Program, image generation support apparatus, image generation support system, and image generation support method

Abstract

To allow a user to easily obtain a generated image as intended regardless of design knowledge or the scope of ideas.SOLUTION: A program causes a computer to execute: a text generation step (Step S3) of causing AI to generate second text data to be provided to image generation AI, from first data; and an image generation step (Step S4) of causing the image generation AI to generate a generated image from the second text data.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to a program, an image generation support device, an image generation support system, and an image generation support method. [Background technology]

[0002] Traditionally, when salespeople at printing companies discuss images of printed materials with customers, the customers often do not have a clear image of what they want to achieve. Therefore, salespeople have to draw out an image from "nowhere." Therefore, the salesperson creates images several times and has the customer check them, which allows the salesperson to determine the direction of the design, and then passes the images to the designer to complete the official images.

[0003] On the other hand, as described in Patent Document 1, for example, machine learning is used to generate images based on input text data. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Publication No. 2024-033903 Summary of the Invention [Problem to be solved by the invention]

[0005] Conventional image generation instructions via text input require the input of words that describe the desired motif or description of the image, and this depends on the user's design knowledge and range of ideas, making it difficult for users without specialized design knowledge, such as sales staff at printing companies, to generate images with simple operations and obtain the desired image.

[0006] Therefore, an object of the present invention is to enable a user to easily obtain an intended generated image regardless of the range of design knowledge or ideas. [Means for solving the problem]

[0007] In order to solve the above problem, the program according to the present invention comprises: To your computer a text generation step of generating second text data to be given to the image generation AI from the first data by the AI; an image generation step of generating an image from the second text data by an image generation AI; Execute the following.

[0008] Furthermore, the image generation support device according to the present invention comprises: a text generation unit that generates second text data to be provided to the image generation AI from the first data by an AI; an image generation unit that generates a generated image from the second text data using an image generation AI; Equipped with.

[0009] Furthermore, the image generation support system according to the present invention comprises: a text generation unit that generates second text data to be provided to the image generation AI from the first data by an AI; an image generation unit that generates a generated image from the second text data using an image generation AI; Equipped with.

[0010] Further, the image generation support method according to the present invention comprises: An image generation support method performed by an image generation support device, a text generation step of generating second text data to be given to the image generation AI from the first data by the AI; an image generation step of generating an image from the second text data by an image generation AI; Includes. [Effects of the Invention]

[0011] According to the present invention, a user can easily obtain an intended generated image regardless of the user's knowledge of design or range of ideas. [Brief explanation of the drawings]

[0012] [Figure 1] FIG. 1 is a block diagram illustrating a configuration of an image generation support system. [Figure 2] FIG. 10 is a flow diagram of image generation support processing. [Figure 3] FIG. 10 is an image diagram of an image generation support screen. [Figure 4] FIG. 10 is an image diagram of an image generation support screen. [Figure 5] FIG. 10 is a flow diagram of image generation support processing. [Figure 6] FIG. 10 is an image diagram of an image generation support screen. [Figure 7] FIG. 10 is an image diagram of an image generation support screen. [Figure 8] FIG. 10 is an image diagram of an image generation support screen. [Figure 9] FIG. 10 is a flow diagram of image generation support processing. [Figure 10] FIG. 10 is an image diagram of an image generation support screen. [Figure 11] FIG. 10 is an image diagram of an image generation support screen. [Figure 12] FIG. 10 is an image diagram of an image generation support screen. [Figure 13] FIG. 10 is a flow diagram of image generation support processing. [Figure 14] FIG. 10 is an image diagram of an image generation support screen. [Figure 15] FIG. 10 is an image diagram of an image generation support screen. [Figure 16] FIG. 10 is an image diagram of an image generation support screen. [Figure 17] FIG. 10 is an image diagram of an image generation support screen. [Figure 18] FIG. 10 is an image diagram of an image generation support screen. [Figure 19] FIG. 10 is an image diagram of a history screen. DETAILED DESCRIPTION OF THE INVENTION

[0013] Hereinafter, embodiments of the present invention will be described with reference to the drawings. However, the scope of the present invention is not limited to the following embodiments and drawings.

[0014] <Image generation support system 100> First, the configuration of an image generation support system 100 will be described using Fig. 1. As shown in Fig. 1, the image generation support system 100 includes an image generation support device 1, a cloud server 2, and an external device 3. The image generation support device 1, the cloud server 2, and the external device 3 transmit and receive information via a communication network N.

[0015] Image generation support device 1 is a device that supports image generation by users such as salespeople at printing companies and their customers. Specifically, image generation support device 1 displays a generated image generated by cloud server 2 based on text data (first text data) and / or image data input by the user, and accepts various operations on the generated image by the user.

[0016] Cloud server 2 is an AI device that uses a sentence generation AI to generate second text data from first text data and / or image data, and uses an image generation AI to generate generated images from the second text data. The second text data is a so-called prompt. Furthermore, the cloud server 2 is an AI device that uses an image generation AI to extract design elements, which will be described later, from the generated image. The cloud server 2 also serves as a device for analyzing the saliency of the generated image, which will be described later. The cloud server 2 may be an on-premise device.

[0017] The external device 3 is a device on which the designer displays the generated images, concepts, and the like.

[0018] The communication network N is a LAN (Local Area Network), a WAN (Wide Area Network), the Internet, or the like.

[0019] <Image generation support device 1> Next, the configuration of the image generation support device 1 will be described with reference to Fig. 1. As shown in Fig. 1, the image generation support device 1 includes a control unit 11, a display unit 12, an operation unit 13, a communication unit 14, and a storage unit 15.

[0020] The control unit 11 is configured with a CPU (Central Processing Unit), RAM (Random Access Memory), etc. The CPU of the control unit 11 reads out various programs stored in the storage unit 15, expands them in the RAM, executes various processes in accordance with the expanded programs, and controls the operation of each unit of the image generation support device 1. The control unit 11 functions as a first acquisition unit that acquires text data generated based on an image. The control unit 11 functions as a first display control unit that displays the acquired text data on the display unit. The control unit 11 functions as an editing unit that edits the text data after displaying the text data. The control unit 11 functions as a second acquisition unit that acquires a generated image generated based on edited text data. The control unit 11 functions as a second display control unit that displays the acquired generated image on the display unit.

[0021] The display unit 12 is configured with a monitor such as an LCD (Liquid Crystal Display), and displays various screens and the like according to instructions of a display signal input from the control unit 11.

[0022] The operation unit 13 is a keyboard equipped with cursor keys, numeric input keys, various function keys, etc., a pointing device such as a mouse, a touch panel laminated on the surface of the display unit 12, etc. The operation unit 13 is configured to be operable by an operator. The operation unit 13 also outputs various signals to the control unit 11 based on operations performed by the operator.

[0023] The communication unit 14 is capable of transmitting and receiving various signals and various data to and from other devices connected via the communication network N.

[0024] The storage unit 15 is configured by a non-volatile semiconductor memory, a hard disk, or the like, and stores various programs executed by the control unit 11, parameters required for executing the programs, various data, and the like.

[0025] <Cloud Server 2> Next, the configuration of the cloud server 2 will be described with reference to Fig. 1. As shown in Fig. 1, the cloud server 2 includes a control unit 21, a communication unit 22, and a storage unit . Cloud Server 2 has the functions of an image generation AI and a saliency analysis module.

[0026] The control unit 21 is configured with a CPU, RAM, etc. The CPU of the control unit 21 reads out various programs stored in the storage unit 23, expands them in the RAM, executes various processes in accordance with the expanded programs, and controls the operation of each unit of the cloud server 2. The control unit 21 generates second text data from the first text data and / or image data using a sentence generation AI, and generates an image from the second text data using an image generation AI. The first text data and / or image data are referred to as first data. The control unit 21 functions as a first generation unit that generates text data based on an image. Specifically, the control unit 21 uses an image generation AI to extract design elements, which will be described later, from the image. The control unit 21 functions as a reception unit that receives a designation by the user using a designation means for a region of interest in the first image. The control unit 21 functions as a generating unit that generates and outputs text data corresponding to the first image according to the region of interest received in the receiving step. The control unit 21 functions as a display control unit that displays, on the display unit, a second image that indicates the features read from the first image when generating the text data. The control unit 21 functions as a setting unit that sets the priority of the design requirements. The control unit 21 functions as a text generation unit that causes the AI ​​to generate second text data to be given to the image generation AI from the first data. The control unit 21 functions as an image generation unit that causes an image generation AI to generate a generated image from the second text data. The control unit 21 functions as a display control unit that causes the display unit to display only the history of related information linked to the same keyword so that it can be referenced. The control unit 21 functions as an association unit that associates image information, including the generated image and information used when generating the image, with a restriction condition that restricts sharing of the image information. The control unit 21 functions as a restricting unit that restricts the generation or display of image information based on restrictive conditions. The control unit 21 functions as an acquisition unit that acquires, based on the first client information, second client information that is in a competitive relationship with the client of the first client information. Examples of image generation AI include Stable Diffusion, DALL·E2, Midjourney, starryai, and Dream by WOMBO. Examples of sentence generation AI include ChatGPT, Bard, Gemini, and Bloom. The control unit 21 can analyze the saliency of the generated design image. That is, the control unit 21 can display the analysis result regarding saliency in the following cases. Specifically, this is the case when an evaluator (user, etc.) wants to know whether a part of an image that the evaluator (user, etc.) wants to make particularly visually noticeable has been made to stand out. When a part that the evaluator wants to make particularly visually noticeable has been made to stand out, it is said to be saliency. This is also the case when an evaluator wants to know what to do to make that part of the image to be evaluated more noticeable.

[0027] (Saliency analysis method) Here, the saliency analysis method will be described. First, the functions of the control unit 21 will be described in detail. The control unit 21 functions as a feature extraction unit and a generation unit.

[0028] The control unit 21 as a feature amount extraction unit extracts low-order image feature amounts and high-order image feature amounts from the acquired evaluation target image. A specific method by which the control unit 21 serving as a feature extracting unit extracts low-order image feature amounts and high-order image feature amounts from the evaluation target image will be described later.

[0029] Here, the low-order image feature is a physical image feature including, for example, color, brightness, direction (edge ​​orientation, shape), etc., and is a component that externally or passively guides a person's gaze to gaze. In this embodiment, the low-order image feature is a concept that broadly includes at least one of color, brightness distribution contrast, etc., and movement. The impact an image has on the viewer and the degree to which it draws attention (prominence, salience) varies depending on factors such as the color contrast used in each part of the image (for example, color differences in the red-green direction and yellow-blue direction), the distribution of brightness contrast (luminance difference) between parts, and directional (orientation) contrast.

[0030] For example, a portion (such as a boundary portion) with a large color difference along the red-green direction or the yellow-blue direction tends to attract the eye and become highly salient. Furthermore, for example, if the entire image is arranged in a fixed direction and there is an object arranged in a different direction (edge ​​direction), the line of sight tends to be drawn to that part.

[0031] Furthermore, the image to be evaluated is not limited to a still image, but may be a moving image. When the image to be evaluated is a moving image, various actions (movements, movements) within the image also affect the degree of gaze of the viewer. For example, if an object moves at a different speed or in a different direction in an image in which the entire image is moving at a substantially constant speed and in a constant direction, the viewer's gaze tends to be drawn to that part.

[0032] Higher-order image features are physiological and mental image features that reflect people's memories, experiences, knowledge, etc., and are components that intrinsically or actively guide people's gaze to focus. More specifically, they are components derived from people's mental or psychological tendencies, gaze movement tendencies, etc. that are thought to affect the impact an image has on a viewer and the degree of gaze (prominence, salience). In this embodiment, the higher-order image features include the degree of at least one of position bias, processing fluency, and face components.

[0033] For example, positional bias includes the following tendencies in eye movement. Specifically, there is a center bias, where the gaze tends to be drawn to objects in the center of an image. Also, for example, when viewing magazines or web pages, the gaze tends to move from the upper left to the lower right of the image, with the gaze tending to be drawn to the upper left. Also, when viewing a document written vertically, the gaze tends to move from the upper right to the lower left, with the gaze tending to be drawn to the upper right. Also, when considering a store such as a supermarket, for example, the gaze tends to be drawn to parts of the store layout that are close to eye level. As described above, the position bias affects the degree of gaze (prominence, salience) of a viewer of an image or the like.

[0034] Furthermore, processing fluency indicates that people generally find it easy to process simple things that are easy to recognize, and difficult to process complex things that are difficult to understand. In this embodiment, people tend to look at and focus on parts of an image that are easy to recognize and have high processing fluency, while they tend not to focus on parts that are difficult to recognize and have low processing fluency. As mentioned above, processing fluency affects the degree of attention (prominence, salience) of the viewer to an image.

[0035] In this embodiment, the degree of processing fluency includes those determined by at least one of the complexity, the density of the depicted object, and the spatial frequency of the luminance distribution. In other words, parts that are difficult to recognize are cluttered and complex parts, that is, parts where the objects depicted in the image are densely packed and difficult to understand. In parts of the image where objects are densely packed, there are sudden changes in the image, such as edges, and the spatial frequency of the luminance distribution is high in such parts. Processing fluency decreases in parts where the complexity, density of depicted objects, and spatial frequency of the luminance distribution are too high. On the other hand, areas where the complexity, density of depicted objects, and spatial frequency of the luminance distribution are too low, in other words, areas that do not contain information, are difficult to read, are hard for the human brain to process, and tend not to be focused on.

[0036] Furthermore, when there is a part in an image that is recognized as a face, people generally tend to focus on that part. In other words, parts that are recognized as a face tend to be highly salient.

[0037] Furthermore, the high-order image features may include characters and fonts. When the elements that make up an image are legible characters, the type and size of the font also affect the degree to which the viewer focuses their attention. Fonts contain characters of a specific typeface, and come in a variety of types, including print, block, and cursive. The type of font used can affect the degree to which the viewer focuses their attention. Also, even within the same typeface, larger characters tend to attract more attention than smaller characters.

[0038] In addition, the control unit 21 as a generation unit generates a feature saliency map for each type of image feature that indicates the saliency in the image to be evaluated based on the image feature, and generates a saliency map that integrates all the feature saliency maps. A specific method by which the control unit 21 serving as a generating unit generates the saliency map will be described later. The control unit 21 then performs blurring processing on the evaluation target image using a Gaussian filter. This blurring processing is a process of reducing the image resolution. Specifically, a group of images (multi-resolution representation of the image, Gaussian pyramid) is generated for each low-order image feature by applying multiple Gaussian filters with varying degrees of blurring to the evaluation target image in a stepwise manner.

[0039] When the control unit 21 generates a group of images (Gaussian pyramid) for each component of the image feature, it uses this multi-resolution representation to obtain (calculate) the inter-image differences at different scales for each element of the image feature. In the process of calculating the inter-image difference, the control unit 21 converts at least one of the color difference and the luminance difference from the RGB data to L * a * b * Calculated based on color space. * a * b * The color space is more suited to human perception of color difference than the RGB color space. Therefore, at least one of the color difference and luminance difference is expressed as L * a * b *Calculation using a color space provides the following benefits. Specifically, the luminance contrast and chromaticity contrast values ​​extracted from the image to be evaluated can be expressed as brightness differences and color differences that match human perception. Therefore, the saliency indicated by the final saliency map can be more closely matched to human perception.

[0040] After the difference image is acquired, the control unit 21 normalizes it and combines feature maps of all scales for each component of the image feature amount, and then generates a feature saliency map according to the feature maps. When viewed for each low-level image feature, the feature saliency map shows that areas with large color contrasts (red-green or yellow-blue) are highly salient for color components. For luminance components, the map shows that the boundary between the black screen and the white area of ​​a laptop computer is highly salient. For directional components, the edges of the notebook or laptop computer are highly salient.

[0041] In this embodiment, processing fluency (complexity, etc.), position bias, and face components are extracted as high-order image features by the control unit 21 as a feature extraction unit. Although processing fluency, position bias, and face components are exemplified as high-order image features here, as mentioned above, high-order image features are not limited to these and may include various other elements (components).

[0042] As mentioned above, processing fluency can be measured by the degree of complexity, and can be analyzed and quantified using a technique called fractal dimension, for example. That is, the image to be evaluated is divided into multiple meshes, and an analysis is performed to determine which parts have dense structures, as represented by dots, and which parts have sparse structures. As a result, parts with a high fractal dimension are evaluated as complex and messy. Parts with a low fractal dimension are evaluated as simple and with little information.

[0043] As mentioned above, the background areas with almost no information have a low fractal dimension, but are not highly noticeable and therefore have low saliency. Therefore, in the feature saliency map for processing fluency, the background areas with almost no information or overly complex areas are evaluated as having low saliency, while areas with moderate complexity are evaluated as having the highest saliency.

[0044] Furthermore, the control unit 21 generates a feature saliency map of position bias according to the location and direction to which the gaze is likely to be guided, taking into account the psychological characteristics of humans, in accordance with the characteristics and type of the image to be evaluated. The characteristics and type of the image include, for example, whether it is an image intended to be published in a book or on a web page, an image to be inserted into a vertically written document, etc. For example, if the image to be evaluated is one that will be posted on a web page, the map will show high salience in the upper left corner of the screen and low salience in the lower right corner.

[0045] Furthermore, the control unit 21 extracts areas that can be recognized as faces from the image to be evaluated using a face area detection AI or the like, and generates a face component feature saliency map, which is a map in which areas that can be recognized as faces have high saliency.

[0046] Then, once the feature saliency maps are generated for each of the low-order and high-order image features, the control unit 21 integrates the feature saliency maps. Then, the control unit 21 calculates where the gaze will go as a whole when a person looks at the image to be evaluated, and which parts will attract high attention and gaze.

[0047] Furthermore, in the process of integrating all feature amount saliency maps, the control unit 21 generates the saliency map so that the sum of the similarities between the saliency map and all feature amount saliency maps is maximized. Specifically, the control unit 21 generates the saliency map so as to satisfy the following formula (1).

number

[0048] By generating a saliency map so as to satisfy the above formula (1), the following saliency map can be generated. Specifically, when a plurality of feature amount saliency maps emphasize the same portion in the image to be evaluated, the saliency map emphasizes the same portion to the same degree as the feature amount saliency map. In addition, we will explain a case where one feature saliency map overemphasizes a part of the image to be evaluated, while another feature saliency map does not overemphasize that part. In this case, the other feature saliency map does not overemphasize that part.

[0049] Return to the explanation of the configuration of Cloud Server 2. The communication unit 22 is capable of transmitting and receiving various signals and various data to and from other devices connected via the communication network N.

[0050] The storage unit 23 is configured by a non-volatile semiconductor memory, a hard disk, or the like, and stores various programs executed by the control unit 11, parameters required for executing the programs, various data, and the like.

[0051] <External device 3> Next, the configuration of the external device 3 will be described with reference to Fig. 1. As shown in Fig. 1, the external device 3 includes a control unit 31, a display unit 32, an operation unit 33, a communication unit , and a storage unit .

[0052] The control unit 31 is configured with a CPU, RAM, etc. The CPU of the control unit 31 reads out various programs stored in the storage unit 35, expands them in the RAM, executes various processes in accordance with the expanded programs, and controls the operation of each part of the external device 3.

[0053] The display unit 32 is configured with a monitor such as an LCD, and displays various screens and the like in accordance with instructions of a display signal input from the control unit 31.

[0054] The operation unit 33 is a keyboard equipped with cursor keys, numeric input keys, various function keys, etc., a pointing device such as a mouse, a touch panel laminated on the surface of the display unit 32, etc. The operation unit 33 is configured to be operable by an operator. The operation unit 33 also outputs various signals to the control unit 31 based on operations performed by the operator.

[0055] The communication unit 34 is capable of transmitting and receiving various signals and various data to and from other devices connected via the communication network N.

[0056] The storage unit 35 is configured by a non-volatile semiconductor memory, a hard disk, or the like, and stores various programs executed by the control unit 11, parameters required for executing the programs, various data, and the like.

[0057] <Image generation support processing> Next, the image generation support process will be described using the flows shown in FIG. 2 (input process), FIG. 5 (design editing process), FIG. 11 (layout editing process), and FIG. 15 (catchphrase generation process). The image generation support process is a process that supports a user in generating an image or a catchphrase based on first text data and / or image data input by a user such as a salesperson of a printing company or a customer of the printing company. Although not shown in the flows shown in FIGS. 2, 5, 11, and 15, the various types of generated data are ultimately transmitted to the designer's terminal (external device 3). It is assumed that the image generation support screen D1 shown in FIG.

[0058] (Input processing) Here, the image generation support screen D1 shown in FIG. 3 will be described. Area A1 is an area where an image input (uploaded) by the user is displayed. Area A2 is an area where the user inputs first text data. The first text data is information about the product to be designed. In other words, the first text data does not have to be words that are typically input to image generation AI, such as direct motifs or descriptions desired for the image to be generated by the image generation AI. The first text data may be, for example, the product name, product type (such as the "name" under the Food Sanitation Act), target (anticipated purchasing demographic, users), product concept, etc. The first text data may include a keyword for grouping image generation jobs. An image generation job refers to a series of processes from the start to the end of image generation support processing. The keyword for grouping is, for example, client information such as the company name of the user's customer ("Customer's Company Name" in Figure 3). As an internal process, the client information is managed using a client ID. Button B1 is a button for the user to input (upload) an image. Button B2 is a button for causing the cloud server 2 to generate an image (generate a key image). A key image is a generated image that serves as a key for subsequent image generation, and subsequent image generation is performed based on a policy of deriving from the key image. Note that when the fine adjustment mode described below is selected, the degree of derivation is narrower than usual; in other words, images that are closer to the key image and similar to each other are generated.

[0059] The control unit 21 associates various image information input or generated in the image generation support process with a restriction condition that restricts sharing of the image information at the time of input or generation, and stores the information in the storage unit 23 (related step). For example, the restriction condition is a keyword for grouping. In this embodiment, the various information is associated with client information. Furthermore, the control unit 21 restricts the generation or display of image information based on the restriction conditions (restriction step). Specific processing details will be described later.

[0060] First, the user inputs first text data and / or image data using the operation unit 13. The control unit 11 accepts the input first text data and / or image data (step S1). Next, the user presses button B2 using operation unit 13. Control unit 11 transmits the first text data and / or image data to cloud server 2 via network N (step S2).

[0061] Next, the control unit 21 generates second text data by a sentence generation AI using the first text data and / or image data received via the network N (step S3; text generation step). The control unit 21 generates one or more pieces of second text data from the first text data. In particular, the control unit 21 may generate a plurality of different second text data from the first text data. Here, "different" refers to differences in one or more of the following: "motif (object to be drawn, noun)," "descriptor (atmosphere, style)," and "color name (color instruction)." In particular, if there are multiple words of the same type in the second text data, it is preferable that the word order of at least one of the words of the same type in the second text data is different for less than half the number of words of the same type. Here, "same type" refers to, for example, nouns. Furthermore, "less than half" refers to, for example, the first noun if there are two nouns in the prompt, or the first two nouns if there are five nouns. Furthermore, it is advisable to generate the second text data by prioritizing combinations that result in a greater number of different words.

[0062] Furthermore, the control unit 21 may generate the second text data so that it is different from the second text data associated with information about a second client that is in a competitive relationship with the first client (competitive client information). To explain the specific processing, the control unit 21 acquires second client information that is in a competitive relationship with the client of the first client information based on the first client information (acquisition step), and then the control unit 21 restricts the generation of second text data that is identical to or similar to the second text data associated with the second client information (restriction step). The competing client information may be input by the user using the operation unit 13 and stored in the storage unit 23 by the control unit 21. Furthermore, the control unit 21 may automatically determine competing client information from the client information and store it in the storage unit 23. For example, the control unit 21 can acquire competing client information from the client information and information on the Internet by using AI or the like.

[0063] Next, the control unit 21 generates generated images by the image generation AI using the second text data (step S4; image generation step). The control unit 21 generates generated images as many times as the number of second text data.

[0064] Next, the control unit 21 extracts (generates) design elements from the generated image using the image generation AI (step S5). The design elements are elements that represent the generated image, and are expressed by combining nouns and modifiers for the nouns. Specifically, the control unit 21 extracts design elements from the generated image generated using "cat peeking into a cup" (second text data), and sets the priority of the design elements in order of salience using the above-mentioned salience analysis. Note that the priority of design elements may be set without performing saliency analysis. For example, suppose the second text data is in the form of a "simple background illustration of a cat peering intently into a cup." The control unit 21 may extract design elements from the beginning of the second text data and set the priority of the design elements according to the order of the extracted design elements.

[0065] Furthermore, the control unit 21 may generate design elements so that the combination of design elements is different from that associated with the information of a second client that is in a competitive relationship with the first client (competitive client information). To explain the specific processing, the control unit 21 acquires second client information that is in a competitive relationship with the client of the first client information based on the first client information (acquisition step), and then the control unit 21 restricts the generation of a combination of design elements that is the same as or similar to the combination of design elements associated with the second client information (restriction step).

[0066] Next, the control unit 21 transmits the data (the generated image and design elements) generated in steps S4 and S5 to the image generation support device 1 via the network N (step S6).

[0067] Next, the control unit 11 receives the data (the generated image and the design elements) via the network N (step S7; first acquisition step).

[0068] Next, the control unit 11 uses the data received via the network N to cause the display unit 12 to display the image generation support screen D2 shown in FIG. 4 (step S8; first display step).

[0069] (Design editing process) Here, the image generation support screen D2 shown in FIG. 4 will be described. Area A3 is an area where the generated generated image is displayed. When multiple generated images are generated, the selected generated image is displayed large on the top row, and the other generated images are displayed small and side-by-side on the bottom row. The initially selected generated image is the left generated image on the bottom row. Area A4 is an area for editing the generated image. The content displayed in area A4 changes when the user selects one of tabs TB1 to TB4. In FIG. 4, the design element tab TB1 is selected, and in area A4, the design elements are displayed vertically by noun, with modifiers for each noun displayed horizontally to the right of each noun. The nouns at the top of the design elements have a higher priority, and the modifiers to the left have a higher priority. The control unit 21 weights the design elements according to their priority and generates the second text data. Note that the method of expressing the priority of the design elements is not limited to this example; the nouns at the bottom may have a higher priority, and the modifiers to the right may have a higher priority. Furthermore, although the design elements are expressed using nouns and modifiers, this is not limiting. The contents of the other tabs will be described later. Button B3 is a button for causing cloud server 2 to generate a new generated image. Button B4 is a button for causing the cloud server 2 to extract design elements again. Buttons B5 and B6 are buttons that allow the user to manually add design elements. When these buttons are pressed, a design element field is added, and the user can enter data into that field. The user can also delete a design element by selecting it using the operation unit 13 and then pressing, for example, the DELETE button on the keyboard (operation unit 13). The button B8 is a button for recording the user's evaluation (OK (good), NG (bad)) of the generated image. When button B3 is pressed to generate an image again, button B11 is a button for generating an image by fine-tuning (fine-tuning mode) based on the image selected in area A3, rather than generating an image from scratch. The image generation process by fine-tuning performed by control unit 21 will be described later.

[0070] First, the user edits the design elements in area A4 using operation unit 13. Control unit 11 accepts the edited design elements (step S11; editing step). Specifically, users can change the priority order by dragging and dropping design elements. Users can drag and drop rows (sets of nouns and modifiers). Users can also drag and drop modifiers within a row to change their order. Users can also drag and drop modifiers from one row to move them to modifiers in another row. Additionally, as shown in Figure 6, by clicking on a design element, the user can display a list of related terms associated with that design element. The user can replace the design element by selecting a word from the list of related terms. In this case, the control unit 11 causes the cloud server 2 to generate and acquire related terms associated with the clicked design element. For example, the control unit 21 of the cloud server 2 generates a list of related terms, such as two proposals for changing the keyword expression of the design element "expressive" into modifiers ("transparent" and "glass-like"), two modifiers that match the list of design elements displayed in area A4 ("shining in the moonlight" and "having a soft, airy feel"), and one random modifier unrelated to the list of design elements ("heralding the arrival of spring"). Furthermore, the user can add new design elements using buttons B5 and B6 as described above, and can delete design elements using operation unit 13 as described above. The image generation support screen D2 shown in FIG. 7 is an example in which the row of cups has been moved above the row of cats in the area A4 of the image generation support screen D2 shown in FIG.

[0071] The control unit 21 may display related terms of the design elements in a manner different from the combination of design elements associated with the information of the second client that is in a competitive relationship with the first client (competitive client information). To explain the specific processing, the control unit 21 acquires second client information that is in a competitive relationship with the client of the first client information based on the first client information (acquisition step), and then the control unit 21 restricts the generation of related terms so that the related terms are different from combinations of design elements that are the same as or similar to the combination of design elements associated with the second client information (restriction step).

[0072] Next, the user presses the image generation button B3 using the operation unit 13. The control unit 11 transmits the edited design elements and the priority order to the cloud server 2 via the network N (step S12).

[0073] Next, the control unit 21 generates a generated image using the image generation AI using the design elements and priorities received via the network N (step S13). Note that the control unit 21 generates the generated image taking into account the content of the image generation process in step S4. Next, the control unit 21 transmits the image data generated in step S13 to the image generation support device 1 via the network N (step S14).

[0074] If the button B11 is pressed, in step S13, the control unit 21 generates an image by making fine adjustments based on the image selected in the area A3. When the fine adjustment mode is selected, the newly drawn images are closer to the original images and more similar to each other than in normal cases (when the button B11 is not pressed). When the fine adjustment mode is selected, the control unit 21 generates an image by making fine adjustments by changing the image generation method from the normal mode as follows. The second text data (prompt) given to the image generation AI is the same, and only the random number seed value is changed. Reduce the number and / or percentage of words to be changed in the second text data (prompt) given to the image generation AI. -In the prompts given to the image generation AI, the words to be changed should be adjectives only (not nouns) - Reduce the weighting of words that change in the prompts given to the image generation AI - Move the position of the word to be changed later in the prompt given to the image generation AI Do It should be noted that a plurality of the above-mentioned specific methods may be applied simultaneously. Furthermore, the control unit 21 may make fine adjustments by directly processing the prompt, or by processing the prompt generation instruction for the image generation AI to the sentence generation AI.

[0075] Next, the control unit 11 receives the image data via the network N (step S15; second acquisition step).

[0076] Next, the control unit 11 uses the data received via the network N to cause the display unit 12 to display the image generation support screen D2 shown in FIG. 8 (step S16; second display step). In area A3 of image generation support screen D2 shown in FIG. 8, a generated image generated by prioritizing the cup is displayed.

[0077] Next, the control unit 11 determines whether or not the user has made an input to area A4 (step S17). If there has been an input (step S17; YES), the control unit 11 determines that the design editing should be re-executed, and proceeds with the image generation support process to step S11. If there has not been an input (step S17; NO), the control unit 11 proceeds with the image generation support process to step S21.

[0078] (Layout editing process) First, the user uses the operation unit 13 to select tab TB2 as shown in Fig. 10, and selects layout data in area A4 to edit the layout of the image displayed in area A3. The control unit 11 accepts the selected layout data (step S21; reflection step). The layout data is data on the layout to be applied to the image displayed in area A3, such as data on a generation area mask, a display area mask, and background-transparent characters. The generation area mask is a mask that indicates the area (shaded area) where the cloud server 2 is to generate an image. The generation area mask is selected by the user from the pull-down menu PD1. The contents of the pull-down menu PD1 include, for example, "No specification," "Circle," "Circle (reverse)," "Circle gradation," and "Gradation." The display area mask is a mask that indicates the area (shaded area) where you want to display the image. The display area mask is selected by the user from the pull-down menu PD2. The contents of the pull-down menu PD2 include, for example, "No Selection," "Circle," "Circle (Inverted)," "Circle Gradient," and "Gradation." The background-transparent text is an image in which the background other than the text portion is transparent. For example, the text portion is a catchphrase. The catchphrase may be a catchphrase (third text data) generated in step S33 described later. The background-transparent text is selected by the user from the pull-down menu PD3. The layout data can be registered at any time, such as before or during the image generation support process. Furthermore, the control unit 21 may display only the layout data associated with the client information. Furthermore, the control unit 21 acquires second client information that is in a competitive relationship with the client of the first client information based on the first client information (acquisition step), and may restrict the display of layout data that is the same as or similar to the layout data associated with the second client information (restriction step). The image generation support screen D2 shown in FIG. 10 is a screen display when "Circle" is selected in the pull-down menu PD1 for the generation area mask. Next, the user presses the image generation button B2 using the operation unit 13. The control unit 11 transmits the layout data to the cloud server 2 via the network N (step S22).

[0079] Next, the control unit 21 generates an image in which the layout data received via the network N is reflected in the generated image (step S23). Next, the control unit 21 transmits the image data generated in step S23 to the image generation support device 1 via the network N (step S24).

[0080] Next, the control unit 11 receives the image data via the network N (step S25).

[0081] Next, the control unit 11 uses the image data received via the network N to cause the display unit 12 to display the image generation support screen D2 shown in FIG. 4 (step S26; third display step). Image generation support screen D2 shown in FIG. 11 is a screen display when "Circle" is selected in the display area mask pull-down PD2 and an image reflecting the display area mask is displayed in area A3. The image generation support screen D2 shown in Figure 12 is the screen display when "Character Sample 1" is selected in the background-transparent character pull-down menu PD3, and a generated image reflecting the background-transparent character is displayed in area A3. Even if the image generation button B2 is pressed again and image generation is performed again, the content edited in the layout tab TB3 is maintained in the generated image displayed in the area A3.

[0082] Next, the control unit 11 determines whether or not the user has input into area A4 (step S27). If there has been input (step S27; YES), the control unit 11 determines that layout editing is to be re-executed, and proceeds with the image generation support process to step S21. If there has been no input (step S27; NO), the control unit 11 proceeds with the image generation support process to step S31.

[0083] (Catchphrase generation process) Next, the user uses the operation unit 13 to select tab TB3 as shown in Fig. 14 and specify an area of ​​interest A5 in area A3. The control unit 11 accepts the specified area of ​​interest A5 (step S31; accepting step). For example, the image generation support screen D2 shown in Fig. 14 is a screen display when a glass is surrounded by the area of ​​interest A5. The attention area is an area to be focused on when generating a catchphrase. In the example of Fig. 14, attention area A5 is specified by enclosing it on the generated image, but it may be configured so that a sentence such as "the cup on the right side of the generated image" can be input, or the coordinates of the attention area in the generated image can be input, or it may be configured so that an area can be selected from areas obtained by dividing the image (for example, into 16 parts). Next, the user presses the catchphrase generation button B10 using the operation unit 13. The control unit 11 transmits data indicating the attention area to the cloud server 2 via the network N (step S32). Furthermore, when generating a catchphrase, text data may be input into area A6. In this case, control unit 11 also transmits the text data to cloud server 2.

[0084] Next, the control unit 21 generates a catchy slogan (third text data) based on the data indicating the attention area received via the network N (step S33; generation step). Specifically, the control unit 21 generates a prompt (second text data) for the specified area, such as "Pay attention to the object in the area from coordinates x1:y1 to x2:y2," and instructs the image generation AI. The catchphrase is generated by inputting image data into an image recognition AI capable of generating explanatory text for objects in an image, and then configuring the prompt to generate the catchphrase by multiplying the explanatory text obtained by the image data with an instruction prompt regarding the area of ​​interest (in the above example, "Pay attention to the object in the area between coordinates x1:y1 and x2:y2"), and then using a text generation AI. The prompt indicating the area is described here using coordinates, but it is also possible to specify it using different wording in internal processing. For example, the prompt indicating the area could be to divide the image into four parts and focus on the bottom right. When the text data input in the area A6 is transmitted, the control unit 21 generates a catchphrase based on the text data as well.

[0085] Furthermore, the control unit 21 may generate a catch phrase so that it is different from a catch phrase associated with information about a second client that is in a competitive relationship with the first client (competitive client information). Specifically, the control unit 21 makes the prompt for generating the catch phrase different. To explain the specific processing, the control unit 21 acquires second client information that is in a competitive relationship with the client of the first client information based on the first client information (acquisition step), and then the control unit 21 regulates the generation of a catch phrase so that the catch phrase is different from a catch phrase that is the same as or similar to the catch phrase associated with the second client information (regulation step).

[0086] Next, the control unit 21 transmits the catchphrase (third text data) generated in step S33 to the image generation support device 1 via the network N (step S34).

[0087] Next, the control unit 11 receives the catch phrase (third text data) via the network N (step S35).

[0088] Next, the control unit 11 uses the data received via the network N to display an image generation support screen D2 shown in Fig. 15 on the display unit 12 (step S36). In the image generation support screen D2 shown in Fig. 15, a catchphrase is generated in an area A7.

[0089] Next, the user presses (ON) the attention analysis button B9 using the operation unit 13. The control unit 11 instructs the cloud server 2 to perform attention analysis via the network N (step S37).

[0090] Next, the control unit 21 generates a characteristic image (second image) (step S38). A typical example of a feature image that visualizes human attention is a saliency map. A saliency map is used as a computational visualization method to determine which parts of an image are most noticeable, and displays the areas where the gaze first turns as a heat map. Next, the control unit 21 transmits the characteristic image generated in step S38 to the image generation support device 1 via the network N (step S39).

[0091] Next, the control unit 11 uses the data received via the network N to display an image generation support screen D2 shown in Fig. 16 on the display unit 12 (step S311; display step). The image generation support screen D2 shown in Fig. 16 is an example in which a characteristic image is displayed in the area A3. Note that the image display can be restored to normal by pressing (OFF) the attention analysis button B9.

[0092] Next, the control unit 11 determines whether or not the user has input into area A4 (step S312). If there has been input (step S312; YES), the control unit 11 determines that the area of ​​interest designation should be re-executed, and proceeds with the image generation support process to step S31. If there has been no input (step S312; NO), the control unit 11 ends the image generation support process.

[0093] The execution order of the design editing process, layout editing process, and catchphrase generation process is not limited to the above example, and they can be executed in response to the user switching between tabs TB1 to TB4. In addition, in the catch phrase generation process, the catch phrase generation (steps S31 to S36) and the characteristic image generation (steps S37 to S311) may be performed in reverse order. In other words, they can be performed in response to a button press by the user.

[0094] <Other> In the above, the image generation support process has been described with the image generation support device 1 and the cloud server 2 as separate devices, but the image generation support device 1 may be given various functions of the cloud server 2, and the image generation support process may be executed in the image generation support device 1. In this case, the processing content of each step is the same as each step of the image generation support processing, but there is no need to send and receive data between the image generation support device 1 and the cloud server 2. For example, the control unit 11 does not receive data via the network N, but instead acquires the data directly.

[0095] It should be noted that an image may be uploaded directly to area A3 on the image generation support screen in Fig. 4. Furthermore, an attention area A5 may be set in the image, and a catchy slogan may be generated.

[0096] Furthermore, in the above example, a catch phrase is described, but it is not limited to a catch phrase and may be any text data.

[0097] In the above description, the control unit 21 generates a catchphrase directly from an image, but the catchphrase may also be generated based on design elements extracted during image generation. When generating a catchphrase from design elements only, step S31 of setting a region of interest is not required. The control unit 21 may also generate a catchphrase based on the priority associated with the design elements. For example, the user may initially set "cat" as the design element with the highest priority and generate a catchphrase that focuses on "cat," but then switch the priorities of "cat" and "cup" and generate a catchphrase that focuses on "cup" as the highest priority.

[0098] Furthermore, the control unit 21 may manage data including a set of first data (first text data and / or image data) and at least one generated image corresponding to the first data as a history. The control unit 11 may then display a separate history screen when the button B7 is pressed. Note that design elements and the like may also be included in the history in addition to the generated image. The history may also be acquired each time an image is generated. If there are multiple sets, they may be displayed in chronological order. The first data may also include a keyword for grouping image generation requests. In this case, related information related to the generated image may be associated with the keyword and managed. The related information may include the generated image and information used when generating the image. When a separate history screen is displayed upon pressing button B7, control unit 11 may display on display unit 12 only the history of related information linked to the same keyword, so that the user can refer to the history of the related information linked to the same keyword, depending on the keyword entered on image generation support screen D1 shown in FIG. 3. The set may also include "reactions to the generated image (evaluations such as OK / NG and free-form comments)." The set may also include "input details of correction instructions for the previous generated image (such as changes to design elements and layout)."

[0099] Furthermore, in the above description, various pieces of information input or generated in the image generation support process are associated with restrictive conditions when they are input or generated, but they do not have to be associated. If there is image information that is not associated with the restrictive condition, in the restricting step, the control unit 21 restricts the generation or display of image information that is the same as or similar to the image information that is not associated with the restrictive condition. To explain the specific processing, the control unit 21 restricts the generation of second text data that is the same as or similar to second text data that is not associated with the restrictive condition (restriction step). To explain the specific processing, the control unit 21 restricts the generation of a combination of design elements that is the same as or similar to a combination of design elements that is not associated with the restrictive condition (restriction step). To explain the specific processing, the control unit 21 restricts the generation of related terms so as to differentiate them from combinations of design elements that are the same as or similar to combinations of design elements that are not associated with the restrictive condition (restriction step). Furthermore, the control unit 21 restricts the display of layout data that is the same as or similar to layout data that is not associated with the restrictive condition (restriction step). To explain the specific processing, the control unit 21 regulates the generation of a catchy slogan so that the catchy slogan is different from a catchy slogan that is the same as or similar to a catchy slogan that is not associated with the regulating condition (regulating step).

[0100] The display area mask may also be a 3D image, a 2D image of a 3D image viewed from a certain direction, a template image of a standard product such as a 350 mm can, or a template image of a direct mail piece. Furthermore, the display area mask and the background transparent character image may be glossed or shaded depending on the template image. For example, in the case of a 350 mm can, the can itself or the characters such as a logo formed on the can may be glossed or shaded. FIG. 17 is an image diagram of the image generation support screen when "350 ml can" is selected in the display area mask and "TWOCAT BEER" is selected in the background transparent text. In this way, by combining the product appearance image with the generated image, the user can get a concrete idea of ​​the finished image. In addition, the user can change the product appearance image in the upper part of area A3 by selecting a candidate image displayed in the lower part of area A3, allowing the user to select an image including the appearance elements. In other words, this invention contributes to reducing the total cost and time of design production by eliminating rework in design production. Specifically, if 100 designs are ordered from a designer and the final candidates are narrowed down to 10 and one proposal is selected at a final meeting, using this invention can reduce investment in rejected designs. The display area mask and the background-transparent character images may be generated using an image generation AI. For example, the image generation AI may be caused to read the appearance of a 350 mm can and generate the display area mask and the background-transparent character images.

[0101] 18 is an image diagram of the image generation support screen when the tone and manner tab TB4 is selected. In the example of Fig. 18, the tone and manner tab TB4 allows the user to select style, composition, and color tone using pull-down menus. Style is the element that characterizes an image, such as color expression, texture, and illustration technique. Composition is a design element that refers to the arrangement of elements within an image, the overall balance of the image, and the relative positions of objects and elements. Hue refers to the qualities and characteristics of color, such as color scheme, color balance, and color contrast. The control unit 21 can change the tone of the image displayed in the area A3 according to the style, composition, and color tone selected by the user.

[0102] FIG. 19 is an image diagram of the history screen D3. The history is acquired at a predetermined timing, such as when an image is generated or when a user selects an evaluation (OK / NG). Area A8 displays the first text data entered by the user in area A2. In the area A9, the generated image displayed in the area A3 and the contents of each of the tabs TB1 to TB4 are displayed. Area A10 is a field where the user can enter any comments. The button B12 is a button for selecting a history of an arbitrary timing from the recorded history and restarting the image generation support process from that timing.

[0103] <Effect 1> As described above, the program causes the computer to execute a first acquisition step (step S7) of acquiring text data generated based on an image, a first display step (step S8) of displaying the acquired text data on a display unit, an editing step (step S11) of editing the text data after displaying it, a second acquisition step (step S15) of acquiring a generated image generated based on the edited text data, and a second display step (step S16) of displaying the acquired generated image on a display unit. This allows users to generate the image they want with simple operations (such as rearranging the order of design elements).Also, by editing the text generated from the image generated based on the customer's request and regenerating the image, it is possible to quickly obtain the desired image that is likely to satisfy the customer.

[0104] The image generation support device 1 also includes a first acquisition unit (control unit 11) that acquires text data generated based on an image, a first display control unit (control unit 11) that displays the acquired text data on a display unit, an editing unit (control unit 11) that edits the text data after displaying it, a second acquisition unit (control unit 11) that acquires a generated image generated based on the edited text data, and a second display control unit (control unit 11) that displays the acquired generated image on a display unit. This allows the user to generate the image they want with a simple operation.

[0105] The image generation support system 100 also includes a first acquisition unit (control unit 11) that acquires text data generated based on an image, a first display control unit (control unit 11) that displays the acquired text data on a display unit, an editing unit (control unit 11) that edits the text data after displaying it, a second acquisition unit (control unit 11) that acquires a generated image generated based on the edited text data, and a second display control unit (control unit 11) that displays the acquired generated image on a display unit. This allows the user to generate the image they want with a simple operation.

[0106] In addition, the image generation support method is an image generation support method performed by an image generation support device, and includes a first acquisition step (step S7) of acquiring text data generated based on an image, a first display step (step S8) of displaying the acquired text data on a display unit, an editing step (step S11) of editing the text data after displaying it, a second acquisition step (step S15) of acquiring a generated image generated based on the edited text data, and a second display step (step S16) of displaying the acquired generated image on a display unit. This allows the user to generate the image they want with a simple operation.

[0107] <Effect 2> As described above, the program causes the computer to execute a reception step (step S31) of receiving a designation by the user using a designation means for an area of ​​interest in the first image, and a generation step (step S33) of generating and outputting text data corresponding to the first image in accordance with the area of ​​interest received in the reception step. This allows the user to generate the intended text data with simple operations. In recent years, text generation AI capable of generating text data such as catchphrases has emerged, but in order to generate the intended catchphrase, it is necessary to construct appropriate prompts, which requires training. With the present invention, such training is no longer necessary.

[0108] The image generation support device 1 also includes a reception unit (control unit 11) that receives a user's designation of an area of ​​interest in the first image using a designation means, and a generation unit (control unit 11) that generates and outputs text data corresponding to the first image according to the area of ​​interest received in the reception step. This allows the user to generate the intended text data with simple operations.

[0109] The image generation support system 100 also includes a reception unit (control unit 21) that receives a user's designation of an area of ​​interest in the first image using a designation means, and a generation unit (control unit 21) that generates and outputs text data corresponding to the first image according to the area of ​​interest received in the reception step. This allows the user to generate the intended text data with simple operations.

[0110] In addition, the image generation support method is an image generation support method performed by an image generation support system, and includes a reception step (step S31) of receiving a designation by a user using a designation means for an area of ​​interest in a first image, and a generation step (step S33) of generating and outputting text data corresponding to the first image in accordance with the area of ​​interest received in the reception step. This allows the user to generate the intended text data with simple operations.

[0111] <Effect 3> As described above, the program causes the computer to execute a text generation step (step S3) in which the AI ​​generates second text data to be given to the image generation AI from the first data, and an image generation step (step S4) in which the image generation AI generates an image from the second text data. This allows users to easily obtain the image they want, regardless of their design knowledge or range of ideas. Conventional image generation instructions using prompt input require input of words that describe the desired motif, description, and other elements of the image, which depend on the user's design knowledge and range of ideas. Furthermore, generating prompts requires intuition and skill, and depends on the user's level of proficiency. As a result, it has been difficult for users without specialized design knowledge, such as salespeople at printing companies, to generate images with simple operations and obtain the intended image. The present invention eliminates this need for proficiency.

[0112] The first data is text data, and the text data is information about a product that is the subject of the design. This allows users to easily obtain the image they want, regardless of their design knowledge or range of ideas.

[0113] The image generation support device 1 also includes a text generation unit (control unit 11) that causes the AI ​​to generate second text data to be given to the image generation AI from the first data, and an image generation unit (control unit 11) that causes the image generation AI to generate a generated image from the second text data. This allows users to easily obtain the image they want, regardless of their design knowledge or range of ideas.

[0114] The image generation support system 100 also includes a text generation unit (control unit 21) that causes the AI ​​to generate second text data to be given to the image generation AI from the first data, and an image generation unit (control unit 21) that causes the image generation AI to generate a generated image from the second text data. This allows users to easily obtain the image they want, regardless of their design knowledge or range of ideas.

[0115] In addition, the image generation support method is an image generation support method performed by an image generation support device, and includes a text generation step (step S3) in which second text data to be given to the image generation AI is generated from the first data by the AI, and an image generation step (step S4) in which an image is generated from the second text data by the image generation AI. This allows users to easily obtain the image they want, regardless of their design knowledge or range of ideas.

[0116] <Effect 4> As described above, the program causes the computer to execute an association step of associating image information, including the generated image and information used when generating the image, with regulatory conditions that regulate the sharing of the image information, and a regulation step of regulating the generation or display of the image information based on the regulatory conditions. This prevents information leaks between users and prevents users from creating similar images, while allowing users to create the images they intend with simple operations.

[0117] The program also causes the computer to execute an acquisition step of acquiring second client information that is in a competitive relationship with the client of the first client information based on the first client information, and a restriction step of restricting the generation or display of image information that is identical to or similar to the image information associated with the second client information. This allows users to generate the image they intend with simple operations, while preventing information leakage between users who are in a competitive relationship and preventing similar images from being generated between users who are in a competitive relationship.

[0118] In addition, the image generation support device 1 is an image generation support device that generates an image from information input by a user, and is equipped with an association unit (control unit 11) that associates image information including the generated image and information used when generating the image with regulatory conditions that regulate the sharing of the image information, and a regulation unit (control unit 11) that regulates the generation of image information based on the regulatory conditions. This prevents information leaks between users and prevents users from creating similar images, while allowing users to create the images they intend with simple operations.

[0119] In addition, the image generation support system 100 is an image generation support system that generates an image from information input by a user, and is equipped with an association unit (control unit 21) that associates image information including the generated image and information used when generating the image with regulatory conditions that regulate the sharing of the image information, and a regulation unit (control unit 21) that regulates the generation of image information based on the regulatory conditions. This prevents information leaks between users and prevents users from creating similar images, while allowing users to create the images they intend with simple operations.

[0120] In addition, the image generation support method is an image generation support method that causes a computer to generate an image from information input by a user, and includes an association step of associating image information including the generated image and information used when generating the image with regulatory conditions that regulate the sharing of the image information, and a regulation step of regulating the generation or display of the image information based on the regulatory conditions. This prevents information leaks between users and prevents users from creating similar images, while allowing users to create the images they intend with simple operations.

[0121] The present invention has been specifically described above based on an embodiment, but it goes without saying that the present invention is not limited to the above embodiment and can be modified as appropriate within the scope of the gist of the present invention. For example, in the above description, examples have been disclosed in which a hard disk or a semiconductor nonvolatile memory is used as a computer-readable medium for the program according to the present invention, but the present invention is not limited to this example. Other computer-readable media may also be portable recording media such as a CD-ROM.

[0122] In addition, the detailed configuration and operation of each device can be modified as appropriate without departing from the spirit of the invention. [Explanation of symbols]

[0123] 100 Image Generation Support System 1. Image generation support device 11 control unit (first acquisition unit, first display control unit, editing unit, second acquisition unit, second display control unit) 12 Display section 13 Control section 14 Communication unit (first generation unit, reception unit, generation unit, setting unit, text generation unit, image generation unit, display control unit, related unit, regulation unit, acquisition unit) 15 Storage section 2. Cloud Server 21 Control Unit 22 Communications Department 23 storage section 3 External device 31 Control Unit 32 Display section 33 Operation section 34 Communications Department 35 Storage section N Communication Network

Claims

1. On your computer a text generation step of generating second text data to be provided to the image generation AI from the first data by the AI; an image generation step of generating an image by an image generation AI from the second text data; A program that executes the following.

2. 2. The program according to claim 1, wherein the first data is text data relating to a product to be designed.

3. the text generating step generates a plurality of different second text data from the first data; 2. The program according to claim 1, wherein the image generating step generates a plurality of generated images from a plurality of different second text data.

4. the first data further includes a keyword for grouping the generation requests of the generated image; Related information relating to the generated image is managed in association with the keyword; The computer, a display step of displaying only the history of the related information linked to the same keyword on a display unit so that the history can be referenced; The program according to claim 1, which executes the above.

5. a text generation unit that generates second text data from the first data by AI to be provided to the image generation AI; an image generation unit that generates a generated image from the second text data using an image generation AI; An image generation support device comprising:

6. a text generation unit that generates second text data from the first data by AI to be provided to the image generation AI; an image generation unit that generates a generated image from the second text data using an image generation AI; An image generation support system comprising:

7. An image generation support method performed by an image generation support device, a text generation step of generating second text data to be provided to the image generation AI from the first data by the AI; an image generation step of generating an image from the second text data by an image generation AI; An image generation assistance method including:

Images