Image generation system, image generation method, and program

Abstract

The goal is to easily generate image data that users desire, based on their image concept. [Solution] The image generation system 1 comprises an acquisition unit, a generation control unit, an evaluation unit, and a selection unit. The acquisition unit acquires concept information, including the image concept desired by the user, and a target value for the evaluation value of the image data. The generation control unit causes the generation AI model to generate multiple image data based on prompts corresponding to the acquired concept information. The evaluation unit evaluates each of the generated image data. The selection unit selects at least one image data whose evaluated value is closest to the target value.

Description

Technical Field

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

Background Art

[0002] Conventionally, it has been difficult for non-designers to create effective advertisements, flyers, etc., and they had no choice but to order them externally at a high cost. Therefore, a support service that provides various templates and creates image data such as advertisements based on the templates is known. However, this is only about providing templates and is not a method for finally finishing with an effective design.

[0003] Therefore, techniques for presenting designs are known. For example, a design correction device that performs attention evaluation and impression evaluation on an input image and presents amendments to the color and brightness of the design based on the evaluation results is known (see Patent Document 1).

[0004] Also, a design generation device that displays an area changed to a color correlated with an input keyword using pre-mapped word-color correlation information is known (see Patent Document 2).

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0006] The conventional design modification and generation devices described above do not always produce modifications or changes that are desired by the user (especially non-designers). Furthermore, the proposed modifications and changes only alter the colors of existing images and cannot propose new image data.

[0007] The objective of this invention is to easily generate image data desired by the user, based on the concept of the image. [Means for solving the problem]

[0008] To solve the above problem, the image generation system of the invention described in claim 1 is: An acquisition unit that acquires concept information including the image concept desired by the user and a target value in the evaluation value of the image data, A generation control unit that causes a generation AI model to generate multiple image data based on prompts corresponding to the acquired concept information, An evaluation unit that evaluates each of the generated image data, The system includes a selection unit that selects at least one image data image whose evaluated value is closer to the target value.

[0009] The invention described in claim 2 is an image generation system described in claim 1, The system includes a display control unit that displays the selected image data on the display unit.

[0010] The invention described in claim 3 is an image generation system described in claim 1, The acquisition unit acquires draft image data, concept information, and target values. The generation control unit causes the generation AI model to generate multiple image data based on prompts corresponding to the acquired draft image data and concept information.

[0011] The invention described in claim 4 is an image generation system described in claim 1, The generation control unit generates a correction prompt to bring the evaluation value closer to the target value based on the selected image data, the evaluation value, and the target value, and inputs the correction prompt to the generation AI model.

[0012] The invention described in claim 5 is an image generation system described in claim 4, The generation control unit repeats the generation of the correction prompt and the generation of image data based on the correction prompt until the evaluation value falls within a predetermined threshold from the target value.

[0013] The invention described in claim 6 is an image generation system described in claim 1, The evaluation is an evaluation of at least one of the following: the impression of the generated image data and the region of interest.

[0014] The invention described in claim 7 is an image generation system described in claim 4, The aforementioned evaluation is an evaluation of the impression of the generated image data. The generation control unit generates the correction prompt so that the evaluation value of the impression falls within a predetermined threshold from the target value.

[0015] The invention described in claim 8 is an image generation system described in claim 4, The aforementioned evaluation is an evaluation of the region of interest in the generated image data. The generation control unit generates the modification prompt to modify areas other than the area of ​​interest if the evaluation value of the area of ​​interest is within a predetermined threshold.

[0016] The invention described in claim 9 is an image generation system described in claim 1, The system includes a unit that provides the aforementioned generated AI model.

[0017] The image generation method of the invention described in claim 10 is: An acquisition step of acquiring concept information including the concept of an image desired by a user and a target value in the evaluation value of the image data, A generation control step of causing a generation AI model to generate a plurality of image data based on a prompt corresponding to the acquired concept information, An evaluation step of evaluating each of the generated image data, And a selection step of selecting at least one image data whose evaluated evaluation value is closer to the target value.

[0018] The program of the invention according to claim 11 is To cause a computer An acquisition unit that acquires concept information including the concept of an image desired by a user and a target value in the evaluation value of the image data, A generation control unit that causes a generation AI model to generate a plurality of image data based on a prompt corresponding to the acquired concept information, An evaluation unit that evaluates each of the generated image data, A selection unit that selects at least one image data whose evaluated evaluation value is closer to the target value, To function as.

Advantages of the Invention

[0019] According to the present invention, image data desired by a user can be easily generated from the concept of an image.

Brief Description of the Drawings

[0020] [Figure 1] It is a block diagram showing an image generation system according to an embodiment of the present invention. [Figure 2] It is a block diagram showing the functional configuration of a server. [Figure 3] It is a block diagram showing the functional configuration of a terminal device. [Figure 4] It is a flowchart showing an image providing process. [Figure 5] It is a diagram showing a second image and a heatmap image. [Figure 6] This is a radar chart of impressions. [Modes for carrying out the invention]

[0021] The advantages and features provided by one or more embodiments of the present invention will be better understood from the following detailed description and accompanying drawings. However, these drawings are for illustrative purposes only and are not intended to define any limitations of the invention. Embodiments of the present invention will be described below with reference to the drawings. However, the scope of the invention is not limited to the disclosed embodiments.

[0022] Embodiments of the present invention will be described with reference to Figures 1 to 6. First, the device configuration of the image generation system 1 of this embodiment will be described with reference to Figures 1 to 3. Figure 1 is a block diagram of the image generation system 1 of this embodiment. Figure 2 is a block diagram of the functional configuration of the server 10. Figure 3 is a block diagram of the functional configuration of the terminal device 20.

[0023] As shown in Figure 1, the image generation system 1 is a system that generates image data of a concept intended by the user. The image generation system 1 comprises a server 10, a terminal device 20, and an AI (Artificial Intelligence) providing device 30. The AI ​​providing device 30 functions as a providing unit. The server 10, terminal device 20, and AI providing device 30 are connected to each other so as to be able to communicate via a communication network 40. The communication network 40 is, for example, the Internet, but is not limited to this. The communication network 40 may be a wired communication network, a wireless communication network, or a LAN (Local Area Network).

[0024] Server 10 is an information processing device that provides image data of the user's intended concept to terminal device 20. Terminal device 20 is a desktop PC (Personal Computer) used by the user. Terminal device 20 is not limited to a desktop PC, but may be other information processing devices such as a palmtop PC or a smartphone.

[0025] The AI ​​provider device 30 is an information processing device that provides a generative AI model to external devices such as the server 10. The generative AI model is a trained model of a so-called image generation AI. When a prompt is input, the generative AI model automatically generates image data. The generated image data is image data that closely resembles the concept information of the input prompt. The prompt is natural language text that describes the task that the generative AI should perform. The prompt includes concept information that mainly indicates the finished image or atmosphere intended by the user. The concept information includes at least one concept in the finished image intended by the user.

[0026] The AI ​​provider 30 receives a prompt from the server 10, for example. The AI ​​provider 30 inputs the prompt into the generated AI model and generates image data that closely resembles the concept information. The AI ​​provider 30 sends the generated image data to the server 10 that sent the prompt.

[0027] Next, with reference to Figure 2, the internal functional configuration of the server 10 will be explained. As shown in Figure 2, the server 10 has a control unit 11, an operation unit 12, a storage unit 13, a display unit 14, and a communication unit 15. Each part of the server 10 is connected to the others via a bus. The control unit 11 functions as an acquisition unit, a generation control unit, an evaluation unit, a selection unit, and a display control unit.

[0028] The control unit 11 controls various parts of the server 10. The control unit 11 has a CPU (Central Processing Unit) and RAM (Random Access Memory). The control unit 11 reads various programs stored in the memory unit 13, loads them into RAM, and executes various processes in cooperation with the CPU and the loaded programs.

[0029] The operation unit 12 includes a keyboard and a pointing device such as a mouse. The operation unit 12 receives key input and position input from the user and outputs the operation information to the control unit 11.

[0030] The storage unit 13 is an HDD (Hard Disk Drive), SSD (Solid State Drive), etc. The storage unit 13 stores information such as data in a read-and-write manner. In particular, the storage unit 13 stores an image provision program. The image provision program is a program for executing the image provision process described later.

[0031] The display unit 14 has a display panel such as an LCD (liquid crystal display) or an ELD (electro-luminescent display). The display unit 14 displays display information on the display panel in accordance with the control of the control unit 11.

[0032] The communication unit 15 is a communication module such as a network card, and performs wired or wireless communication with external devices such as terminal devices 20 and AI provisioning devices 30 on the communication network 40. The control unit 11 transmits and receives information with the terminal devices 20 and AI provisioning devices 30 via the communication unit 15.

[0033] Next, with reference to Figure 3, the internal functional configuration of the terminal device 20 will be explained. As shown in Figure 3, the terminal device 20 has a control unit 21, an operation unit 22, a storage unit 23, a display unit 24, and a communication unit 25. Each part of the terminal device 20 is connected to the others via a bus.

[0034] The control unit 21 controls various parts of the terminal device 20. The control unit 21 has a CPU and RAM. The control unit 21 reads various programs stored in the memory unit 23, loads them into RAM, and executes various processes in cooperation with the CPU and the loaded programs.

[0035] The operation unit 22 includes a keyboard and a pointing device such as a mouse. The operation unit 22 receives key input and position input from the user and outputs the operation information to the control unit 21.

[0036] The storage unit 23 is an HDD, SSD, etc. The storage unit 23 stores various types of information in a read-and-write manner.

[0037] The display unit 24 has a display panel such as an LCD or ELD. The display unit 24 displays display information on the display panel in accordance with the control of the control unit 21.

[0038] The communication unit 25 is a communication module such as a network card, and performs wired or wireless communication with external devices such as the server 10 on the communication network 40. The control unit 21 sends and receives information with the server 10 and other devices via the communication unit 25.

[0039] Next, the operation of the image generation system 1 will be explained with reference to Figures 4 to 6. Figure 4 is a flowchart of the image provisioning process. Figure 5 shows the second image 50 and the heatmap image 60. Figure 6 is an impression radar chart 70.

[0040] Referring to Figure 4, the image provision process performed on server 10 will be explained. The image provision process is the process of generating a second image data that is close to the user's intended concept, or the concept and the first image data, and providing it to the terminal device 20. First, the control unit 21 of the terminal device 20 receives input from the user via the operation unit 22, which includes concept information and target values, or concept information, target values, and the first image data. The concept information is text that includes the concept of the image (design) of the second image data that the user wishes to generate. For example, if the user wishes to receive image data of an advertisement, the concept information includes the concept of the finished image and atmosphere that the user wants to include in the advertisement image. The first image data is draft image data that has image elements (keywords, items, background, etc.) that the user wishes to include in the finished image. The first image data is stored in, for example, the storage unit 23 and selected for input.

[0041] The target value is the target value for the evaluation of the generated second image data. The evaluation of the second image data is, for example, an evaluation of the area of ​​interest within the image, or an evaluation of the overall impression. Regarding the area of ​​interest, the second image data is generated by performing image analysis on the entire image, and the degree of attention each pixel receives is shown in a heat map. For example, when the second image data of the second image 50 shown in Figure 5 is generated, and this second image data is performed image analysis, a heat map image 60 is generated. The heat map image 60 has pixels that are colored with a corresponding color as the degree of attention increases (blue → light blue → green → yellow → red (black → gray → white in the figure)).

[0042] The area of ​​focus is a region within the second image data that the user wants to draw attention to, and is specified by the user. For example, if the user wants to draw attention to a region where a keyword is displayed in the image, the target area of ​​focus will be the region surrounding that keyword. Specifically, the area of ​​focus 61 will be a rectangle surrounding the keyword "Potato Chips" in the heatmap image 60. The evaluation value of the level of attention for the area of ​​focus 61 is, for example, the ratio of red and yellow pixels [%] to all pixels within the area of ​​focus 61. The target value of the evaluation value for the area of ​​focus is specified, for example, by keyword + a predetermined target value [%] for the evaluation value of the area of ​​focus. The specification of the area of ​​focus is not limited to keywords; it may also be the coordinate information of the area of ​​focus in the second image data.

[0043] In evaluating impressions, an impression evaluation value is calculated based on the image analysis of the second image data. An impression may have multiple impression items. As shown in Figure 6, for example, the impression radar chart 70 may have impression items such as "natural," "handmade," "luxury," "down-to-earth," "fresh," and "homely." The impression evaluation value is expressed as a percentage for the degree of each impression item. However, the type, number, and method of expressing the degree of impression are not limited to these. The target value of the impression evaluation is specified, for example, by the impression item + a predetermined target value [%] for the evaluation value of that impression item. The target value of the impression is not limited to this and may also be the coordinate information of the area of ​​interest within the image.

[0044] The control unit 21 transmits the input concept information, target value and area of ​​interest, or concept information, target value and first image data, to the server 10 via the communication unit 25. The control unit 11 of the server 10 begins receiving the input concept information, target value and area of ​​interest, or concept information, target value and first image data, from the terminal device 20 via the communication unit 15. Triggered by this start of reception, the control unit 11 executes image provision processing according to the image provision program stored in the storage unit 13.

[0045] First, the control unit 11 determines whether or not the reception of the first image data has started (step S11). If the reception of the first image data has started (step S11; YES), the control unit 11 completes the reception of the first image data (step S12). The control unit 11 completes the reception of the concept information and target value (step S13). If the reception of the first image data has not started (step S11; NO), the process proceeds to step S13.

[0046] The control unit 11 generates a prompt including the concept information of step S13, or the concept information of step S13 and the image data of step S12 (step S14). The prompt may include at least a portion of the target value and the number of second image data to be generated (a predetermined number in step S15, described later). The control unit 11 transmits the prompt of step S14, S21, or S22 to the AI ​​providing device 30 for the generation AI model via the communication unit 15 (step S15). The AI ​​providing device 30 receives the prompt of step S14, S21, or S22 from the server 10. The AI ​​providing device 30 inputs the prompt of step S14, S21, or S22 to the generation AI model to generate a predetermined number of second image data. The AI ​​providing device 30 transmits the generated predetermined number of second image data to the server 10. In step S15, the control unit 11 receives the generated predetermined number of second image data from the AI ​​providing device 30 via the communication unit 15.

[0047] The control unit 11 evaluates the area of ​​interest for each generated second image data and evaluates the overall impression (step S16). In step S16, the control unit 11 performs image analysis of the second image data and calculates the evaluation value of the area of ​​interest included in the target value of step S13 as the evaluation result. Specifically, the control unit 11 quantifies the degree of visibility (attention level) for each pixel of the second image data (splendor mapping process). Splendor mapping is an image processing method that represents each pixel in the image with a value (attention level score) indicating the degree of visibility of that pixel. Specifically, in splendor mapping, areas with color contrast in the red-green direction and yellow-blue direction, areas with brightness contrast, and areas with linear components (edges) that coincide with a predetermined direction are represented with high values ​​as areas that are easily visible. The predetermined direction is, for example, the direction from 0 to 315 degrees in 45-degree increments when the angle is taken from 0 to 360 degrees.

[0048] Having a color contrast in the red-green direction corresponds, for example, to the difference in the values ​​indicating the red-green color between adjacent pixels being greater than or equal to a predetermined value. Having a color contrast in the yellow-blue direction corresponds, for example, to the difference in the values ​​indicating the yellow-blue color between adjacent pixels being greater than or equal to a predetermined value. Having a luminance contrast corresponds, for example, to the difference in the values ​​indicating the luminance between adjacent pixels being greater than or equal to a predetermined value. Furthermore, the angles indicating the predetermined direction, 0 degrees and 180 degrees (horizontal direction), 45 degrees and 225 degrees (upward diagonal direction), 90 degrees and 270 degrees (vertical direction), and 135 degrees and 315 degrees (downward diagonal direction), each correspond to the linear component in the same direction. The control unit 11 generates a heatmap image in which the image of the second image data is converted into pixels of the color corresponding to the quantified level of attention. The control unit 11 uses the ratio of red and yellow pixels to all pixels of the area of ​​interest in the heatmap image as the evaluation value of the area of ​​interest.

[0049] Furthermore, in step S16, the control unit 11 performs image analysis of the second image data and calculates an evaluation value for the impression of the image as an evaluation result. Specifically, the control unit 11 performs a color reduction process on the color of each pixel constituting the image of the second image data, grouping similar colors together into the same color. The control unit 11 then determines the proportion (area ratio) that each of the grouped colors (color schemes) occupies in the image. The control unit 11 then calculates the similarity between the color scheme of the second image data and the color scheme of the impression correspondence table (not shown).

[0050] The impression correspondence table associates impression words that describe the impression an image gives with combinations of multiple colors (e.g., three colors) as image features. Examples of impression words include "natural," "handmade," "luxury," "common," "fresh," and "homely." Each color is represented by its respective RGB grayscale value. The impression correspondence table may also include features other than color as image features associated with each impression word. The impression correspondence table is pre-generated and stored in the storage unit 13, for example. The control unit 11 estimates the impression words corresponding to the color scheme patterns as impression items of the second image data, along with an evaluation value (impression score) [%] corresponding to the similarity.

[0051] This section describes the case using an impression correspondence table, but a correlation formula created based on the correspondence between impression words of a sample image and the features of the sample image may also be used. Alternatively, a pre-trained machine learning model may be used. This pre-trained model is trained using the features of multiple sample images and the impression words and impression scores of those sample images, as evaluated by multiple subjects, as training data. When image data is input to this pre-trained model, it outputs the impression words and impression scores corresponding to the image data.

[0052] The control unit 11 selects from a predetermined number of second image data the second image data in which the evaluation result of the area of ​​interest in step S16 and the evaluation result of the impression are closest to the target value (step S17). In step S17, for example, the single second image data is selected in which the sum of the difference (difference) between the evaluation value in the area of ​​interest and the target value and the difference between the evaluation value in the impression and the target value is smallest. However, the selection method is not limited to this.

[0053] The control unit 11 determines whether the difference between the evaluation value and the target value of the area of ​​interest is within a first predetermined value, and whether the difference between the evaluation value and the target value of the impression is within a second predetermined value (step S18). The first predetermined value is the threshold for the allowable difference between the evaluation value and the target value of the area of ​​interest. The second predetermined value is the threshold for the allowable difference between the evaluation value and the target value of the impression. If the difference between the evaluation value and the target value of the area of ​​interest is not within the first predetermined value, and the difference between the evaluation value and the target value of the impression is not within the second predetermined value (step S18; NO), the process proceeds to step S19.

[0054] The control unit 11 determines whether the difference between the evaluation value and the target value of the area of ​​interest is within a first predetermined value (step S19). If it is within the first predetermined value (step S19; YES), the process proceeds to step S20. The control unit 11 sets the area other than the area of ​​interest in the second image data as the modification area (step S20). The modification area is the area that is modified when generating new second image data based on the second image data selected in step S17. The control unit 11 generates a prompt to modify (correct) the set modification area (step S21). The prompt generated in step S21 includes the second image data selected in step S17 and text indicating that the modification area should be changed. The process proceeds to step S15.

[0055] The control unit 11 determines whether the difference between the evaluation value and the target value of the impression is within a second predetermined value (step S22). If it is not within the first predetermined value (step S19; NO), the process proceeds to step S22. If it is within the second predetermined value (step S22; YES), the control unit 11 generates a (correction) prompt regarding the impression (step S23). The prompt generated in step S23 includes text, for example, a message to generate a second image data that increases the evaluation value of the impression item for the target value. The process proceeds to step S15. If it is not within the second predetermined value (step S22; NO), the process proceeds to step S15.

[0056] If the difference between the evaluation value and the target value of the area of ​​interest is within a first predetermined value, and the difference between the evaluation value and the target value of the impression is within a second predetermined value (step S18; YES), the process proceeds to step S24. The control unit 11 transmits the second image data selected in step S17 to the terminal device 20 via the communication unit 15 (step S24). The control unit 21 of the terminal device 20 receives the second image data from the server 10 via the communication unit 25 and displays it on the display unit 24. The image provision process ends.

[0057] Furthermore, the second image data selected in step S17 and transmitted and displayed in step S23 is not limited to one image; it may be multiple images whose evaluation value is closer to the target value. For example, if there are multiple second image data that satisfy the conditions of step S18, these multiple second image data may be selected, transmitted, and displayed.

[0058] Here, with reference to Figures 5 and 6, a specific example of image provision processing will be explained. The user prepares a first image data of a front view of potato chips containing the keyword "Potato Chips". Based on the first image, the user requests a second image data with a design that is appealing to men and women in their teens and twenties. Prior to this, the user inputs the desired concept information, along with the target value of the area of ​​focus and the target value of the impression, into the terminal device 20. The target value of the area of ​​focus is set as: Target value of area of ​​focus: Area of ​​focus of keyword "Potato Chips" + AA [%]. The target value of the impression is set as: Impression item "down-to-earth" 95% + Impression item "fresh" 95%.

[0059] In step S14 of the image provision processing, a prompt is generated. The prompt will be, for example, the text shown in (1) below. (1) This is the product packaging design. Please make it a design that is appealing to men and women in their teens and twenties. It should be down-to-earth and have a handmade feel. Please create 100 images for the second image file.

[0060] In step S15, 100 second image data are generated. In step S16, the evaluation value of the area of ​​interest and the impression evaluation value are calculated for each second image data. In step S17, one second image data of second image 50 is selected. The evaluation value of the area of ​​interest in the second image data of second image 50 is, for example, the ratio of red-yellow pixels (AA-α) [%] in the area of ​​interest 61 of the heatmap image 60. The impression evaluation value in the second image data of second image 50 is, for example, the numerical value [%] of each impression item in the impression radar chart 70.

[0061] In step S18, assume that the first predetermined value is, for example, β(β<α)[%]. In this case, the difference between the evaluation value (AA-α)[%] of the area of ​​interest 61 and the target value AA will not be within the first predetermined value. In step S21, a prompt is generated that includes the second image data of the second image 50 and the text (2) below. (2) Create 100 images of the second image, excluding the area of ​​interest in the second image 50, and modify (correct) them.

[0062] Alternatively, in step S18, suppose the first predetermined value is β (β > α) [%]. In this case, the difference between the evaluation value (AA - α) [%] of the area of ​​interest 61 and the target value AA is within the first predetermined value. Now, suppose the second predetermined value is 5 [%]. In this case, the difference between the evaluation values ​​of the impressions ("down-to-earth" 95%, "fresh" 95%) and the target values ​​("down-to-earth" 94%, "fresh" 46%) is not within the second predetermined value. In steps S18 and S19, for example, the difference in the average values ​​of the impression scores for the impression items "down-to-earth" and "fresh" is compared with the second predetermined value. In step S22, for example, the impression scores for the impression items "down-to-earth" and "fresh" are compared with the target values, and the comparison results are also reflected in the prompt. In this way, the prompt with the text (3) below is generated. (3) Please create a design that is stronger and fresher in impression. Please maintain the current down-to-earth feel. Please create 100 images of this second image data.

[0063] As described above, according to this embodiment, the image generation system 1 includes a control unit 11. The control unit 11 obtains concept information, which includes the image concept desired by the user, and target values ​​for the evaluation of the image data, from the terminal device 20. Based on prompts corresponding to the acquired concept information, the control unit 11 causes the generation AI model to generate a plurality of second image data. The control unit 11 evaluates each of the generated image data. The control unit 11 selects one second image data whose evaluated value is closest to the target value. Therefore, it is possible to easily generate a second image data desired by the user from the image concept.

[0064] The control unit 11 transmits the selected second image data to the display unit 24 for display. Therefore, the user can visually confirm the generated second image data.

[0065] The control unit 11 acquires the first image data, concept information, and target values ​​of the draft image. Based on prompts corresponding to the acquired first image data and concept information, the control unit 11 causes the generation AI model to generate multiple second image data. Therefore, the second image data desired by the user can be easily generated from the first image data and the concept of the image data.

[0066] The control unit 11 generates a new prompt to bring the selected second image data, evaluation value, and target value closer to the target value. The control unit 11 inputs the new correction prompt into the generation AI model. This allows the user to easily and accurately generate the second image data they desire.

[0067] The control unit 11 repeatedly generates a new prompt and a second image data based on the new prompt until the evaluation value falls within a predetermined range from the target value. This allows for the easy and accurate generation of the second image data desired by the user.

[0068] The evaluation involves assessing the impression of the generated second image data and the degree of attention given to the area of ​​focus. The control unit 11 generates a new prompt so that the impression evaluation value falls within a second predetermined value from the target value. Therefore, the second image data with the impression desired by the user can be easily generated from the new prompt.

[0069] The control unit 11 generates a new prompt to change areas other than the area of ​​interest if the evaluation value of the area of ​​interest is within a first predetermined value. Therefore, a second image data of the area of ​​interest desired by the user can be easily generated from the new prompt.

[0070] The image generation system 1 includes an AI providing device 30 that provides a generation AI model. This reduces the processing load on the server 10.

[0071] The above description discloses an example in which a storage unit (HDD, SSD) is used as a computer-readable medium for the program according to the present invention, but the invention is not limited to this example. Other computer-readable mediums that can be used include non-volatile memory such as flash memory and portable recording media such as CD-ROMs. Furthermore, carrier waves can also be used as a medium for providing program data according to the present invention via a communication line.

[0072] The above-described embodiments are merely examples of the image generation system, image generation method, and program according to the present invention, and are not limited thereto. For example, the first and second embodiments may be combined as appropriate.

[0073] Furthermore, although the above embodiment describes an AI provisioning device 30 that provides generated AI, the configuration is not limited to this. The server 10 itself may also have a function to provide generated AI.

[0074] Furthermore, in the above embodiment, the number of second image data files to be generated in step S15 was a predetermined number, but it is not limited to this. For example, when the control unit 21 receives concept information input from the user via the operation unit 22, it may also receive input for setting concept information, including the number of image data files to be generated. The number of image data files to be generated input by the terminal device 20 is transmitted to the server 10. The control unit 11 of the server 10 sets the predetermined number of received image data files to the predetermined number in step S15 of the image provision process. In other words, the predetermined number of image data files is the number of files input by the user via the operation unit 22. Therefore, the user can specify the number of image data files to be generated. By increasing the predetermined number, the completeness of the second image data for the desired concept can be improved. Alternatively, by decreasing the predetermined number, the time required to generate the second image data can be shortened.

[0075] While embodiments of the present invention have been described and illustrated in detail, the disclosed embodiments are for illustrative purposes only and are not limiting. The scope of the present invention should be interpreted by the terms of the appended claims. [Explanation of symbols]

[0076] 1. Image generation system 10 servers 11 Control Unit 12 Control section 13 Storage section 14 Display section 15 Communications Department 20 Terminal devices 21 Control Unit 22 Control section 23 Memory section 24 Display section 25 Communications Department 30 AI providing device 40 Communication Networks

Claims

1. An acquisition unit that acquires concept information including the image concept desired by the user and a target value in the evaluation value of the image data, A generation control unit that causes a generation AI model to generate multiple image data based on prompts corresponding to the acquired concept information, An evaluation unit that evaluates each of the generated image data, An image generation system comprising a selection unit that selects at least one image data whose evaluated value is closer to the target value.

2. The image generation system according to claim 1, further comprising a display control unit that displays the selected image data on a display unit.

3. The acquisition unit acquires draft image data, concept information, and target values. The image generation system according to claim 1, wherein the generation control unit causes the generation AI model to generate a plurality of image data based on prompts corresponding to the acquired draft image data and concept information.

4. The image generation system according to claim 1, wherein the generation control unit generates a correction prompt to bring the evaluation value closer to the target value based on the selected image data, the evaluation value, and the target value, and inputs the correction prompt to the generation AI model.

5. The image generation system according to claim 4, wherein the generation control unit repeats generating the correction prompt and generating image data based on the correction prompt until the evaluation value falls within a predetermined threshold from the target value.

6. The image generation system according to claim 1, wherein the evaluation is an evaluation of at least one of the impression of the generated image data and the region of interest.

7. The aforementioned evaluation is an evaluation of the impression of the generated image data. The image generation system according to claim 4, wherein the generation control unit generates the correction prompt so that the evaluation value of the impression falls within a predetermined threshold from the target value.

8. The aforementioned evaluation is an evaluation of the region of interest in the generated image data. The image generation system according to claim 4, wherein the generation control unit generates a correction prompt to modify areas other than the area of ​​interest when the evaluation value of the level of attention of the area of ​​interest is within a predetermined threshold.

9. The image generation system according to claim 1, further comprising a providing unit for providing the generated AI model.

10. An acquisition process to obtain concept information including the image concept desired by the user and a target value in the evaluation value of the image data, A generation control step in which the generation AI model generates multiple image data based on prompts corresponding to the acquired concept information, An evaluation step for evaluating each of the generated image data, An image generation method comprising a selection step of selecting at least one image data whose evaluated value is closer to the target value.

11. Computers An acquisition unit that acquires concept information including the image concept desired by the user and a target value in the evaluation value of the image data. Based on prompts corresponding to the acquired concept information, a generation control unit causes the generation AI model to generate multiple image data. An evaluation unit that evaluates the generated image data, A selection unit that selects at least one image data whose evaluated value is closer to the target value. A program designed to function as such.

Images