and 6 may also be implemented in the form of a computer-readable medium that stores computer-executable commands and data. In this case, the commands and data may be stored in the form of program codes, and when executed by a processor, may generate a predetermined program module to perform a predetermined operation. In addition, the computer-readable medium may be any available medium that can be accessed by a computer, and includes both volatile and nonvolatile media, removable and non-removable media. In addition, the computer-readable medium may be a computer recording medium, and the computer recording medium may include both volatile and nonvolatile, removable and non-removable media implemented by any method or technology for storing information such as computer-readable commands, data structures, program modules, or other data. For example, the computer recording medium may be a magnetic storage medium such as an HDD or an SSD, an optical recording medium such as a CD, a DVD, and a Blu-ray disc, or a memory included in a server that is accessible via a network.

[0121] In addition, the method for providing a customized image according to the embodiment described through FIGS. 5 and 6 may be implemented as a computer program (or computer program product) including computer-executable commands. The computer program includes programmable machine commands processed by a processor, and may be implemented in a high-level programming language, an object-oriented programming language, an assembly language, a machine language, etc. In addition, the computer program may be recorded on a tangible computer-readable recording medium (e.g., a memory, a hard disk, a magnetic / optical medium, or a solid-state drive (SSD), etc.).

[0122] Accordingly, the method for providing a customized image according to the embodiment described through FIGS. 5 and 6 can be implemented by executing the above-described computer program on a computing device. The computing device may include at least some of a processor, memory, a storage device, a high-speed interface connecting the memory and a high-speed expansion port, and a low-speed interface connecting the low-speed bus and the storage device. Each of these components is connected to one another using various buses and may be mounted on a common motherboard or in another suitable manner.

[0123] Here, the processor can process instructions within the computing device, such as instructions stored in a memory or storage device to display graphical information for providing a graphical user interface (GUI) on an external input / output device, such as a display connected to a high-speed interface. In another embodiment, multiple processors and / or multiple buses may be utilized, as appropriate, together with multiple memories and memory types. The processor may also be implemented as a chipset comprising multiple independent analog and / or digital processors.

[0124] Memory also stores information within a computing device. For example, memory may consist of volatile memory units or a collection of volatile memory units. For another example, memory may consist of nonvolatile memory units or a collection of nonvolatile memory units. Memory may also be another form of computer-readable media, such as magnetic or optical disks.

[0125] A storage device can provide a large amount of storage space to a computing device. The storage device can be a computer-readable medium or a configuration including such a medium, and can include, for example, devices within a storage area network (SAN) or other configurations, and can be a floppy disk device, a hard disk device, an optical disk device, a tape device, flash memory, or other similar semiconductor memory device or device array.

[0126] The embodiments described above are provided for illustrative purposes only, and those skilled in the art will readily appreciate that the embodiments described above can be readily modified into other specific forms without altering the technical concepts or essential characteristics of the embodiments described above. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive. For example, components described as being single may be implemented in a distributed manner, and similarly, components described as being distributed may be implemented in a combined manner.

[0127] The scope of protection sought through this specification is indicated by the claims described below rather than the detailed description above, and should be interpreted to include all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts.

Claims

1. In a customer-customized image provision system including an image generation device that generates an image according to a request from an operator terminal of a shopping mall operator, the image generation device, A control unit that extracts unique data and variable data from shopping mall information corresponding to the shopping mall and product information for one or more products sold in the shopping mall, selectively combines the variable data with the unique data to generate a plurality of prompts, and generates a plurality of images corresponding to each of the generated plurality of prompts; and A system for providing customized images, comprising a memory for storing a plurality of images generated by the above control unit.

2. In paragraph 1, The above control unit, A system for providing a customized image, which generates one or more images corresponding to each of the plurality of prompts, associates corresponding variation data with respect to the generated one or more images, and stores the one or more images in the memory.

3. In paragraph 1, The above control unit, A system for providing customized images, wherein the plurality of prompts are generated by inputting the above-mentioned unique data and the above-mentioned variable data into a prompt generation model, and the plurality of prompts are each input into an image generation model to generate the plurality of images.

4. In paragraph 1, The above control unit, A system for providing customized images to customers, which crawls the webpage of the above shopping mall and collects at least one of the above shopping mall information and the above product information.

5. In paragraph 3, The above customer-tailored image provision system is, Further comprising a shopping mall server that communicates with multiple user terminals and provides the shopping mall to the user terminals, The above shopping mall server is, A system for providing customized images to a customer, which stores the plurality of images, selects an image corresponding to user information of a user corresponding to a user terminal accessing the shopping mall and at least one of search words entered by the user, and configures a web page to be provided to the user terminal.

6. In paragraph 5, The above shopping mall server is, Collecting the user's feedback data on the images that make up the webpage provided to the user terminal and providing the data to the image generation device, The above control unit, A system for providing customized images, wherein at least one of the prompt generation model and the image generation model is further trained based on feedback data for the image.

7. In paragraph 6, The above shopping mall server is, A system for providing customized images to a customer, which classifies users of a shopping mall into multiple groups according to preset criteria, selects related product images based on feedback information of the group to which a user corresponding to a user terminal accessing the shopping mall belongs, and configures a web page to be provided to the user terminal.

8. A method for providing a customized image to a customer, performed by a customized image providing system including an image generating device that generates an image according to a request from an operator terminal of a shopping mall operator, A step in which the image generation device extracts unique data and variable data from shopping mall information corresponding to the shopping mall and product information on one or more products sold at the shopping mall; A step in which the image generation device selectively combines the variable data with the unique data to generate a plurality of prompts; and A method for providing a customized image, comprising a step of generating a plurality of images corresponding to each of the plurality of generated prompts by the image generating device.

9. In paragraph 8, The step of generating the above multiple images is: A step in which the image generating device generates one or more images corresponding to each of the plurality of prompts; and A method for providing a customized image, comprising a step of storing one or more images by associating corresponding variation data with one or more generated images by the image generating device.

10. In paragraph 8, The step of generating the above multiple prompts is: The image generating device includes a step of generating the plurality of prompts by inputting the unique data and the variable data into a prompt generating model, The step of generating the above multiple images is: A method for providing a customized image, wherein the image generation device includes a step of generating the plurality of images by inputting the plurality of prompts into an image generation model, respectively.

11. In paragraph 8, The steps of extracting the above unique data and variable data are: A method for providing a customer-tailored image, comprising a step of the image generation device crawling the webpage of the shopping mall to collect at least one of the shopping mall information and the product information.

12. In paragraph 10, The above customer-tailored image provision system further includes a shopping mall server that communicates with multiple user terminals and provides the shopping mall to the user terminals. The above method of providing customized images to customers is as follows: The above shopping mall server stores the plurality of images; and A method for providing a customized image to a customer, further comprising a step of the shopping mall server selecting an image corresponding to at least one of user information of a user corresponding to a user terminal accessing the shopping mall and a search word entered by the user, and configuring a web page to be provided to the user terminal.

13. In paragraph 12, The above method of providing customized images to customers is as follows: A step in which the shopping mall server collects the user's feedback data on the image that constitutes the web page to be provided to the user terminal and provides the data to the image generation device; and A method for providing a customized image, wherein the image generation device further comprises a step of additionally training at least one of the prompt generation model and the image generation model based on feedback data for the image.

14. In paragraph 13, The above method of providing customized images to customers is as follows: A step in which the shopping mall server classifies users of the shopping mall into multiple groups according to preset criteria; and A method for providing a customer-tailored image, further comprising a step of the shopping mall server selecting a related product image based on feedback information of a group to which a user corresponding to a user terminal accessing the shopping mall belongs and configuring a web page to be provided to the user terminal.

15. A computer-readable recording medium having recorded thereon a program for performing the method described in Article 8.

Images