Method, apparatus and program for providing visual cue-based prompt for image generation

Abstract

A method for providing a prompt for image generation based on a visual queue according to various embodiments of the present disclosure is disclosed. The method may include: a step of providing a visual queue selection area including one or more component-specific visual queues and obtaining user input for creating a prompt for image generation; a step of displaying text corresponding to the user input in a prompt panel area in conjunction with obtaining the user input; and a step of storing a prompt corresponding to the currently selected visual queue in the clipboard of a user terminal when a selection input for a copy button included in the prompt panel area is obtained.

Description

Technology Field

[0001] The present disclosure relates to a method, apparatus, and program for providing a prompt for generating a visual queue-based image. Background Technology

[0002] The emergence of Generative AI technology, particularly models that generate images based on text, is bringing revolutionary changes to the fields of visual design and art education. Advanced AI models such as DALL-E and Midjourney are recognized as powerful tools that enable the rapid visualization of ideas and diverse creative experiments, as they allow for the generation of high-quality visual results with just simple text input. This technological advancement has lowered the barrier to entry, enabling anyone—even non-experts—to participate in creative image production, and it is establishing itself as a core element of the creative thinking process.

[0003] However, the core of image generation using generative AI lies in the process of creating prompts that clearly and specifically convey the user's intent, and this process remains difficult for many users due to its complexity and abstractness.

[0004] The commonly used sentence-based prompt method causes a high cognitive burden because users must directly describe various visual attributes of an image, such as the subject, composition, style, and lighting, in detail using sentences.

[0005] In addition, novice users often go through repeated trial and error to achieve the desired results due to a lack of knowledge regarding the technical terms and various expression techniques required for effective image generation. This process can lead to attempts to produce standardized results rather than creative and original ideas, and also raises the issue that it is difficult to predict the outcome due to the discrepancy between abstract linguistic expressions and the actual generated images.

[0006] Therefore, there is a demand in the industry for a new interface that enhances the efficiency and intuitiveness of creating prompts for image generation. In this regard, Korean Published Patent No. 10-2024-0150019 discloses a prompt conversion device and method for image generation. The problem to be solved

[0007] The present disclosure is devised in response to the aforementioned background technology and aims to provide a method, apparatus, and program for providing a prompt for generating a visual queue-based image.

[0008] The technical problems of the present disclosure are not limited to those mentioned above, and other unmentioned technical problems will be clearly understood by those skilled in the art from the description below. means of solving the problem

[0009] According to one embodiment of the present disclosure for solving the problem described above, a method for providing a prompt for image generation based on a visual queue is disclosed. The method may include: a step of providing a visual queue selection area including one or more component-specific visual queues and obtaining user input for creating a prompt for image generation; a step of displaying text corresponding to the user input in a prompt panel area in conjunction with obtaining the user input; and a step of storing a prompt corresponding to the currently selected visual queue in the clipboard of a user terminal when a selection input for a copy button included in the prompt panel area is obtained.

[0010] In an alternative embodiment, the one or more components include at least one of an object, material and texture, background, composition, style, color and mood, and lighting corresponding to the image to be generated, and the visual cue selection area is composed of one or more interfaces corresponding to the one or more components, and the visual cue may include a plurality of images corresponding to the currently displayed interface and text corresponding to each of the plurality of images.

[0011] In an alternative embodiment, the step of providing a visual queue selection area including one or more component-specific visual queues and obtaining user input for creating a prompt for image generation may include: obtaining a selection input for a specific visual queue; or obtaining a specific text input while a specific component is displayed.

[0012] In an alternative embodiment, the prompt panel area includes a list corresponding to one or more components, and when a specific visual queue included in a specific component is selected or specific text is entered, the specific visual queue or the specific text may be updated to be added to an item in the list corresponding to the specific component.

[0013] In an alternative embodiment, the prompt panel area further includes one or more component-specific attribute buttons, and the method may further include the step of storing a prompt corresponding to the selected attribute button in the clipboard of a user terminal when a selection input for the attribute button is obtained.

[0014] In an alternative embodiment, the method may further include the step of acquiring a mouseover input for a specific visual queue among the one or more component-specific visual queues; and the step of enlarging and displaying an image included in the specific visual queue in response to the mouseover input.

[0015] In an alternative embodiment, the step of providing a visual queue selection area including one or more component-specific visual queues and obtaining user input for creating a prompt for image generation may further include: when a first selection input for a first component is obtained, determining a recommended visual queue for a second component associated with a first visual queue corresponding to the first selection input; and when providing an interface corresponding to the second component, arranging and displaying a plurality of visual queues corresponding to the second component based on the recommended visual queue.

[0016] In an alternative embodiment, the method may further include the steps of: storing the prompt corresponding to the currently selected visual queue in the clipboard of a user terminal, and then obtaining an output image corresponding to the prompt; generating a new visual queue based on the output image; determining at least one component among a plurality of components included in the prompt to be mapped to the new visual queue; and mapping the new visual queue to the at least one component to add the new visual queue to the interface of each of the at least one component.

[0017] According to one embodiment of the present disclosure for solving the above-described problem, an apparatus is disclosed. The apparatus comprises: a memory for storing one or more instructions; and a processor for executing the one or more instructions stored in the memory, and the processor can perform the above-described methods by executing the one or more instructions.

[0018] According to one embodiment of the present disclosure for solving the above-described problem, a computer program stored on a computer-readable recording medium is disclosed, which is coupled with a computer as hardware to perform the above-described methods.

[0019] Other specific details of the present disclosure are included in the detailed description and drawings. Effects of the invention

[0020] The present disclosure can provide an environment in which even users without specialized knowledge can efficiently and accurately create prompts for an image generation AI through an intuitive visual interface.

[0021] The effects of the present disclosure are not limited to those mentioned above, and other unmentioned effects will be clearly understood by a person skilled in the art from the description below. Brief explanation of the drawing

[0022] FIG. 1 is a drawing illustrating a system according to one embodiment of the present disclosure. FIG. 2 is a hardware configuration diagram of a computing device according to one embodiment of the present disclosure. FIGS. 3 to 8 are drawings for explaining a method for providing a prompt for generating a visual queue-based image according to one embodiment of the present disclosure. Specific details for implementing the invention

[0023] Various embodiments are now described with reference to the drawings. In this disclosure, various descriptions are provided to facilitate understanding of the disclosure. However, it is evident that these embodiments can be practiced without such specific descriptions.

[0024] As used in this disclosure, terms such as “component,” “module,” “system,” etc. refer to computer-related entities, hardware, firmware, software, combinations of software and hardware, or executions of software. For example, a component may be, but is not limited to, a procedure executed on a processor, a processor, an object, an execution thread, a program, and / or a computer. For example, both an application executed on a computing device and the computing device itself may be a component. One or more components may reside within a processor and / or an execution thread. A component may be localized within a single computer. A component may be distributed among two or more computers. Additionally, these components may be executed from various computer-readable media having various data structures stored therein. Components may communicate through local and / or remote processes, for example, according to signals having one or more data packets (e.g., data from a component interacting with another component in a local system or distributed system, and / or data transmitted through signals to other systems and networks such as the Internet).

[0025] Furthermore, the term "or" is intended to mean an implicit "or" rather than an exclusive "or." That is, unless otherwise specified or evident from the context, "X uses A or B" is intended to mean one of the natural implicit substitutions. In other words, if X uses A; if X uses B; or if X uses both A and B, "X uses A or B" may apply to any of these cases. Additionally, the term "and / or" as used in this disclosure should be understood to refer to and include all possible combinations of one or more of the listed related items.

[0026] Additionally, the terms “comprising” and / or “comprising” should be understood to mean that such features and / or components are present. However, the terms “comprising” and / or “comprising” should be understood not to exclude the presence or addition of one or more other features, components and / or groups thereof. Furthermore, unless otherwise specified or clearly evident from the context to indicate a singular form, the singular in this disclosure and claims should generally be interpreted to mean “one or more.”

[0027] Those skilled in the art should recognize that the various exemplary logical blocks, configurations, modules, circuits, means, logics, and algorithmic steps described in connection with the embodiments disclosed herein may be implemented in electronic hardware, computer software, or a combination of both. To clearly exemplify the interchangeability of hardware and software, various exemplary components, blocks, configurations, means, logics, modules, circuits, and steps have been generally described above in terms of their functionality. Whether such functionality is implemented in hardware or software depends on the specific application and design constraints imposed on the overall system. Skilled technicians may implement the described functionality in various ways for each specific application. However, such decisions regarding implementation should not be construed as going beyond the scope of this disclosure.

[0028] Description of the presented embodiments is provided to enable those skilled in the art to use or practice the present disclosure. Various modifications to these embodiments will be apparent to those skilled in the art. The general principles defined herein may be applied to other embodiments without departing from the scope of the present disclosure. Thus, the present disclosure is not limited to the embodiments presented herein. The present disclosure should be interpreted in the broadest possible scope consistent with the principles and novel features presented herein.

[0029] In the present disclosure, "computer" refers to any type of hardware device comprising at least one processor, and may be understood to include software configurations operating on said hardware device according to embodiments. For example, "computer" may be understood to include smartphones, tablet PCs, desktops, laptops, and user clients and applications running on each of these devices, but is not limited thereto.

[0030] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the attached drawings.

[0031] Each step described in this disclosure is described as being performed by a computer, but the subject of each step is not limited thereto, and depending on the embodiment, at least some of each step may be performed on different devices.

[0032] FIG. 1 is a drawing illustrating a system according to one embodiment of the present disclosure.

[0033] Referring to FIG. 1, a system according to one embodiment of the present disclosure may include a computing device (100), a user terminal (200), and an external server (300). The system illustrated in FIG. 1 is according to one embodiment, and its components are not limited to the embodiment illustrated in FIG. 1 and may be added, changed, or deleted as needed.

[0034] In one embodiment, the computing device (100) may provide a prompt for generating an image based on a visual cue. For example, the computing device (100) may provide a web-based interface that allows a user to configure a prompt for generating an image simply by selecting a visual element. Specifically, the web-based interface may display a visual cue selection area configured in the form of a plurality of tabs or menus in one area of ​​the screen, and the visual cue selection area may include categories for image generation components, such as a subject, material and texture, background, composition, style, color tone and mood, lighting, etc. In each category, a plurality of visual cues may be presented in a combined form of image and text, and the user may select one or more visual cues through user input such as a mouse click, touch input, or keyboard selection.

[0035] Here, a visual cue may refer to a visual or multimodal representation element provided to enable a user to intuitively recognize and select visual attributes during the creation process of a prompt for image generation. A visual cue may include static images such as photos, icons, illustrations, graphics, and thumbnails, as well as dynamic content such as videos, animations, and GIFs; it may also include three-dimensional objects such as 3D models, augmented reality (AR) objects, and virtual reality (VR) objects. Furthermore, a visual cue may be implemented as a multimodal representation combined with text, audio, or haptic feedback, and these various forms may be provided individually or in combination. The visual cue may serve to explain or assist components of the prompt, such as objects, materials and textures, backgrounds, compositions, styles, colors and moods, and lighting, but is not limited thereto.

[0036] In the present disclosure, a visual cue may be a visual prompt element in the form of a combination of an image and text corresponding to the image, and the user can intuitively understand the meaning of the text through the image. The prompt may be a text-based command that instructs an image generation AI model to generate a specific image.

[0037] For example, a computing device (100) provides a component-specific interface such as a predefined target, style, composition, etc., and can obtain a user's selection input for a visual cue such as 'teenage girl', 'jumping motion', etc., among a plurality of visual cues included in each interface.

[0038] Text corresponding to the selected visual queue is reflected and accumulated in the prompt panel area in real time, and the prompt panel displays items by selected component in a list format, allowing the user to intuitively check the status of the prompts written so far.

[0039] For example, the computing device (100) can display the text ‘teenage girl’ in the ‘target’ item within the prompt panel area when the user selects the ‘teenage girl’ visual cue, and then additionally display the text ‘jumping action’ in the same item when the user selects the ‘jumping action’ visual cue.

[0040] Additionally, the user may, if necessary, click the 'Copy Text' button to save the entire accumulated prompt to the user terminal's clipboard, or select a specific attribute button to save only the part corresponding to that attribute to the clipboard.

[0041] For example, when a user selects a copy button while text such as 'Target: teenage girl, jumping motion', 'Style: 3D rendering' is displayed in the prompt panel area of ​​the computing device (100), the entire text can be saved to the clipboard of the user terminal (200) and provided as input to an external AI model.

[0042] Meanwhile, the user can generate a desired image by utilizing the prompt stored in the clipboard of the user terminal (200) and pasting it into an external image generation model.

[0043] Accordingly, the computing device (100) of the present disclosure can support even a user without specialized knowledge in generating an image generation prompt with only an intuitive selection.

[0044] In addition, the prompt panel provides the ability to delete all items at once, including a 'Select Reset' button, and allows for the removal of items individually by entering a deselect input for specific items.

[0045] Furthermore, the computing device (100) supports intuitive recognition of features such as detailed texture, composition, and color tone even within a limited display area by magnifying and displaying the resolution of the image when the user places a mouse over a specific visual cue. In addition, in an embodiment with added recommendation logic, when the user selects a specific style, associated lighting or background elements may be prioritized and displayed, thereby supporting a more complete prompt configuration.

[0046] Hereinafter, an example of a method in which a computing device (100) provides a prompt for generating a visual queue-based image is described with reference to FIGS. 3 to 8.

[0047] In various embodiments, the computing device (100) may provide Web or Application-based services. However, it is not limited thereto.

[0048] The computing device (100) may include any type of computer system or computer device, such as, for example, a microprocessor, a mainframe computer, a digital processor, a portable device, and a device controller. However, it is not limited thereto.

[0049] Hereinafter, the hardware configuration of the computing device (100) will be described with reference to FIG. 2.

[0050] Meanwhile, the user terminal (200) may be connected to the computing device (100) via a network (400) and may be a terminal of a user who uses an image generation prompt provided by the computing device (100). For example, the user terminal (200) may include a user who connects to the interface of the computing device (100) to select a visual queue, a user who copies text generated in the prompt panel area, and a user who transmits the copied text to an external image generation model.

[0051] Here, the user terminal (200) may include, for example, various types of computer devices. Specifically, for example, the user terminal (200) may refer to various terminal devices such as smartphones, tablet PCs, desktops, and laptops.

[0052] The user terminal (200) includes a display on at least a part of the terminal and may include an operating system for running applications or extension-based services provided by the computing device (100). For example, the user terminal (200) may be a smartphone, but is not limited thereto, and the user terminal (200) may include all kinds of handheld-based wireless communication devices such as navigation, PCS (Personal Communication System), GSM (Global System for Mobile communications), PDC (Personal Digital Cellular), PHS (Personal Handyphone System), PDA (Personal Digital Assistant), IMT (International Mobile Telecommunication)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (W-Code Division Multiple Access), Wibro (Wireless Broadband Internet) terminal, smartpad, tablet PC, etc., as wireless communication devices that ensure portability and mobility.

[0053] An external server (300) may be connected to a computing device (100) via a network (400), and may transmit and receive various information / data necessary for the computing device (100) to provide various functions, technologies, or services related to the method of the present disclosure, and may store and manage various information / data generated as the computing device (100) provides various functions, technologies, or services related to the method of the present disclosure.

[0054] For example, the external server (300) may be a database server that stores information used in various functions, technologies, or services related to the method of the present disclosure. As another example, the external server (300) may be a server that provides information used in various functions, technologies, or services related to the method of the present disclosure.

[0055] The network (400) may refer to a connection structure capable of exchanging information between each node, such as computing devices, multiple terminals, and servers. For example, the network (400) includes a Local Area Network (LAN), a Wide Area Network (WAN), the World Wide Web (WWW), a wired and wireless data network, a telephone network, a wired and wireless television network, etc.

[0056] Wireless data communication networks include, but are not limited to, 3G, 4G, 5G, 3GPP (3rd Generation Partnership Project), 5GPP (5th Generation Partnership Project), LTE (Long Term Evolution), WIMAX (World Interoperability for Microwave Access), Wi-Fi, Internet, LAN (Local Area Network), Wireless LAN (Wireless Local Area Network), WAN (Wide Area Network), PAN (Personal Area Network), RF (Radio Frequency), Bluetooth network, NFC (Near-Field Communication) network, satellite broadcasting network, analog broadcasting network, DMB (Digital Multimedia Broadcasting) network, etc.

[0057] FIG. 2 is a hardware configuration diagram of a computing device according to one embodiment of the present disclosure.

[0058] Referring to FIG. 2, a computing device (100) according to one embodiment of the present disclosure may include one or more processors (110), a memory (120) for loading a computer program (151) executed by the processor (110), a bus (130), a communication interface (140), and a storage (150) for storing the computer program (151). Here, FIG. 2 illustrates only the components related to the embodiment of the present disclosure. Accordingly, a person skilled in the art to which the present disclosure pertains will understand that other general-purpose components may be included in addition to the components illustrated in FIG. 2.

[0059] The processor (110) controls the overall operation of each component of the computing device (100). The processor (110) may be composed of one or more cores and may include various processors that support data analysis and deep learning operations, such as a Central Processing Unit (CPU), a General Purpose Graphics Processing Unit (GPGPU), and a Tensor Processing Unit (TPU). Alternatively, it may be configured to include any type of processor well known in the art of the present disclosure.

[0060] Additionally, the processor (110) may run at least one application or program for executing a method for providing a prompt for generating a visual queue-based image according to embodiments of the present disclosure. Specifically, the processor (110) may obtain a visual queue selection input or a text input from a user terminal, display text corresponding to the input in real time in a prompt panel area, and process a selection input for a copy button, an attribute button, a selection reset button, etc. included in the prompt panel. In addition, it may perform operations such as enlarging a specific visual queue image based on a mouse over input, and may also perform functions such as determining an associated recommended visual queue based on a selection input for a specific component, or creating a new visual queue based on the mapping result between the output image and the prompt and reflecting it in the system's database.

[0061] In various embodiments, the processor (110) may further include RAM (Random Access Memory) and ROM (Read-Only Memory) to temporarily and permanently store signals or data processed internally, and may be implemented in the form of a System on Chip (SoC) including at least one of a graphics processing unit, RAM, and ROM. This configuration can reliably provide a web-based or application-based user interface, process multiple visual queues in real time, and support a computational environment that analyzes and recommends large-scale image data and prompt text in parallel.

[0062] Memory (120) is a component that stores various data, commands, and information, and can load a computer program (151) from storage (150) to execute a method for providing a prompt for generating a visual queue-based image according to various embodiments of the present disclosure. When the computer program (151) is loaded into memory (120), the processor (110) can execute one or more instructions constituting the program to obtain a visual queue selection input or text input from a user terminal, display text corresponding to the selected visual queue in real time in a prompt panel area, and perform an operation to process selection inputs for a copy button, an attribute button, a selection reset button, etc. Additionally, memory (120) may store data for displaying a specific visual queue image in enlargement according to a mouse over input, association data for determining a recommended visual queue, and training data for generating a new visual queue by mapping an output image and a prompt.

[0063] The memory (120) can be implemented as volatile memory such as RAM, but is not limited thereto and can be implemented as various storage media such as ROM, flash memory, etc. This configuration can contribute to reliably providing a web-based or application-based interface, processing multiple visual queues in real time, and analyzing and managing large-scale image data and prompt text in parallel.

[0064] The bus (130) provides communication functions between components of the computing device (100). The bus (130) can be implemented as various types of buses, such as an address bus, a data bus, and a control bus.

[0065] The communication interface (140) supports wired and wireless internet communication of the computing device (100). In addition, the communication interface (140) may support various communication methods in addition to internet communication, such as a local area network (LAN), a wide area network (WAN), a mobile communication network (e.g., 4G, 5G), Wi-Fi, Bluetooth, and Near Field Communication (NFC). To this end, the communication interface (140) may be configured to include a communication module well known in the art of the present disclosure, and wired and wireless network interfaces may be used in parallel as needed.

[0066] Specifically, the communication interface (140) supports interaction between the user terminal (200) and the computing device (100) to receive user input signals, such as selected visual queue inputs, text inputs, and copy / initialization commands from the user terminal, and transmit updated prompt data based on the inputs to the user terminal. Additionally, the communication interface (140) may transmit and receive output images for creating new visual queues, update training data for determining recommended visual queues, or mapping data between prompts and image results through data exchange between the computing device (100) and an external server (300).

[0067] In some embodiments, the communication interface (140) may be omitted, in which case the computing device (100) may perform the visual queue-based prompt providing function offline while operating in a standalone application or local environment.

[0068] Storage (150) may store a computer program (151) non-temporarily. When performing a process according to an embodiment of the present disclosure through a computing device (100), storage (150) may store various information necessary to execute a method according to the disclosed embodiment or to provide a service. For example, storage (150) may store a plurality of images to be displayed in a visual queue selection area and text data corresponding to the images, a prompt record accumulated in a prompt panel, an association database for determining a recommended visual queue, a mapping result between an output image and a prompt, and training data for creating a new visual queue.

[0069] Storage (150) may be configured to include non-volatile memory such as ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), flash memory, a hard disk, a removable disk, or any form of computer-readable recording medium well known in the art to which this disclosure belongs.

[0070] A computer program (151) may include one or more instructions that cause a processor (110) to execute each step of a visual queue-based prompt provision method according to various embodiments of the present disclosure when loaded into memory (120). That is, by executing the instructions, the processor (110) may perform operations such as processing user input, updating a prompt panel, mouse-over enlargement display, copy processing by attribute, determining a recommended visual queue, and creating a new visual queue.

[0071] In one embodiment, the computer program (151) may include one or more instructions configured to perform various tasks related to the training of a neural network model, thereby enabling personalized recommendations or the creation of new queues based on the user's prompt selection pattern or output image data.

[0072] The steps of the method or algorithm described in connection with the embodiments of the present disclosure may be implemented directly in hardware, implemented as a software module executed by hardware, or implemented by a combination thereof. The software module may reside in RAM (Random Access Memory), ROM (Read Only Memory), EPROM (Erasable Programmable ROM), EEPROM (Electrically Erasable Programmable ROM), Flash Memory, a hard disk, a removable disk, a CD-ROM, or any form of computer-readable recording medium well known in the art to which the present disclosure belongs.

[0073] The components of the present disclosure may be implemented as a program (or application) and stored on a medium to be executed in combination with a computer, which is hardware. The components of the present disclosure may be implemented as software programming or software elements, and similarly, embodiments may be implemented in programming or scripting languages ​​such as C, C++, Java, assembler, etc., including various algorithms implemented as combinations of data structures, processes, routines, or other programming configurations. Functional aspects may be implemented as algorithms executed on one or more processors.

[0074] FIGS. 3 to 8 are drawings for explaining a method for providing a prompt for generating a visual queue-based image according to one embodiment of the present disclosure.

[0075] Referring to FIG. 3, the computing device (100) provides a visual queue selection area containing one or more component-specific visual queues and can obtain user input for creating a prompt for image generation (S110). Here, the user input may include, for example, an input selecting a specific component tab in the component selection area, an input specifying a specific category in the sub-category selection area, an input clicking or touching one or more visual queue images in the visual queue display area, or an input directly writing prompt elements through a text input area. Additionally, the user input may include an input requesting the enlargement of a specific visual queue by pointer actions such as mouse over, an input deleting or initializing some of the items displayed in the prompt panel, or a selection input to copy the entire prompt or the attribute-specific prompt to the clipboard.

[0076] Specifically, the computing device (100) can obtain a selection input for a specific visual cue. For example, if a user clicks or touches a ‘teacher’ visual cue among a plurality of visual cues displayed on an interface corresponding to a ‘target’ component, the computing device (100) recognizes this as a selection input for the corresponding visual cue. At this time, the selection input is not limited to a simple click or touch, but can be provided in various forms such as a double-click, drag and drop, stylus pen input, voice command, etc., and the system can process these inputs as visual cue selections.

[0077] When a selection input is obtained, the computing device (100) can extract text data mapped to the corresponding visual queue and automatically add it to the corresponding component item in the prompt panel area. For example, if the ‘teacher’ visual queue is selected, the text “teacher” can be reflected in real time in the ‘target’ item of the prompt panel. If the user subsequently selects the ‘jumping action’ visual queue, the corresponding text can be accumulated and recorded in the same item.

[0078] Furthermore, it can be implemented as a toggle mechanism where selecting the same visual cue again deselects the item and removes it from the prompt panel, and when multiple visual cues are selected consecutively, their respective texts can be accumulated to form a complex prompt. In this way, the selection input for a specific visual cue goes beyond a simple click action and serves as a core interface that supports users in effectively specifying desired images and intuitively creating prompts.

[0079] Additionally, the computing device (100) can obtain specific text input while a specific component is displayed. For example, if a user wants to add a prompt element that is not provided by default by the system while the 'target' component is displayed, the user can directly enter a phrase such as 'smiling doctor' into a separate text input window and select the 'Add' button. In this case, the computing device (100) recognizes the text as an item corresponding to the 'target' component and reflects it in the prompt panel area.

[0080] Text input can be allowed not only as a single word but also in the form of phrases or sentences, and the text entered by the user is immediately accumulated in the corresponding component items of the prompt panel. Additionally, if the user enters text into the same item multiple times, it can be added sequentially to form complex attributes, and can be modified through the functions of deleting individual items or resetting the entire list if necessary. Furthermore, the entered text is stored and managed internally by the system and can be utilized as training data for generating recommended visual cues or suggesting personalized prompts to the user.

[0081] More specifically, the computing device (100) may provide a user interface to a user terminal (200) that includes the visual queue selection area and the prompt panel area to be described later. And, the computing device (100) may obtain user input through the user interface.

[0082] One or more components of the present disclosure may be a classification system for systematically organizing and selecting visual attributes that constitute an image. For example, one or more components may include at least one of an object, material and texture, background, composition, style, color and mood, and lighting corresponding to the image to be created. Here, the 'object' component is an attribute that defines the central subject of the image, such as a person, object, animal, or character, and may include detailed elements such as age, gender, facial expression, and movement. The 'material and texture' component may include attributes representing visual and tactile characteristics, such as materials like metal, glass, wood, or fabric, as well as smooth or rough surfaces and shine. The 'background' component establishes the spatial context of the image, such as an indoor, outdoor, natural landscape, or urban environment, and may include mood control elements such as blur effects, perspective, and depth. The 'composition' component may form the spatial structure of the image by defining the placement of the subject or the viewpoint (e.g., front, side, bird's-eye view, close-up, etc.). The 'Style' component specifies artistic styles such as painterly, watercolor, cartoonish, 3D rendering, cyberpunk, and Gothic, and can reflect the atmosphere of the era or genre characteristics. The 'Color and Mood' component can determine the color palette and emotional atmosphere, including warm tones, cool tones, pastels, monotone, dramatic, and dreamy. Finally, the 'Lighting' component includes attributes regarding the direction, intensity, color, and diffusion method of light—such as natural light, studio lighting, soft light, backlighting, and colored lighting—and can influence the depth and impression of the image.

[0083] Additionally, the visual cue selection area may consist of one or more interfaces corresponding to one or more components. For example, each component, such as 'Object', 'Background', 'Composition', and 'Style', may be provided as an individual interface tab or menu, and when a user selects a specific tab, visual cues corresponding to that component may be displayed on the screen. In this case, each interface may include functions such as scrolling, filters, and search bars to support quick navigation to the desired item among multiple visual cues.

[0084] A visual cue may include multiple images corresponding to the currently displayed interface and text mapped to each of those images. For example, in the 'Style' component interface, multiple images representing different styles, such as watercolor, cyberpunk, and 3D rendering, may be presented as samples, and text such as "watercolor," "cyberpunk," and "3D rendering" may be displayed at the bottom of each image. Users can intuitively understand the visual attributes through the images, and when an image is selected, the corresponding text is automatically reflected in the prompt panel. Therefore, the visual cue functions not merely as a simple example image, but as a prompt configuration unit combining images and text.

[0085] That is, one or more components may include 'Subject' defining the core subject of the image, 'Material / Texture' describing the texture of the surface, 'Background' setting the spatial environment, 'Composition' determining the viewpoint and placement, 'Style' specifying the artistic style, 'Color / Mood' determining the overall color tone and atmosphere, and 'Lighting' setting light effects. The above components may be provided in the form of menus or tabs on the user interface, and the user can view a list of multiple visual cues corresponding to each component by selecting the relevant menu or tab. For example, when the 'Subject' tab is selected, visual cues such as people, animals, objects, and characters are displayed, and when the 'Style' tab is selected, visual cues expressing different artistic styles such as watercolor, 3D rendering, cartoon style, and cyberpunk may be displayed. In this way, the components form a systematic and intuitive classification system, supporting the user to select and combine various attributes necessary for image creation step by step.

[0086] For example, referring to FIG. 4, the visual queue selection area (10) of the user interface may include a component selection area (11), a subcategory selection area (12), a text input area (13), and a visual queue display area (14). However, the present disclosure is not limited thereto, and the visual queue selection area (10) may include other areas other than those described above, or may include fewer areas than those described above.

[0087] For example, the user can select a desired item from a plurality of component tabs, such as 'Target', 'Style', 'Background', and 'Lighting', in the component selection area (11). When the user selects a specific component tab, the computing device (100) can load and present a plurality of visual cues belonging to that component in the visual cue display area (14). At this time, each visual cue may be displayed in the visual cue display area (14) arranged as a combination of an image and its corresponding text, and the user can intuitively select the desired cue through mouse clicks, touch input, or keyboard selection. For example, when the user clicks 'Target' in the component selection area (11), various visual cues related to targets, such as people, animals, objects, and characters, are displayed in the interface, and when the user selects one of them, the corresponding text is automatically reflected in the prompt panel.

[0088] Additionally, the user can filter the visual cues displayed in the visual cue display area (14) through the sub-category selection area (12). For example, if the 'Person' sub-category is selected while the 'Target' component is selected, the computing device (100) can refer to a pre-classified database to sort and display only the visual cues related to people in the visual cue display area (14). At this time, the sub-category selection area can be provided in various ways, such as a drop-down menu, a checkbox, or a toggle button, and the user can adjust the range of the displayed visual cues by specifying multiple sub-categories simultaneously or by disabling specific conditions.

[0089] As a result of the filtering, various sub-visual cues related to people (e.g., children, adolescents, adults, the elderly, males, females, etc.) are arranged in the visual cue display area in the form of a combination of images and text, allowing the user to intuitively select the desired visual cue from among these items. In this way, the sub-category selection area supports direct access to the attributes required by the user among multiple visual cues, enabling the prompt creation process to be performed more quickly and efficiently.

[0090] Subsequently, the user can provide a selection input by clicking or touching a specific visual cue displayed in the visual cue display area (14), for example, the ‘teacher’ visual cue (14’). This selection input is not limited to a single click or touch, but can be implemented in a toggle method where the same item is deselected when selected again, or in a method where multiple visual cues are selected consecutively and accumulated, and the result of the user's input is reflected in the prompt panel area in real time. For example, when the ‘teacher’ visual cue is selected, the text “teacher” is automatically added to the ‘target’ item of the prompt panel, and then when the ‘jumping action’ visual cue is selected, the corresponding text can be accumulated and recorded in the same item.

[0091] Alternatively, the user may directly input specific attributes not provided by the system using the text input area (13). For example, if the user inputs the phrase "smiling doctor" into the text input window and selects the "Add" button, the computing device (100) recognizes this as a user-defined prompt element for the "target" component and reflects it in the prompt panel. The text entered at this time can be in units of words, phrases, or sentences, and complex attributes can be constructed through multiple inputs. Additionally, the item may be deleted or modified as needed, and the entered text may be stored internally by the system and subsequently used as training data for recommended visual cues.

[0092] In various embodiments, the computing device (100) may acquire a mouseover input for a specific visual queue among one or more component-specific visual queues. The mouseover input may include the act of detecting that a pointer of the user terminal (200), such as a mouse cursor, a touchpad pointer, or the pointing coordinates of a stylus pen, is positioned over a specific visual queue image for a certain period of time or longer. The system may set a predetermined delay time to distinguish such input from simple screen movement, and may be implemented to recognize it as a valid mouseover input only when the cursor remains continuously on the image for a certain period of time.

[0093] When mouse over input is recognized, the computing device (100) can convert and display an image included in the corresponding visual queue to a larger size or improved resolution than the original. The enlarged image may appear as a pop-up over the existing visual queue display area or be provided through a separate overlay window, and may automatically disappear when the user moves the cursor out of the image, so that the visual queue of the original size is displayed again. This pop-up or overlay method allows the user to check the detailed visual characteristics of a specific queue even when multiple visual queues are listed simultaneously in a limited screen area.

[0094] As a specific example, when a user places a mouse cursor over an 'astronaut' visual cue image displayed in the visual cue display area (14), the computing device (100) can generate and display an enlarged image of the astronaut image with increased resolution or size. The enlarged image clearly shows visual features such as detailed texture, tone, and composition of the original, allowing the user to intuitively review whether the image matches the properties intended by the user.

[0095] Furthermore, the magnified display may provide auxiliary information in addition to simple image magnification. For instance, text labels mapped to the visual cue (e.g., "astronaut"), related attribute values ​​(e.g., "target"), or associated tags recommended by the system (e.g., "space background," "neon lights") may be displayed. This auxiliary information helps users quickly understand the meaning of the image and compare it with other visual cues. Moreover, the system can record the frequency or duration of mouseovers on specific visual cues to utilize as training data for recommendation algorithms, thereby enabling the development of personalized user interfaces.

[0096] The mouse-over zoom feature can be particularly useful when selecting attributes where even small differences, such as composition, style, and texture, significantly impact the overall mood or interpretation of an image. Since users can accurately perceive and compare minute details through the zoomed-in image, it can enhance the precision of the final prompt selection. Therefore, beyond simple user convenience, this feature can serve as an important means to simultaneously improve the efficiency of prompt creation and the quality of the resulting image.

[0097] In one embodiment, the computing device (100) can display text corresponding to user input in the prompt panel area (20) in conjunction with acquiring user input (S120). This means that as the user selects a visual queue or directly inputs text, the prompt panel area is updated in real time so that the input is not merely reflected temporarily, but can be systematically accumulated and managed by component.

[0098] Specifically, the computing device (100) can map a specific visual cue included in a specific component or input specific text in the previous step (S110) to the corresponding component and add it to the list (21) item in the prompt panel area. For example, if the user selects a visual cue named 'teacher' from the 'Target' component tab, the text "teacher" can be newly reflected next to the 'Target' item in the list in the prompt panel area. Then, if the user selects 'transparent background' from the 'Background' component, the corresponding text can be accumulated next to the 'Background' item in the list. In this way, as the user sequentially selects visual cues across multiple components, the prompt panel can gradually accumulate text corresponding to each component item, and finally form a completed prompt string.

[0099] More specifically, whenever the computing device (100) obtains a selected input of a visual cue or directly entered text from a user, it may prioritize identifying the component to which the input belongs. If the identified component is, for example, a 'target', the system may activate the area of ​​the 'target' item in the list (21) and add the text of the selected visual cue or directly entered text in real time next to the item. The method of addition may be implemented as an additive method rather than a simple overwrite method, so that when multiple items are selected for the same component, multiple texts are displayed together using a comma, slash, or other separator. Such cumulative updates can help the user gradually expand the attributes required for image generation while still allowing them to grasp the entire structure at a glance.

[0100] Furthermore, updates in the prompt panel area can be configured to provide visually intuitive feedback, going beyond simply listing text. For example, newly selected items are indicated by short animations or changes in highlight color, allowing users to immediately recognize that their recent selection has been successfully reflected in the panel. Conversely, when a user deselects or deletes a specific item, the corresponding text can be implemented by gradually fading away from the list before being completely removed. This visual feedback can help users understand the prompt creation process more intuitively.

[0101] The prompt panel area may include a plurality of component lists (21), component-specific attribute buttons (22), text copy buttons (23), selection initialization buttons (24), etc., as shown in (A) and (B) of FIG. 5.

[0102] The list (21) lists items corresponding to each component, and next to each item, the visual cue text selected by the user can be displayed in a stacked manner. The attribute buttons (22) can be implemented with the names of each component (e.g., 'target', 'background', 'material / texture'), and when the user clicks the button, only the text corresponding to that attribute can be selectively copied. For example, if the user clicks the 'material / texture' button, only the prompt text related to all textures selected so far is saved to the clipboard.

[0103] Meanwhile, the text copy button (23) can provide the function of copying the entire prompt string accumulated in the prompt panel area to the clipboard. When the user clicks this button, the computing device (100) integrates all component items displayed in the list (21) into one complete prompt string and stores it in the clipboard of the user terminal. Thus, the user can obtain the resulting image immediately by simply pasting the copied prompt into an external image generation AI model (ChatGPT, Gemini, Grok, Copilot, etc.).

[0104] The selection reset button (24) can perform the function of returning the prompt panel to its initial state. When the user clicks this button, all text displayed in the list (21) can be deleted and each component item can be initialized to an empty state. This function is useful when the user wants to start writing the prompt again from the beginning and can reduce unnecessary modification or deletion processes.

[0105] To explain in detail with an example, when a user selects the ‘teacher’ visual cue belonging to the ‘target’ component in the visual cue selection area (10), the computing device (100) can automatically display the text “teacher” next to the ‘target’ item in the list (21) of the prompt panel area (20). Then, when the user selects the ‘transparent background’ visual cue in the ‘background’ component interface, the corresponding text is added and displayed next to the ‘background’ item in the list (21). Subsequently, when ‘front view’ is selected in the ‘composition’ tab and ‘photorealistic’ in the ‘style’ tab, the list (21) gradually expands to complete the final list structure as illustrated in (B) of FIG. 5. Through this series of processes, the user can visually track the process of constructing the entire prompt, which can provide a much more intuitive and systematic prompt creation experience than simple sentence-based input.

[0106] Additionally, the prompt panel area may support a function to modify selected items to enhance user convenience. For example, an 'X' mark button may be placed next to a specific component item in the list (21) to allow the user to easily remove individual items, or the order of items may be rearranged through a drag-and-drop function. Such an editable structure allows the user to quickly and flexibly correct the prompts even after multiple trials and errors.

[0107] Furthermore, the computing device (100) may provide real-time feedback based on data accumulated in the prompt panel. For example, if a user selects 'watercolor' from the 'style' component and 'neon light' from 'lighting', the system may detect the possibility that the two attributes are an incompatible combination and present a warning message or alternative recommendation to the user. This feedback function supports the creation of more complete prompts beyond simple input recording.

[0108] Furthermore, data accumulated in the prompt panel can be utilized as training data. The system records which visual cues a user frequently selects and which combinations they prefer, and subsequently adjusts the priority of recommended visual cues based on the user's preference patterns. For instance, if a user frequently selects 'Portrait' and 'Photorealistic' styles, those items can be automatically sorted and displayed at the top in the next session. This can provide a foundation for evolving into a user-customized interface.

[0109] In some embodiments of the present disclosure, when the computing device (100) obtains a selection input for a copy button included in a prompt panel area, it may store a prompt corresponding to the visual queue selected so far in the clipboard of the user terminal (S130). This allows the prompt, which the user has progressively written across multiple components, to be integrated into a single completed string and easily transmitted to an external system.

[0110] Specifically, the computing device (100) can combine the entire text contained in the list of all components (21) displayed in the prompt panel area (20) into a single prompt string. In this combination process, the text is not merely concatenated, but a string can be generated in a structurally distinct form by attaching the item name of each component as a prefix. For example, by integrating in a format such as "Target: teacher / Background: transparent background," the image generation AI model can interpret the attributes more clearly.

[0111] When the user selects the text copy button (23) after completing the prompt, the computing device (100) can convert all items included in the list into a single string according to a set format and save it to the clipboard. At this time, the conversion format may be a pre-set standard template or a format customized by the user. Since the converted string is in a simple text format, it can be input into various image generation AI models directly through the paste function on the user terminal (200). This allows the user to immediately utilize the prompt without a separate editing process.

[0112] For example, when a user sequentially selects visual cues for components such as 'target', 'style', 'composition', and 'lighting' and then clicks the text copy button (23), the computing device (100) may display a confirmation window titled "Prompt Copy Complete" as illustrated in (A) of FIG. 6. This confirmation window clearly informs the user that the copy operation has been successfully performed and may also provide a preview of some of the copied content. Through this, the user can review the copied prompt again and modify it if necessary.

[0113] Meanwhile, the following prompt may be stored in the clipboard of the user terminal (200).

[0114] Create an image based on the list prompt below :

[0115] Target: Teacher

[0116] Material / Texture: Transparent glass

[0117] Background: Transparent background

[0118] Composition: Front view

[0119] Style: Photorealistic

[0120] Color / Mood: Vivid tones

[0121] Lighting: Diffused light

[0122] Such structured prompts help external AI models understand the user's intent more accurately, and allow users to obtain the desired image without repetitive trial and error.

[0123] In various embodiments, when the computing device (100) obtains a selection input for an attribute button, it may save only the prompt of a specific component corresponding to the selected attribute button to the clipboard of the user terminal. This is a function to allow the user to edit or test only specific attributes separately without copying the entire prompt.

[0124] Specifically, the computing device (100) may provide each of the multiple component names displayed in the list (21) of the prompt panel as selectable attribute buttons (22). When a user selects a specific attribute button, the computing device (100) may save the selected text to the clipboard only for that component, excluding the content of other components. This feature may be useful when the user wishes to modify or test only some elements of the entire prompt.

[0125] For example, as illustrated in (B) of FIG. 6, when a user clicks the 'material / texture' item in the list (21), the computing device (100) may display a confirmation window titled "'material / texture' copy complete". This confirmation window may clearly inform the user that the text contained in the component has been successfully saved to the clipboard.

[0126] This attribute-level copying feature can significantly enhance the flexibility of prompt creation. Users can easily compare various results by changing only specific parts without modifying the entire prompt. In particular, when changes in detailed attributes significantly impact the overall outcome during the image generation process, this feature enables efficient experimentation and iteration.

[0127] In addition, the computing device (100) of the present disclosure may provide various additional functions in conjunction with a copy function. For example, the content copied to the clipboard may be maintained for a certain period of time, or recent copy history may be saved as a history so that the user can recall past prompts at any time. Furthermore, it is possible to implement a function that automatically backs up the prompts stored in the clipboard to a cloud-based storage by linking with an external server, or shares them with other users. These additional functions can greatly expand the usability of the prompts.

[0128] According to various embodiments of the present disclosure, a computing device (100) can support the easy configuration of a more complete prompt combination by recommending associated prompt elements based on a user's prompt selection.

[0129] Referring to FIG. 7, when the computing device (100) obtains a first selection input for a first component, it can determine a recommended visual queue for a second component associated with a first visual queue corresponding to the first selection input (S210). This is a function in which the system actively analyzes and presents how a specific attribute individually selected by the user can be combined with other attributes to provide the most natural and complete result.

[0130] Specifically, the computing device (100) analyzes text or image features of a first visual cue corresponding to a user's first selection input and, based on a pre-built association database or machine learning model, determines other visual cues that are highly associated with the analyzed features as recommended visual cues of a second component. The association database may be built based on the frequency of use of specific prompt elements together or semantic similarity by pre-analyzing multiple images and prompt data, and the user's preference patterns or learned results may be continuously updated.

[0131] More specifically, when the computing device (100) analyzes the text of the first visual queue, it can perform a statistical analysis to measure the frequency of co-occurrence between specific words or phrases. For example, it can calculate the frequency with which the text "neon light" appears together with the text "cyberpunk" in prompt data containing the text "cyberpunk", and calculate an association score based on this. Items with high association scores are positioned at the top as the recommendation visual queue of the second component.

[0132] In various embodiments, the computing device (100) may go beyond simple frequency analysis and use a natural language processing (NLP) model to convert the text of each visual cue into a vector, and use a mathematical measure such as cosine similarity within the vector space to determine the visual cue of another component that is semantically closest as a recommendation target. For example, even if the words "space" and "galaxy" do not frequently appear together in the same context, they may yield high semantic similarity in the vector space, in which case the system may recommend "galaxy" as a background element when the user selects "space."

[0133] Additionally, when the computing device (100) analyzes an image of the first visual cue, it can extract various visual features such as the color distribution of the image, object detection results, and texture patterns, and determine a visual cue containing another image with similar features as a recommendation target. To do this, a deep neural network model such as a Convolutional Neural Network (CNN) can be applied, or a pre-trained Vision Transformer (ViT) model can be utilized. For example, if a user selects a "cyberpunk" style image, the system can detect that the main color palette of the image is neon-colored and suggest "neon lighting" as a recommendation target in the "lighting" component and "android" as a recommendation target in the "target" component.

[0134] For example, if the user selects "cyberpunk" in the 'style' component, the computing device (100) may refer to an association database to determine "neon light" in the 'lighting' component, "android" in the 'target' component, and "futuristic city" in the 'background' component as recommended visual cues. These recommended items may be displayed at the top when the user next opens the corresponding component tab.

[0135] Next, when the computing device (100) provides an interface corresponding to the second component, it can arrange and display a plurality of visual queues corresponding to the second component based on the recommended visual queue (S220). This is implemented not merely by highlighting the recommended item, but by arranging the entire list of visual queues so that the recommended item is positioned at the top first, and the remaining general items are displayed sequentially below it.

[0136] Specifically, when a user selects the 'Lighting' component tab, the computing device (100) can retrieve the entire list of visual cues included in the interface and rearrange them so that the recommended item generated in step (S210) is displayed first. For example, if 'Neon Lighting' is recommended, 'Neon Lighting' is displayed at the top of the list, followed by 'Natural Light', 'Studio Lighting', 'Soft Light', etc., in sequence. Through this, the user can prioritize checking the item deemed most suitable when searching for the desired lighting.

[0137] In various embodiments, recommended items can be distinguished and highlighted using visual effects in addition to simply aligning them at the top. For example, recommended items may be displayed with a highlight background color or have a separate icon attached, or a tooltip reading "Recommended Item" may be provided when the mouse hovers over them. This allows users to intuitively recognize that the item is not merely one of the items in a list, but a recommendation derived from the system's analysis results.

[0138] Furthermore, the system can record behavioral data, such as which recommended items the user actually selected or whether they ignored recommendations and chose other items, and incorporate this into machine learning models. Over time, the accuracy of recommendations gradually improves; for example, even if the same "Cyberpunk" style is selected, if a specific user shows a tendency to prefer "blue light" over "neon lighting," "blue light" may be displayed with a higher priority in subsequent sessions.

[0139] Accordingly, the computing device (100) of the present disclosure can support the system in easily and quickly configuring a harmonious and complete combination of prompts through a recommendation function, even if the user does not need to be familiar with the complex relationships between the prompt elements. This can lower the entry barrier for novice users and provide the effect of increasing work efficiency for skilled users.

[0140] According to various embodiments of the present disclosure, the computing device (100) may also provide a feedback structure that utilizes the user's image generation results as input assets of the system. For example, if a user generates an image by selecting "cyberpunk" style and "neon lighting" and then uploads the result back to the system, the system may map the resulting image to the corresponding prompt and register it as a new visual cue. The data accumulated in this way may be provided as recommendation items to other users in the future and may be utilized as an asset to improve the learning quality of the entire system.

[0141] Referring to FIG. 8, the computing device (100) can obtain an output image corresponding to the prompt after storing the prompt corresponding to the currently selected visual queue in the clipboard of the user terminal (S310).

[0142] Specifically, the computing device (100) may provide an interface for receiving an output image uploaded from a user terminal (200). The interface may include a file upload button or a drag-and-drop area. When an image is uploaded, the computing device (100) may acquire and store the image as an output image associated with the prompt most recently stored in the clipboard of the user terminal (200). Through this, a data pair between a specific prompt and the resulting output image can be obtained.

[0143] For example, it is assumed that the computing device (100) inputs a prompt generated by the user through the system of the present disclosure into an external AI model to generate an image of a 'cat flying in space' and saves the image file to its user terminal (200). Subsequently, the user can transmit the saved 'cat flying in space' image file to the computing device (100) through the 'result image upload' button provided by the system of the present disclosure. The computing device (100) receives the transmitted image and obtains it as an output image for the 'cat flying in space' prompt.

[0144] Additionally, the computing device (100) can generate a new visual queue based on the output image (S320).

[0145] Specifically, the computing device (100) can generate a new visual queue by combining the output image obtained in step (S310) with the prompt text associated with the output image. In this case, the computing device (100) may use the original output image as is, or generate a thumbnail image by resizing, cropping, or compressing it to fit the interface specifications of the visual queue selection area, and use this as the image part of the new visual queue. The text part of the new visual queue may be assigned the entire prompt string used to generate the output image as is.

[0146] For example, if the computing device (100) obtains an output image of a ‘cat flying in space’ in step (S310), it can generate a thumbnail of the image. Then, it can create a new visual queue by combining this thumbnail image and the prompt used to generate the image (e.g., ‘target: cat, flying action, background: space, style: oil painting’) into a single data pair.

[0147] Additionally, the computing device (100) can determine at least one component among the plurality of components included in the prompt to be mapped to a new visual queue (S330).

[0148] Specifically, the computing device (100) can analyze the prompt text included in the new visual queue to identify multiple components (e.g., 'target', 'style') to which the text is assigned. Subsequently, the computing device (100) can obtain user input to provide a list of the identified multiple components to the user and to select one or more target components to which the new visual queue will be added.

[0149] For example, if the computing device (100) contains text for the 'target', 'background', and 'style' components in the prompt of the 'Cat Flying in Space' visual cue generated in step (S320), it may provide the user with options for 'target', 'background', and 'style' in the form of a checkbox interface. The user may complete the input by selecting only the 'target' component, which is considered to be the most representative attribute of the corresponding visual cue.

[0150] In various embodiments, the computing device (100) may automatically determine the key main components through parsing the prompt text or analyzing the features of the output image.

[0151] Specifically, the computing device (100) can use natural language processing (NLP) technology to analyze the key keywords or sentence structure of the prompt to infer the main components, or use a pre-trained image recognition model (e.g., CNN) to recognize the main objects, backgrounds, and styles of the output image and map them to the most suitable components.

[0152] For example, the computing device (100) inputs the output image of a ‘cat flying in space’ into an image recognition model, and if the model recognizes the ‘cat’ as the primary object with the highest confidence, it can determine the ‘object’ component to which the ‘cat’ belongs as the target component of a new visual queue.

[0153] And, the computing device (100) can map a new visual queue to at least one component and add the new visual queue to the interface of each of the at least one component (S340).

[0154] Specifically, the computing device (100) can store a new visual queue (a data pair of a thumbnail image and a prompt text) in a database corresponding to the target component determined in step (S330). Subsequently, when a user requests an interface for the target component, the computing device (100) can load the new visual queue stored in the database along with the existing visual queues and display them together in the visual queue display area (14).

[0155] For example, if the computing device (100) decides to map the ‘Flying Cat’ visual cue to a ‘Target’ component, it can add the visual cue to the list of ‘Target’ components. Then, when the user selects the ‘Target’ tab, the ‘Flying Cat’ visual cue is newly displayed in the visual cue display area (14) along with the existing visual cues such as ‘Teacher’ and ‘Policeman’, allowing the user to select it.

[0156] Accordingly, the computing device (100) of the present disclosure can provide a personalized prompt generation environment optimized for the user through a virtuous cycle structure in which the user's creative activity continuously expands the system's assets and the system again promotes the user's creativity.

[0158] Although embodiments of the present disclosure have been described above with reference to the attached drawings, those skilled in the art will understand that the present disclosure may be implemented in other specific forms without altering its technical concept or essential features. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive.

Claims

Claim 1 A method performed by a computing device comprising at least one processor, comprising: providing a visual queue selection area comprising one or more component-specific visual queues and obtaining user input for creating a prompt for image generation; displaying text corresponding to the user input in a prompt panel area in conjunction with obtaining the user input; when a selection input for a copy button included in the prompt panel area is obtained, storing a prompt corresponding to the currently selected visual queue in the clipboard of a user terminal; after the prompt is stored in the clipboard of the user terminal, obtaining an output image corresponding to the prompt from the user terminal and storing the output image as the output image associated with the prompt most recently stored in the clipboard of the user terminal; generating a thumbnail image by resizing, cropping, or compressing the output image to conform to the interface specifications of the visual queue selection area, and generating a new visual queue by grouping the thumbnail image and the text of the prompt used to generate the output image into a single data pair; and determining at least one component among a plurality of components included in the prompt to be mapped to the new visual queue. A method for providing a prompt for generating a non-visual queue-based image, comprising the step of mapping the new visual queue to the at least one component, storing the new visual queue in a database corresponding to the at least one component, and, when an interface of each of the at least one component is requested, loading the new visual queue along with existing visual queues from the database and displaying it on the interface of each of the at least one component. Claim 2 A method for providing a prompt for generating a visual cue-based image according to claim 1, wherein the one or more components include at least one of an object, material and texture, background, composition, style, color and mood and lighting corresponding to the image to be generated, the visual cue selection area is composed of one or more interfaces corresponding to the one or more components, and the visual cue includes a plurality of images corresponding to the currently displayed interface and text corresponding to each of the plurality of images. Claim 3 A method for providing a non-visual queue-based image generation prompt, wherein, in claim 1, the step of providing a visual queue selection area including one or more component-specific visual queues and obtaining user input for creating an image generation prompt comprises: obtaining a selection input for a specific visual queue; or obtaining a specific text input while a specific component is displayed. Claim 4 A method for providing a prompt for generating a visual queue-based image according to claim 1, wherein the prompt panel area includes a list corresponding to one or more components, and when a specific visual queue included in a specific component is selected or specific text is entered, the item of the list corresponding to the specific component is updated to include the specific visual queue or the specific text. Claim 5 A method for providing a prompt for visual queue-based image generation, wherein, in claim 1, the prompt panel area further includes one or more component-specific attribute buttons, and the method further includes the step of, when a selection input for the attribute button is obtained, storing a prompt corresponding to the selected attribute button in the clipboard of a user terminal. Claim 6 A method for providing a prompt for generating a non-visual queue-based image, wherein the method further comprises: a step of obtaining a mouseover input for a specific visual queue among the one or more component-specific visual queues; and a step of enlarging and displaying an image included in the specific visual queue in response to the mouseover input. Claim 7 A method for providing a visual queue-based image generation prompt according to claim 1, wherein the step of providing a visual queue selection area including one or more component-specific visual queues and obtaining user input for creating an image generation prompt further comprises: a step of determining a recommended visual queue of a second component associated with a first visual queue corresponding to the first selection input when a first selection input for a first component is obtained; and a step of arranging and displaying a plurality of visual queues corresponding to the second component based on the recommended visual queue when providing an interface corresponding to the second component. Claim 8 delete Claim 9 A device comprising: a memory for storing one or more instructions; and a processor for executing the one or more instructions stored in the memory, wherein the processor performs the method of claim 1 by executing the one or more instructions. Claim 10 A computer program stored on a computer-readable recording medium that is combined with a computer, which is hardware, to perform the method of claim 1.

Images