Information processing device, information processing method, and program

Abstract

Generates production data that brings together the impressions of multiple productions more closely. [Solution] The information processing device according to the present disclosure is an information processing device that generates production data, and comprises a generation means for generating a second production data which has background information for a first production data and is at least different in size from the first production data, wherein the generation means generates the background information for the second production data by enlarging or reducing the background information for the first production data so that the size ratio of the background information for the first production data to the background information for the second production data is a value between the ratio of the size of the first production data to the size of the second production data and 1.

Description

Technical Field

[0001] The present invention relates to an information processing apparatus, an information processing method, and a program.

Background Art

[0002] Conventionally, a method has been proposed in which a template for storing information such as the shape and arrangement of images, characters, graphics, etc. that make up a poster is prepared, and the information processing apparatus arranges the images, characters, graphics, etc. according to the template to generate a poster.

[0003] In Patent Document 1, a system for generating a postcard is described in which templates are selected in ascending order of the difference between the impression evaluation value of the template and the impression evaluation value of the image to be arranged on the template.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

[0007] The information processing device relating to this disclosure is an information processing device that generates production data, and comprises a generation means for generating a second production data which has background information for a first production data and is at least different in size from the first production data, wherein the generation means generates the background information for the second production data by enlarging or reducing the background information for the first production data such that the size ratio of the background information for the first production data to the background information for the second production data is a value between the ratio of the size of the first production data to the size of the second production data and 1. [Effects of the Invention]

[0008] According to this disclosure, it is possible to generate production data that brings together the impressions of multiple productions more closely. [Brief explanation of the drawing]

[0009] [Figure 1] This is a block diagram showing the hardware configuration of the product generation device. [Figure 2] This is a software block diagram for a product creation application. [Figure 3] This is a diagram explaining the skeleton. [Figure 4] This is a diagram explaining the background design. [Figure 5] This is a diagram illustrating color scheme patterns. [Figure 6] This is a diagram showing the app launch screen provided by the product creation app. [Figure 7] This diagram shows the preview screen provided by the product creation app. [Figure 8] This is a flowchart showing the process of quantifying impressions. [Figure 9] This is a diagram illustrating subjective evaluation of impressions. [Figure 10] This is a flowchart showing the product generation process. [Figure 11] This is a diagram illustrating the color scheme list. [Figure 12] It is a diagram for explaining the acquisition of a color scheme pattern. [Figure 13] It is a diagram for explaining the subjective evaluation of color schemes. [Figure 14] It is a diagram for explaining the method of selecting a skeleton. [Figure 15] It is a diagram for explaining the method of selecting a color scheme pattern. [Figure 16] It is a diagram for explaining the scaling of the background pattern. [Figure 17] It is a diagram for explaining the complementation and adjustment of the background pattern. [Figure 18] It is a diagram for explaining the method of selecting a font. [Figure 19] It is a software block diagram for explaining the layout section in detail. [Figure 20] It is a flowchart showing the layout process. [Figure 21] It is a diagram for explaining the input of the layout section. [Figure 22] It is a diagram for explaining the operation of the layout section. [Figure 23] It is a diagram showing a modified example of the preview screen provided by the merchandise creation app. [Figure 24] It is a software block diagram of the merchandise creation app according to the second embodiment. [Figure 25] It is a flowchart showing the merchandise generation process according to the second embodiment. [Figure 26] It is a diagram for explaining the combination generation section. [Figure 27] It is a diagram for explaining the combination generation section.

Mode for Carrying Out the Invention

[0010] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. Note that the following embodiments do not limit the present disclosure according to the claims, and not all combinations of the features described in the present embodiments are essential for the solution means of the present disclosure. The same reference numerals are assigned to the same components, and the description thereof is omitted.

[0011] In the embodiments described below, a method is described for generating automatically designed product data by running an application for creating products such as merchandise (hereinafter also referred to as the "product creation application") in an information processing device. In this specification, the product data is also referred to as "product data."

[0012] In this specification, "merchandise" includes posters, brochures, menus, postcards, flyers, banners, business cards, shop cards, invitations, membership cards, and other creative works used, for example, as advertising media. It also includes any creative works that include at least one of either image content or text content. Furthermore, the data of the creative works (merchandise data) may be output as print data and used not only by being printed, but also by being used as electronic content on websites, social networking services, virtual spaces, etc.

[0013] Furthermore, in this specification, "brand" refers to the design expression of a company's or store's identity (corporate philosophy, vision, code of conduct, and characteristics). In order for a brand to be widely recognized by customers, it is necessary to convey a message to customers by deploying a consistent worldview in the design across various products. In other words, what is important for people to recognize a brand is a unified design with a consistent worldview. To achieve this, it is necessary to use a consistent design across customer touchpoints such as product packaging, store design, and promotional materials (website, brochures, posters, business cards, postcards, etc.). To create a unified design, it is necessary to include similar design elements across multiple products. Design elements are the elements that make up a design, and examples include the brand's logo (symbol), font, image, and color. When these design elements are used commonly across multiple products, a sense of unity is created, and customers can recognize a consistent worldview.

[0014] The following embodiments describe a product generation device that generates product data having a consistent design across multiple products of different sizes or types, and that expresses the brand impression intended by the user.

[0015] <<First Embodiment>> In the first embodiment, a method for generating product data for multiple different products by running an application (hereinafter also referred to as the "product creation application") in an information processing device will be described. In the following description, unless otherwise specified, "image" includes still images taken by a camera, frame images extracted from videos, and illustrations created with paint tools, etc. Furthermore, an information processing device equipped with a product creation application will be referred to as a product generation device.

[0016] Figure 1 is a block diagram showing the hardware configuration of the product generation device 100. The product generation device 100 is an information processing device, and examples include personal computers (hereinafter referred to as PCs), smartphones, tablets, etc. In this embodiment, the product generation device 100 will be described as a PC. The product generation device 100 has a CPU 101, ROM 102, RAM 103, HDD 104, display 105, keyboard 106, pointing device 107, data communication unit 108, and GPU 109.

[0017] The CPU (Central Processing Unit / Processor) 101 comprehensively controls the product generation device 100 and, for example, reads a program stored in the ROM 102 into the RAM 103 and executes it, thereby realizing the operation of this embodiment. In Figure 1, there is one CPU, but it may be composed of multiple CPUs.

[0018] ROM102 is a general-purpose ROM that stores, for example, programs executed by CPU101. RAM103 is a general-purpose RAM that is used as working memory to temporarily store various information when a program is executed by CPU101.

[0019] The HDD (hard disk drive) 104 is a storage medium (storage unit) for storing image files, a database that holds processing results such as image analysis, and skeletons used by product creation applications.

[0020] The display 105 is a display unit that displays the user interface (UI) of this embodiment, as well as product data, which is the result of the layout of image data (hereinafter also referred to as "images") and text, to the user. The keyboard 106 and pointing device 107 receive instructions and operations from the user. The display 105 may also have a touch sensor function.

[0021] Keyboard 106 is used, for example, when a user inputs the conditions for creating a product they want to create on the UI displayed on display 105.

[0022] The pointing device 107 is used, for example, when a user clicks a button on the UI displayed on the display 105.

[0023] The data communication unit 108 communicates with external devices via a wired or wireless network. For example, the data communication unit 108 transmits the laid-out data, generated by the automatic layout function, to a printer or server that can communicate with the product generation device 100.

[0024] GPU109 is a processor that performs image processing in response to instructions from CPU101. For example, GPU109 analyzes images placed on merchandise, estimates the impression of images or text, estimates the impression of each design element, estimates the impression of the merchandise, lays out images and text on a skeleton, and applies color schemes to generate merchandise data.

[0025] The data bus 110 connects each block in Figure 1 so that they can communicate with each other. Note that the configuration shown in Figure 1 is merely an example and is not limited to this. For example, the product generation device 100 does not have a display 105, and the UI may be displayed on an external display.

[0026] In this embodiment, the product creation application is stored on the HDD 104. The product creation application is launched when the user clicks or double-clicks the application icon displayed on the display 105 using the pointing device 107.

[0027] <Software Block Diagram> Figure 2 is a software block diagram of the product creation application. The product creation application includes a creation condition specification unit 201, a text specification unit 202, an image specification unit 203, a design element specification unit 204, a key design specification unit 205, a generated product display unit 206, and a product generation unit 210. The product generation unit 210 includes an image acquisition unit 211, an image analysis unit 212, a skeleton acquisition unit 213, a design element acquisition unit 214, a color scheme pattern acquisition unit 215, a skeleton selection unit 216, a color scheme pattern selection unit 217, a background pattern scaling unit 218, a logo selection unit 219, a font selection unit 220, a layout unit 221, an impression estimation unit 222, and a product selection unit 223.

[0028] When the product creation application is installed on the product generation device 100, a launch icon appears on the top screen (desktop) of the OS (operating system) running on the product generation device 100. The user operates the launch icon displayed on the display 105 using the pointing device 107. In response to this operation, the product creation application program stored on the HDD 104 is loaded into the RAM 103 and executed by the CPU 101. This starts the product creation application.

[0029] Each program module corresponding to each component shown in Figure 2 is included in the aforementioned product creation application. The CPU 101 then executes each program module, causing the CPU 101 to function as the respective parts shown in Figure 2. Hereafter, each part shown in Figure 2 will be described as performing various processes according to the program. Figure 2 also specifically shows a software block diagram related to the product generation unit 210, which performs the function of automatically creating multiple products.

[0030] The creation condition specification unit 201 specifies the creation conditions for the merchandise to the merchandise generation unit 210 in response to UI operations by the pointing device 107. In this embodiment, the creation conditions specify the types of merchandise to be created, the usage categories, and the assumed observation distance for each merchandise. The size of the merchandise may be pre-set in association with the type of merchandise, or it may be specified by the user. Furthermore, for some merchandise, it may be possible to specify multiple sizes. In that case, the user may specify the actual dimensions of the width and height, or specify paper sizes such as A1 or A2. The usage category is a category that indicates what kind of use the merchandise will be used for, for example, restaurants, school events, sales, etc. The creation condition specification unit 201 outputs the specified creation conditions to the skeleton acquisition unit 213, the color scheme pattern acquisition unit 215, the design element acquisition unit 214, and the background pattern scaling unit 218.

[0031] The text specification unit 202 accepts user specification of text information to be placed on the product via UI operation using the keyboard 106. Text information to be placed on the product is, for example, a string of characters representing the title, date and time, and location if the product is a poster. Text information is associated with a type of text information. The type of text information is information that indicates the type (tag or attribute information), such as whether the text information is a title or information indicating the date and time or location, and is associated with each piece of text information. The type of text information may change depending on the type of product to be created selected in the creation condition specification unit 201. For example, if a poster is selected, the types of text information will be "title," "subtitle," and "body text." If a postcard is selected, the types of text information will be "title," "address," and "contact information." If multiple products are selected as the target of creation, text information that is duplicated across multiple products may be specified separately for each product or combined into one specification. The text specification unit 202 associates each character information with the type of character information and then outputs it to the skeleton acquisition unit 213 and the layout unit 221.

[0032] The image specification unit 203 accepts user specifications for one or more image data to be placed on the product. Image data can be specified based on the structure of the file system containing the image data, such as the device and directory. Image data may also be specified by supplementary information to identify the image, such as the date and time of shooting, or by attribute information. The image specification unit 203 may also specify image data included in and provided as material by the poster creation application (hereinafter also referred to as "application material images"). The image specification unit 203 may also specify image data held by an external image provision service with which the poster creation application is linked (hereinafter also referred to as "linked material images"). The image specification unit 203 outputs the file paths of the specified image and the generated image to the image acquisition unit 211.

[0033] The design element specification unit 204 accepts from the user the specification of design elements to be reflected in the product to be generated (hereinafter referred to as "created product"). The design elements to be specified include at least a target impression and further include at least one of the following: color scheme, background information (hereinafter referred to as "background pattern"), logo, and font. In this embodiment, the target impression, color scheme, background pattern, logo, and font can be specified. Furthermore, the design element specification unit 204 accepts the specification of the degree of design reflection. The degree of design reflection is an indicator that shows how much the specified design elements are reflected in the design of the created product. For color scheme, background pattern, logo, and font, one or more candidate color schemes, background patterns, logos, and fonts are specified in the product generation process described later. The target impression is the impression that the created product is ultimately required to maintain and is the impression that can be given to people who view the created product. In this embodiment, the intensity of the impression to be conveyed is specified by UI operation using the pointing device 107. For example, "luxury," "approachability," "dynamism," and "solidity" are specified as impression factors, respectively. The design element specification unit 204 outputs the specified design element information to the design element acquisition unit 214. Details about impressions will be described later.

[0034] Furthermore, the design element specification section 204 does not necessarily have to accept user specifications. It is sufficient for the specification to be enabled only when the user wishes to specify a design element. Also, the degree of design reflection does not necessarily have to be specified by the user. If the design elements and the degree of design reflection are specified by the user, the user can control the design of the created product.

[0035] The key design specification unit 205 accepts the specification of a key design to be reflected in the generated product. A key design is a design that serves as the basis for the created product and is a design that is commonly used across multiple created products. If the key design is a design of a product that has been created in advance, it may be product data created with a product creation application before the product creation process starts, or design data created with another design creation application. The input format of the key design data file may be raster data such as JPEG or BMP, or vector data containing drawing commands. For example, common PDLs such as PDF (Portable Document Format) proposed by Adobe, XPS proposed by Microsoft, or HP-GL / 2 proposed by HP may be used. One key design may be specified, or multiple key designs may be specified. The key design specification unit 205 outputs the file path of the specified key design to the design element acquisition unit 214.

[0036] Next, the configuration of the product generation unit 210 will be explained in detail. As a prerequisite, differences in the types of products to be created can be achieved by selecting the type of skeleton corresponding to the product to be created in the skeleton acquisition unit 213.

[0037] The image acquisition unit 211 acquires one or more image data specified by the image selection unit 203 from the HDD 104. The image acquisition unit 211 outputs the acquired image data to the image analysis unit 212. It also outputs the number of acquired images to the skeleton acquisition unit 213. Images stored in the HDD 104 include still images and frame images extracted from videos. Still images and frame images are acquired from imaging devices such as digital cameras and smartphones. The imaging device may be provided by the product generation device 100 or it may be an external device. If the imaging device is an external device, the images are acquired via the data communication unit 108. As another example, still images may be illustrations created with image editing software or CG images created with CG production software. Still images and extracted images may be images acquired from a network or server via the data communication unit 108. An example of an image acquired from a network or server is a social networking service image (hereinafter referred to as "SNS image"). Furthermore, the program executed by CPU 101 analyzes the data associated with each image to determine its source. For example, for SNS images, the application may manage the source by obtaining the image from the SNS via the application. Note that the images are not limited to those described above, but may be of other types.

[0038] The image analysis unit 212 performs image data analysis on the image data acquired from the image acquisition unit 211 and obtains information indicating image features. Specifically, the image analysis unit 212 performs object recognition processing and major color extraction processing, which will be described later, to obtain information indicating image features of the image data. The image analysis unit 212 also associates the information indicating image features obtained from the image data with the image data and outputs it to the layout unit 221.

[0039] The skeleton acquisition unit 213 acquires one or more skeletons from the HDD 104 that meet the conditions specified by the creation condition specification unit 201, the text specification unit 202, the design element acquisition unit 214, and the image acquisition unit 211. In this embodiment, a skeleton is information that indicates the arrangement of strings, images, shapes (graphics), etc. to be placed on the product.

[0040] Figure 3 shows an example of a poster skeleton among various merchandise. On the skeleton 301 in Figure 3(a), three graphic objects 302, 303, and 304, one image object 305, and four text objects 306, 307, 308, and 309, which are objects on which text will be placed, are arranged. Each object is associated with metadata necessary for generating the poster, in addition to its position, size, and angle, which indicate where it will be placed. Figure 3(b) shows an example of metadata. For example, text objects 306 to 309 hold metadata attributes indicating what kind of text information will be placed there. Here, text object 306 indicates the title, text object 307 indicates the subtitle, and text objects 307 and 308 indicate the body text. Graphic objects 302 to 304 also hold metadata attributes indicating the shape of the graphic and a color scheme number (color scheme ID) indicating the color scheme pattern. Here, the attributes of graphic objects 302 and 303 indicate a rectangle, and the attribute of graphic object 304 indicates an ellipse. Furthermore, graphic object 302 is assigned color scheme number 1, and graphic objects 303 and 304 are assigned color scheme number 2. Here, the color scheme number is information referenced in the color scheme described later, and different color scheme numbers are assigned different colors. Graphic objects may also be drawn by being painted with a uniform color. Graphic objects may also have background images or background illustrations cut out in the shape of the graphic object drawn on them. Note that the types of objects and metadata are not limited to these. For example, there may be map objects for placing maps, or barcode objects for placing QR codes (registered trademarks) or barcodes. Also, metadata for text objects may include metadata that represents the line spacing and character spacing. The metadata may include the purpose of the skeleton and may be used to control whether the skeleton can be used or not depending on the purpose.

[0041] Skeletons may be classified and managed according to the type of product. For example, there may be skeletons for posters, menus, postcards, tri-fold leaflets, calendars, banners, etc. Furthermore, skeletons may be managed as groups consisting of multiple classified skeletons based on relationships such as the arrangement of objects. For example, one could create a group of skeletons in Group 1 that includes skeletons configured to convey a sense of luxury, and keep the same skeleton group ID in the metadata of each skeleton. This allows the group ID of the skeleton used to create one type of product to determine which skeleton should be applied to other types of products. As a result, when creating data for multiple products, the design can be unified across those products.

[0042] Furthermore, the skeletons may be classified and managed according to their width-to-height ratio. This makes it possible to obtain a skeleton with a width-to-height ratio that matches the size of the product specified in the creation condition specification unit 201.

[0043] The skeleton may be saved on HDD104 in CSV format, for example, or in a database format such as SQL. The skeleton acquisition unit 213 outputs one or more skeletons acquired from HDD104 to the skeleton selection unit 216.

[0044] The design element acquisition unit 214 acquires design elements to be used for design generation from the key design specified by the key design specification unit 205. The design element acquisition unit 214 extracts the color scheme, background pattern, logo, font, and impression value as design elements from the key design file specified by the key design specification unit 205. The extracted impression value is used as the target impression of the created product.

[0045] When extracting impression values ​​from a key design, the design element acquisition unit 214 estimates the impression of the product data, which is the key design, by the impression estimation process described later. Alternatively, if the key design is product data previously created in the product creation application, the impression estimated from that product data at the time of creation is linked to the product data and stored, and that impression value is acquired. Other methods for extracting design elements will be described later.

[0046] Furthermore, the design element acquisition unit 214 acquires design elements based on the design element information specified by the user in the design element specification unit 204 and combines them with the design elements acquired from the key design (hereinafter referred to as "merging"). That is, the design element acquisition unit 214 merges each design element extracted from the product data file designated as the key design with each design element specified in the design element specification unit 204, element by element. Note that if the design element specification is set to invalid, or if a design element is not specified by the user, it may not be merged, and only the design elements extracted from the key design may be used in the created product.

[0047] For example, when merging target impressions, the design element acquisition unit 214 uses, for example, the average value of each impression factor between the impression values ​​extracted from the key design and the target impression specified by the design element specification unit 204 as the target impression value after merging. Alternatively, instead of the average value, a representative value of multiple values, such as the maximum or minimum value, may be used as the target impression value after merging.

[0048] When the user specifies the degree of design reflection in the design element specification unit 204, the design element acquisition unit 214 merges the design elements extracted from the key design according to the degree of design reflection with the design elements specified by the user. This allows for a more accurate reflection of the user's intentions. Specifically, each element of the design elements is merged using the following formula (1). Note that formula (1) represents the target impression. In formula (1), the value of the degree of design reflection is converted to a value range of 0 to 1. The specified target impression value is the target impression received from the user in the design element specification unit 204. The extracted target impression value is the impression value extracted from the key design.

[0049] Target impression value after merging = Reflection degree × Specified target impression value + (1 - Reflection degree) × Extracted target impression value ... (1)

[0050] Furthermore, the design element acquisition unit 214 creates a color scheme list by merging the color scheme extracted from the key design with the color scheme specified in the design element specification unit 204, and outputs it to the color scheme pattern acquisition unit 215. The design element acquisition unit 214 creates a logo list by merging the logo extracted from the key design with the logo specified in the design element specification unit 204, and outputs it to the logo selection unit 219. The design element acquisition unit 214 creates a font list by merging the font extracted from the key design with the font specified in the design element specification unit 204, and outputs it to the font selection unit 220. In these lists, the color scheme, logo, and font specified in the design element specification unit 204 are kept separate from the color scheme, logo, and font extracted from the key design.

[0051] The design element acquisition unit 214 outputs the logo list to the skeleton acquisition unit 213. The design element acquisition unit 214 outputs the merged target impression value and the degree of reflection specified in the design element specification unit 204 to the skeleton selection unit 216, the color scheme pattern selection unit 217, the background pattern scaling unit 218, the logo selection unit 219, the font selection unit 220, and the product selection unit 223.

[0052] Furthermore, regarding the background image, if a background image is specified in the design element specification unit 204, the design element acquisition unit 214 outputs the specified background image to the background image scaling unit 218. If no background image is specified in the design element specification unit 204, the background image extracted from the key design is output to the background image scaling unit 218. For example, if the file specified as the key design is raster data, it may be difficult to accurately extract the entire background image of the key design. In this case, the user specifies the SVG data or raster data that was the source of the background image of the key design as the background image in the design element specification unit 204. This allows the design element acquisition unit 214 to acquire the correct background image.

[0053] Figure 4 shows examples of background patterns. Figures 4(a), 4(b), and 4(c) represent background patterns with repeating designs. Figures 4(d), 4(e), and 4(f) show cases where the pattern or a part of the design includes a repeating design. Thus, in this embodiment, a background pattern refers to something used as the base for a design and containing any pattern.

[0054] The color scheme pattern acquisition unit 215 acquires a list of the main colors of an image from the image analysis unit 212 and a color scheme list from the design element acquisition unit 214. If the number of additional colors is specified in the creation condition specification unit 201, it acquires the specified number of additional colors. Furthermore, the color scheme pattern acquisition unit 215 acquires a color scheme pattern based on the acquired main color list, color scheme list, and number of additional colors of the image. A color scheme pattern is a combination of colors used in the product. Additionally, if a color is specified in the creation condition specification unit 201, the color scheme pattern acquisition unit 215 also acquires a color scheme pattern containing that specified color from the HDD 104 and outputs it to the color scheme pattern selection unit 217.

[0055] Figure 5 shows an example of a table showing a color scheme pattern. In this embodiment, the color scheme pattern is represented as a combination of four colors. The Color ID column in Figure 5 is an ID for uniquely identifying the color scheme pattern. The Color 1 to Color 4 columns store the color values. The color values ​​are represented by the RGB color values ​​in that order from 0 to 255 ((R, G, B) = (0 to 255, 0 to 255, 0 to 255)). In this embodiment, a color scheme pattern consisting of a combination of four colors is used, but other numbers of colors may be used, or multiple numbers of colors may be mixed.

[0056] The skeleton selection unit 216 selects a skeleton from the skeletons acquired from the skeleton acquisition unit 213 that matches the type of product specified in the creation condition specification unit 201 and matches the target impression merged by the design element acquisition unit 214. The skeleton selection unit 216 outputs the selected skeleton to the layout unit 221. The selected skeleton satisfies the following conditions: one or more skeletons are selected for each type of product, and for each type of product, one or more skeletons are selected that match the target impression. Since the overall arrangement of each product is determined by the skeleton, the variations of each product after generation can be increased by preparing various types of skeletons in advance.

[0057] The color scheme pattern selection unit 217 selects one or more color schemes that match the target impression merged by the design element acquisition unit 214 from the color scheme patterns acquired by the color scheme pattern acquisition unit 215, and outputs them to the layout unit 221.

[0058] The background pattern scaling unit 218 scales the background pattern acquired by the design element acquisition unit 214 by different scaling ratios depending on the size or type of the product to be created, as specified by the creation condition specification unit 201. In the following explanation, scaling refers to increasing or decreasing the size. The background pattern scaling unit 218 determines the size ratio R between the background pattern (background information) of the key design (first product data) to be used as the source and the background pattern (background information) of the product to be used as the destination. Then, it generates the background pattern (background information) of the product to be created by scaling the background pattern (background information) of the key design by the determined size ratio R. The background pattern scaling unit 218 sets the size ratio R of the background pattern to the ratio C of the size C1 of the key design and the size C2 of the product to be created. 12 The value is determined to be between 1 and 2. The size ratio between the background image of the key design and the background image of the product to which it will be used is also referred to as the scaling ratio R in the following explanation. The scaling ratio R is expressed by the following relational equation (2).

[0059]

number

[0060] C1 is the length of the diagonal of the first production data, which will be the source for the expansion, and C2 is the length of the diagonal of the second production data, which will be the destination for the expansion. 12 This is not limited to the ratio of diagonal lengths, but can be the ratio of areas, the ratio of the square root of the area ratio, the ratio of the lengths of corresponding sides between the data of the first and second creations, or any other value that represents the ratio of the size of the data of the first creation to the size of the data of the second creation.

[0061] α is a coefficient for determining the scaling ratio R, and the following method of determination is possible. That is, the coefficient α is determined based on the ratio of the distance at which the first production data is observed to the distance at which the second production data is observed. Equation (3) below is the equation for determining the coefficient α.

[0062]

number

[0063] Here, D1 is the assumed observation distance of the first production data, which is the source for the display, and D2 is the assumed observation distance of the second production data, which is the destination for the display. Observation distance is the distance between the person viewing the product and the product itself. For example, posters are often viewed from a distance, while business cards are often viewed at close range. Objects with a large observation distance, i.e., those that are far away, appear smaller in inverse proportion to that distance. Assumed observation distance is the expected observation distance, which is either pre-set for each product or specified by the user.

[0064] Alternatively, the coefficient α is determined based on the ratio of the font size of the representative text used in the first production data to the font size of the representative text used in the second production data. In this case, the coefficient α is determined based on the following equation (4). In equation (4), F1 is the size of the representative text in the first production data from which the data is derived, and F2 is the size of the representative text in the second production data from which the data is derived. The representative text may be the text whose attribute is specified as a title, or it may be the largest font size in the product.

[0065]

number

[0066] Alternatively, the coefficient α may be determined to approximate the impression of the key design. In this case, the background image scaling unit 218 scales the background image by changing the value of α to various values, creates a product layout using the scaled background image through a layout process described later, and estimates the impression of the completed product image. The background image scaling unit 218 uses an α that is close to the impression value of the key design in calculating the scaling ratio R. As mentioned above, the value of α is 1 < 1 / C 12 In this case, 1 < α < 1 / C 12 Let it be one of the values, and 1 / C 12 If <1, then 1 / C 12The value of α should be one of the following: < 1. Furthermore, a smaller impression distance between the estimated impression value and the key design impression value indicates a closer relationship between the two. For example, when determining a single value for α, the α with the smallest impression distance is selected. When determining multiple values ​​for α, N α values ​​are selected in order of increasing impression distance.

[0067] The background image scaling unit 218 outputs the enlarged or reduced background image to the layout unit 221, linking it to the size or type of the product being created.

[0068] The method for determining the coefficient α is not limited to the method described above; any value is acceptable as long as the scaling ratio is determined to be between the ratio of the size of the first production data and the size of the second production data and 1. The scaling of the background image will be discussed later.

[0069] The logo selection unit 219 selects one or more logos from the logo list merged by the design element acquisition unit 214 that match the target impression merged by the design element acquisition unit 214, and outputs them to the layout unit 221.

[0070] The font selection unit 220 selects one or more font patterns from the fonts merged by the design element acquisition unit 214 that match the target impression merged by the design element acquisition unit 214, and outputs them to the layout unit 221. A font pattern is a combination of fonts that includes at least two of the following: the title font, the subtitle font, and the body text font.

[0071] The layout unit 221 combines various data to lay out one or more skeletons obtained from the skeleton selection unit 216. As a result, one or more types of product data specified in the creation condition specification unit 201 are created in a predetermined number or greater. The number of creations may be a value specified by the user in the creation condition specification unit 201, or it may be a value set in advance.

[0072] The layout unit 221 places text obtained from the text specification unit 202 and image data obtained from the image analysis unit 212 onto each skeleton. It then applies a color scheme pattern obtained from the color scheme pattern selection unit 217 and a font pattern selected from the font selection unit 220. Furthermore, the layout unit 221 places a background image obtained from the background image scaling unit 218, scaled according to the size or type of the created product, into the background area of ​​each skeleton. The background area is set on the skeleton as a graphic object, for example. Furthermore, the layout unit 221 places a logo selected from the logo selection unit 219 into the logo area of ​​each skeleton. The logo area is set on the skeleton as a graphic object or text object. The layout unit 221 outputs the generated one or more product data to the impression estimation unit 222.

[0073] The impression estimation unit 222 estimates an impression for each of the product images rendered from multiple product data acquired from the layout unit 221, and associates the estimated impression (estimated impression) with each product data. The impression estimation unit 222 then outputs one or more product data associated with the estimated impression to the product selection unit 223.

[0074] The product selection unit 223 compares the target impression merged by the design element acquisition unit 214 with the estimated impressions of multiple product data linked to estimated impressions, obtained from the impression estimation unit 222, and selects the products to be displayed on the display 105 as the creation result. The selected products are saved on the HDD 104. The product selection unit 223 outputs the selected product data to the generated product display unit 206.

[0075] The product selection unit 223 selects created products in which the distance (impression distance) between the estimated impression associated with the created product data and the target impression is smaller than a predetermined threshold. Furthermore, if the created products are a product set containing multiple products of different types, the product selection unit 223 selects a product set in which the sum of the distances (total impression distance) between the estimated impressions associated with the multiple product data of different types and the target impression is smaller than a predetermined threshold. The value representing the impression is represented by a vector, and the distance is represented by the distance between vectors. In other words, the smaller the distance between the impression estimated from the created product and the target impression, the closer the created product is to the target impression.

[0076] The generated product display unit 206 renders the product data acquired from the product selection unit 223 and outputs a product image for display on the display 105. The product image is, for example, bitmap data. The generated product display unit 206 displays the product image on the display 105.

[0077] The product creation application may also have a function (not shown) that, after displaying the generated product results on the generated product display unit 206, allows the user to edit the arrangement, color, shape, etc. of images, text, and graphics through additional operations to further change the design to the user's desired specifications.

[0078] Furthermore, if the system includes a function to print product data stored on the HDD104 using a printer according to the conditions specified in the creation condition specification unit 201, the user will be able to obtain printed copies of the products they have created.

[0079] <Example of display screen> Figure 6 shows an example of the app launch screen 601 provided by the product creation app. The app launch screen 601 is displayed on the display 105. The user specifies the key design, product creation conditions, content (text and images), and design elements via the app launch screen 601. The creation condition specification unit 201, image specification unit 203, text specification unit 202, design element specification unit 204, and key design specification unit 205 obtain the specifications from the user through this UI screen.

[0080] The content specification area 600 includes a title box 602, a subtitle box 603, a body text box 604, and an image specification area 605. The title box 602, subtitle box 603, and body text box 604 accept the specification of text information to be placed on the product. In this embodiment, three types of text information are accepted, but this is not limited to these. For example, additional text information such as location and date and time may be accepted. Also, it is not necessary to specify text information for all boxes, and some boxes may be left blank. The display and display content of the boxes may be configured to change according to the specification result in the product specification area 612. For example, if a poster is selected, boxes for specifying the title, subtitle, and body text will be displayed. If it is a postcard, boxes for specifying the title, address, and contact information will be displayed. If multiple products are selected in the product specification area 612, the boxes of the overlapping types may each accept the specification of different text information, or they may be combined into one to accept the specification of text information. Also, the display and display content of the boxes may be configured to change according to the specification result in the category specification area 611. For example, if "Food & Drink" is selected, a box for specifying the address and contact information will be displayed, and if "Event" is selected, a box for specifying the venue and date and time will be displayed. The text specification unit 202 obtains text specification information from the user through these UI screens.

[0081] The image selection area 605 is a UI that accepts the selection of images to be placed in a product. A thumbnail 606 of the selected image is displayed in the image selection area 605. The image addition button 607 is a button that is operated when adding an image to be placed in a product. When the user presses the image addition button 607, the image selection unit 203 displays a dialog screen for selecting an image file via the HDD 104 or network, and accepts the user's selection of an image file on the dialog screen. The thumbnail of the selected image is then added to the image selection area 605. If multiple products are selected in the product creation selection area 612, the image selection area 605 may accept the selection of images individually for each product type, or it may accept the selection of images used in common by multiple products all at once. The image selection unit 203 obtains the image selection information from the user through this UI screen.

[0082] The key design specification area 608 is a UI that accepts the specification of key designs to be used in creating merchandise. A thumbnail 609 of the specified key design is displayed in the key design specification area 608. The key design add button 610 is a button that is operated when adding a key design to be placed on the merchandise. When the user presses the key design add button 610, the key design specification unit 205 displays a dialog screen for selecting a file stored on the HDD 104, and accepts the user's selection of a key design file on the dialog screen. The thumbnail of the selected key design is then added to the key design specification area 608. The key design specification unit 205 obtains the key design specification information from the user through this UI screen.

[0083] The creation condition specification area 650 includes a category specification area 611 and a creation product specification area 612.

[0084] The category specification area 611 is a UI that accepts the specification of the usage category of the product to be created. For example, it has a list box that displays a list of selectable usage categories. The category specification area 611 is not required. However, by providing the category specification area 611, the user can control the generation of products that match the category.

[0085] The product creation specification area 612 is a UI that accepts the specification of the type of product to be created. The product creation specification area 612 has, for example, a checkbox 613 and a distance box 614 for each type of product, and accepts the specification of one or more types of products to be created. The check state of the checkbox 613 can be switched by the user's click operation with the pointing device 107. There are three check states for the checkbox 613: "✓" indicates that it has been specified as a product to be created, "-" indicates that it has been specified as a key design, and a blank space indicates that it is disabled.

[0086] The distance box 614 accepts input from the user for the expected observation distance. If a product is specified as a key design, the expected observation distance of that key design is specified. If a product is specified as a created product, the expected observation distance of that created product is specified. The distance box 614 may also display predetermined initial values ​​for each type of product and each size of product. The initial values ​​are the observation distance values ​​that the developer of the product creation application set in advance during the development of the application. Alternatively, the expected observation distance may be stored within the application, and the creation condition specification unit 201 of the product creation application may obtain the expected observation distance according to the product or key design specified in the checkbox 613. In that case, user input to the distance box 614 is unnecessary, and it is not necessary to display the distance box 614. Furthermore, the key design may be specified in the key design specification area 608 instead of the created product specification area 612.

[0087] The design element specification area 615 is a UI that accepts user specification of design elements for each design element. The design element specification area 615 includes a color scheme specification box 617, a background pattern specification box 620, a logo specification box 621, a font specification box 622, impression sliders 624-627, and an influence level slider 629.

[0088] The color scheme selection box 617 is a UI for inputting color information to be used in product creation. Color 618 represents a thumbnail of the selected color. The add color button 619 is a button for adding a color specification. When the add color button 619 is pressed, a list showing multiple selectable colors is displayed, and when any color is selected from this list using the pointing device 107, the selected color is added. Alternatively, a UI for specifying colors, such as a color palette with multiple colors arranged side by side, may be displayed.

[0089] The background image selection box 620 is a UI for inputting information about background images to be used in product creation. In this embodiment, multiple background images are displayed in a list, and the selection of a background image is accepted by clicking with the pointing device 107. Alternatively, a dialog screen (not shown) displaying files in which background images are stored may be displayed, and the background image may be selected by clicking with the pointing device 107. The file format may be an image file format (JPEG, bitmap, etc.) or a vector data format (PDF).

[0090] The logo selection box 621 is a UI for inputting logo information to be used in product creation. In this embodiment, multiple logos are displayed in a list, and the selection of a logo is accepted by clicking with the pointing device 107. Alternatively, a dialog screen (not shown) displaying the files in which the logos are stored may be displayed, and the logo may be selected by clicking with the pointing device 107. The file format may be an image file format (JPEG, bitmap, etc.) or a vector data format (PDF).

[0091] The font selection box 622 is a UI for inputting information about fonts to be used in product creation. In this embodiment, multiple fonts are displayed in a list, and font selection is accepted by clicking with the pointing device 107. Alternatively, a dialog screen (not shown) displaying the files in which the fonts are stored may be displayed, and the font may be selected by clicking with the pointing device 107.

[0092] Radio button 623 is a button used to control the enabling or disabling of settings for each design element. By pressing radio button 623 to set it to on / off, the user can enable or disable the settings for each design element. Figure 6 shows the state where the color scheme and background pattern are enabled.

[0093] The impression sliders 624-627 are UI elements for specifying a target impression for each product being created. In this embodiment, the target impression includes four impression factors: luxury, approachability, dynamism, and gravitas. For example, impression slider 624 is an operation object for setting the value of the impression factor related to luxury. The further to the right it is slid, the higher the luxury level, and the further to the left it is slid, the lower the luxury level (inexpensive) the product is set. Furthermore, by combining the target impression factors set by each slider, a target impression is set that reflects not only the target impression factors set by one slider, but also the target impression factors set by the other sliders.

[0094] For example, consider a scenario where the user sets the impression slider 624, which controls "luxury," to the right of the center, and the impression slider 627, which controls "solidity," to the left of the center. In this case, an elegant impression, characterized by high luxury and low solidity, is set as the target impression, and the product creation app generates a product with an elegant impression. Alternatively, if the impression slider 624, which controls "luxury," is set to the right of the center, and the impression slider 627, which controls "solidity," is also set to the right of the center, a gorgeous target impression, characterized by high luxury and solidity, is set. In this case, the product creation app generates a product with a gorgeous impression.

[0095] Thus, by combining multiple impression factors, even if a common impression factor such as "luxury" is set, it becomes possible to set target impressions with different directions, such as an "elegant" impression and a "gorgeous" impression. In other words, the target impression is composed of and determined by multiple factors that indicate the impression. However, it is not limited to this, and the target impression may be determined by a single factor that indicates the impression. In this embodiment, the state in which the slider is set to the leftmost position is -2, and the state in which it is set to the rightmost position is +2, and the values ​​indicating the impression factors are represented by integer values ​​from -2 to +2. These numbers indicate the degree of that impression factor, with -2 indicating low, -1 indicating slightly low, 0 indicating neither, +1 indicating slightly high, and +2 indicating high. The purpose of correcting and representing the degree of the impression factor within the numerical range of -2 to +2 is to match the scale with the estimated impression described later and to facilitate the distance calculation described later. Therefore, the values ​​of the impression factors are not limited to the numerical range of -2 to +2, but may also be normalized using values ​​from 0 to 1.

[0096] Radio button 628 controls the enabling or disabling of each impression factor. Users can enable or disable each impression factor by pressing radio button 628 to set it to on / off. For example, selecting "off" with radio button 628 excludes that impression factor from the target impression control. For instance, a user who wants to create a calm product with low dynamism but doesn't want to specify other impressions can turn off all radio buttons 628 except for dynamism to generate a product that specializes in low dynamism. Figure 6 shows a state where luxury and approachability are enabled, and dynamism and gravitas are disabled. Radio button 628 allows for highly flexible impression control, such as specifying a target impression that includes all impression factors or a target impression that includes only some impression factors.

[0097] Furthermore, if each impression slider 624-627 is set to the leftmost position, which is equivalent to not having a corresponding impression factor set, the system may be configured not to accept the on / off operation of radio button 628. For example, if impression slider 624 is set to the leftmost position, the sense of luxury will be set to 0. In this case, if there is an impression factor that the user wants to disable, they can disable the setting by setting that slider to the leftmost position.

[0098] The reflection degree slider 629 is a UI for setting the weight at which each design element information set in the design element specification area 615 is reflected in the product generation. When set to the far left, the weight is set to 0%, the entered design element information is ignored and not reflected in the product. When set to the far right, the weight becomes 100%, and the entered design element information is always reflected in the product. For example, as shown in Figure 6, if the reflection degree slider 629 is set to the 40% position, the reflection degree of the design element will be 40%. The design element acquisition unit 214 sets the skeleton selection unit 216, color scheme pattern selection unit 217, logo selection unit 219, and font selection unit 220 so that the usage frequency or probability of each design element specified in the design element specification area 615 is 40%.

[0099] The design element specification area 615 may also include a checkbox 632 to enable the settings in the design element specification area 615. If a user wants to specify design elements individually, they can enable this by checking the checkbox 632. If the checkbox 632 is not checked and the settings in the design element specification area 615 are disabled, the result will be the same as when the reflection level is set to 0%.

[0100] Each design element may reflect design elements extracted from the key design set in the key design specification area 608 and be set in the UI of the design element specification area 615. When the key design reflection button 616 is pressed, the design element acquisition unit 214 acquires the key design from the key design specification unit 205 and extracts the design elements. The extracted design elements are shared with the design element specification unit 204 and reflected in the design element specification area 615. By reflecting the design elements of the key design, the user can set only the missing or changed elements while referring to the design elements of the key design, making it easier to specify design elements. Note that the design elements that can be specified in the design element specification area 615 shown in Figure 6 are just examples, and other design-related items may also be specified as design elements.

[0101] The reset button 630 is a button for resetting the various settings on the app startup screen 601. When the user presses the OK button 631, the creation condition specification unit 201, text specification unit 202, image specification unit 203, design element specification unit 204, and key design specification unit 205 output the information set on the app startup screen 601 to the product generation unit 210. The creation condition specification unit 201 obtains the type of product to be created and the expected observation distance from the product creation specification area 612, and obtains the usage category of the product to be created from the category specification area 611. The design element specification unit 204 obtains the color from the color scheme specification box 617, the background pattern from the background pattern specification box 620, the logo from the logo specification box 621, and the font from the font specification box 622. The design element specification unit 204 further obtains information on whether to enable the design element information from the checkbox 632, and obtains the target impression of the product to be created from the impression sliders 624-627 and radio buttons 623. The design element specification unit 204 further obtains the degree of reflection of the design element information from the degree of reflection slider 629.

[0102] The text specification unit 202 obtains text information to be placed on the product from the title box 602, subtitle box 603, and body box 604. The image specification unit 203 obtains the image file path to be placed on the product from the image specification area 605. The key design specification unit 205 obtains the key design file path from the key design specification area 608.

[0103] Furthermore, the creation condition specification section 201, text specification section 202, image specification section 203, design element specification section 204, and key design specification section 205 may be modified from the values ​​set on the application startup screen 601. For example, the text specification section 202 may be modified to remove unnecessary whitespace characters at the beginning or end of the entered text information. Also, the design element specification section 204 may be corrected for the target impression values ​​specified by the impression sliders 624 to 627.

[0104] Figure 7 shows an example of the preview screen 701. The preview screen 701 is a screen that displays images of products generated by the product generation unit 210, and is displayed on the display 105 by the generated product display unit 206.

[0105] Figures 7(a) to 7(d) show examples of preview screens 701A, 701B, 701C, and 701D, each displaying different content. This illustrates that the content displayed on preview screen 701 changes according to the information specified by the user on the application startup screen 601. In the following explanation, unless otherwise distinguished, each of the preview screens 701A, 701B, 701C, and 701D will be referred to as preview screen 701. When the OK button 631 on the application startup screen 601 is pressed and the product generation by the product generation unit 210 is completed, the screen displayed on the display 105 transitions to preview screen 701. Since the product generation unit 210 generates multiple products, multiple product images are displayed as a list on preview screen 701. The user selects a product by clicking on one of the product images in the list using the pointing device 107. Multiple products may be selectable.

[0106] The edit button 702 is operated to transition to an editing function (not shown). When the edit button 702 is pressed, the user transitions to a UI that provides editing functions, making the selected product editable. The print button 703 is operated to transition to a printer control UI (not shown). When the print button 703 is pressed, the selected product can be printed via the control UI. The save button 704 is operated to save the selected product. When the save button 704 is pressed, the selected product can be saved to the HDD 104 in a predetermined format that allows for re-editing. The predetermined format is CSV format, JSON format, etc. The information saved includes the estimated impression of the product and design elements (logo, background pattern, color scheme, font).

[0107] The Next Candidate Display button 706 is used to display generated product images that are not fully displayed on the preview screen 701. The number of product images that can be displayed on the preview screen 701 is predetermined according to the screen size. The Next Candidate Display button 706 becomes active when the number of products generated by the product generation unit 210 is greater than the number of product images that can be displayed on the screen. When the Next Candidate Display button 706 is pressed, the product images that were not displayed will be displayed on the screen, and the product images that were displayed will disappear from the screen. The product images that disappear may be one at a time or multiple at once. The user may specify the number of product images that disappear in a designated area not shown. The Previous Candidate Display button 705 is used to display product images that have disappeared from the screen due to the operation of the Next Candidate Display button 706. The Previous Candidate Display button 705 becomes active when the Next Candidate Display button 706 is pressed. When the previous candidate display button 705 is pressed, the product image that was displayed before the next candidate display button 706 was pressed will be displayed, and the product image that was displayed when the next candidate display button 706 was pressed will no longer be displayed.

[0108] Checkbox 707 is used to toggle whether or not to display the key design specified in the key design specification section 205 on the preview screen 701. When checked, the display of the key design is enabled.

[0109] The differences in the display content shown in Figures 7(a) to 7(d) will be explained in detail along with the user-specified information on the app startup screen 601.

[0110] The preview screen 701A shown in Figure 7(a) represents the state where one product is displayed. Preview screen 701A is displayed when, on the app startup screen 601, one type of product is specified, one key design is specified, the content is left blank, and the design elements are left blank. The product images 708 and 709 displayed on preview screen 701A are product images generated using design elements extracted from the key design specified in the key design specification area 608. Two patterns of product images are displayed on this screen, and by operating the next candidate display button 706 or the previous candidate display button 705, the other pattern of product image will be displayed on the screen. The two product images 708 and 709 represent that only the background color is different while maintaining the key design. When such specification information is specified in each item on the app startup screen 601, variations of the design similar to the key design are generated, with the background color changed while maintaining the key design.

[0111] The preview screen 701B shown in Figure 7(b) shows the state in which the key design is displayed. Compared to the preview screen 701A in Figure 7(a), the checkbox 707 is enabled and the key design image 710 is displayed. The specified information on the application startup screen 601 is the same as in Figure 7(a). As shown in preview screen 701B, by displaying the product images 708 and 709 generated by the product generation unit 210 and the key design image 710 side by side on the screen, it becomes easy to compare the product images generated by the product generation unit 210 with the key design. In the example of Figure 7(b), the key design image 710 and product images 708 and 709 are enlarged or reduced so that they can be displayed within the preview area of ​​the preview screen 701, respectively. However, the system is not limited to this, and each product may be displayed in the ratio of its actual product size.

[0112] The preview screen 701C shown in Figure 7(c) displays the state where multiple products are displayed. Compared to Figure 7(a), the preview screen 701C shows the state when multiple types of products are selected. Compared to the preview screen 701A in Figure 7(a), product images 711 and 712 of a different type are displayed in addition to product images 708 and 709. The specified information on the app startup screen 601 is assumed to be two types of products to be created, one key design, no content specified, and no design element specified. The product images 711 and 712 displayed on the preview screen 701C are generated using design elements extracted from the key design specified in the key design specification area 608, and the types of products are different from product images 708 and 709. For example, product images 708 and 709 are postcards, and product images 711 and 712 are flyers. Thus, the preview screen 701C displays two patterns each of different types of product images. The difference between the two product images 708 and 709 is that only the background color is different while maintaining the key design. Similarly, the difference between the two product images 711 and 712 is that only the background color is different while maintaining the key design. Furthermore, product image 708 for postcards and product image 711 for flyers have different product types and images used, but the key design and background color are similar. Similarly, product image 709 for postcards and product image 712 for flyers also have different product types and images used, but the key design and background color are similar. In other words, on the preview screen 701C, for different types of products, two similar key design patterns are created for each, while maintaining the key design. By operating the next candidate display button 706 or the previous candidate display button 705, other patterns of product images for each type of product will be displayed on the screen. In this way, by selecting multiple types of products on the app launch screen 601, multiple design patterns for multiple types of products can be generated at once. In other words, it is possible to generate multiple design patterns for combinations of different types of products.

[0113] The preview screen 701D shown in Figure 7(d) illustrates an example of a product image generated when content is specified. The specified information on the app launch screen 601 differs from the specified information in Figure 7(a) in that content settings (image settings) are performed. By specifying content on the app launch screen 601, the user can control the content 713 included in the generated product. This allows for the appropriate generation of product design combinations that express the user's intent, even when using user-specified content. Note that product images 715 and 716 also have different background image sizes. The background image variations can also be compared using the background image scaling process described later.

[0114] <Quantifying the impression of a product> Here, we will explain the process for quantifying the impression of a product (hereinafter referred to as the product impression quantification process). The product impression quantification process is a pre-processing step necessary for executing the product impression estimation process, which will be described later. The product impression estimation process is executed in the product generation process (S1014 in Figure 10(a)), which will be described later. In this embodiment, we will explain the process using the quantification of the impression of a poster as an example of a product. Note that the impression of other products can also be quantified by performing the same process as in the case of a poster. Furthermore, for products that are similar in size and use, the impression quantification result of one product may be directly applied to the other product.

[0115] The process of quantifying the impression of the poster is performed by the application development vendor during the development phase of the product creation application. The process of quantifying the impression of the poster may be performed by the product generation device 100, or by an information processing device different from the product generation device 100. If performed by an information processing device different from the product generation device 100, it will be performed by the CPU of that information processing device.

[0116] The poster impression quantification process quantifies the impressions people have of various posters. It also derives a correspondence between poster images and the impressions they evoke. This makes it possible to estimate the impression of a poster from the generated poster images. Once the impression is estimated, it becomes possible to control the impression of a poster by modifying the poster image, or to search for a poster image that embodies a specific target impression. The poster impression quantification process is performed, for example, by running an impression learning application in a poster generation device before the poster generation process to learn the impressions of poster images.

[0117] Figure 8 is a flowchart showing the process of quantifying the impression of a poster. The flowchart shown in Figure 8 is initiated, for example, when a developer launches an impression learning application, and is realized when the CPU 101 reads the program stored in the HDD 104 into the RAM 103 and executes it. The process of quantifying the impression of a poster will be explained with reference to Figure 8. In the description of each process, the symbol "S" indicates a step in the flowchart. The same applies hereafter in this specification.

[0118] In S801, CPU101 performs the task of obtaining subjective evaluations of impressions of posters. Figure 9 illustrates an example of a method for subjectively evaluating impressions of posters. CPU101 presents posters to subjects and obtains subjective evaluations of the impressions they receive from those subjects. Measurement methods such as the Semantic Differential (SD) method and the Likert scale can be used in this case. Figure 9 shows an example of a questionnaire using the SD method, in which multiple evaluators are presented with pairs of adjectives that express impressions, and a score is assigned to the adjective pairs that are recalled from the target poster. After obtaining subjective evaluation results from multiple subjects for multiple posters, CPU101 calculates the average value of the responses for each pair of adjectives, and uses this average value as the representative score for the corresponding pair of adjectives. Note that the subjective evaluation method for impressions does not have to be the SD method; it is sufficient to have words that express impressions and corresponding scores.

[0119] In S802, the CPU 101 performs factor analysis on the subjective evaluation results obtained in S801. If the subjective evaluation results are left as they are, the number of adjective pairs becomes a dimensionality, making control complex. Therefore, it is desirable to reduce the dimensionality to an efficient level using analytical methods such as principal component analysis or factor analysis. In this embodiment, we will explain the process assuming that the dimensionality has been reduced to four factors through factor analysis. Naturally, this number varies depending on the selection of adjective pairs in the subjective evaluation and the factor analysis method used. Furthermore, the output of the factor analysis is assumed to be standardized. That is, each factor is scaled so that the mean is 0 and the variance is 1 in the poster used for the analysis. This allows the impressions -2, -1, 0, +1, and +2 specified in the design element specification unit 204 to directly correspond to -2σ, -1σ, mean, +1σ, and +2σ in each impression, facilitating the calculation of the distance between the target impression and the estimated impression, as described later. In this embodiment, the four factors are referred to as luxury, familiarity, dynamism, and gravitas, as shown in Figure 6. These are names conveniently assigned to convey an impression to the user through the user interface, and each factor is composed of multiple adjective pairs influencing each other. The CPU 101 also stores the conversion formulas (hereinafter referred to as "impression conversion formulas") obtained through factor analysis, which convert the subjective evaluation results of each adjective pair to the values ​​of each impression, in the HDD 104.

[0120] In S803, CPU101 associates poster images with impressions. While quantitative analysis is possible for posters that have undergone subjective evaluation using the method described above, it is necessary to estimate impressions of posters to be created without subjective evaluation. Associating poster images with impressions can be achieved by training a model that estimates impressions from poster images. Training can be done using deep learning methods such as Convolutional Neural Networks (CNNs), Visual Transformers (ViTs), or decision trees. In this embodiment, CPU101 takes poster images as input and four factors as output, and performs supervised deep learning using a CNN. That is, a deep learning model is created that has been trained with subjectively evaluated poster images and their corresponding impressions as ground truth, and impressions are estimated by inputting unknown poster images into this trained model.

[0121] In S804, CPU101 saves the model configuration and trained parameters of the deep learning model for impression estimation created in S803 to HDD104.

[0122] The impression estimation unit 222 loads the deep learning model stored in the HDD 104 into the RAM 103 and executes it. The impression estimation unit 222 converts the poster data acquired from the layout unit 221 into an image and estimates the impression of the poster by running the deep learning model loaded into the RAM 103 on the CPU 101 or GPU 109. In this embodiment, a deep learning method is used, but it is not limited to this. For example, if a machine learning method such as a decision tree is used, feature quantities such as the average brightness and edge quantity of the poster image may be extracted by image analysis, and a machine learning model that estimates the impression based on these feature quantities may be created.

[0123] <Product Generation Process> Figure 10(a) is a flowchart showing the product generation process executed by the product generation unit 210 of the product creation application. As described above, the flowchart shown in Figure 10(a) starts when the user presses the launch icon of the product creation application. The flowchart shown in Figure 10(a) is realized, for example, by the CPU 101 reading a program stored in the HDD 104 into the RAM 103 and executing it. In this embodiment, it will be explained that each part shown in Figure 2, which functions when the CPU 101 executes the product creation application, performs its processing. The product generation process will be explained with reference to Figure 10(a).

[0124] In S1001, the product creation application displays the application startup screen 601 shown in Figure 6 on the display 105. The user inputs each setting via the UI screen of the application startup screen 601 using the keyboard 106 or pointing device 107.

[0125] In S1002, the creation condition specification unit 201, text specification unit 202, image specification unit 203, design element specification unit 204, and key design specification unit 205 obtain their respective settings from the application startup screen 601.

[0126] The creation condition specification unit 201 obtains the type and category of the product to be created. The type of product to be created is specified in the product specification area 612. In the example in Figure 6, it is shown that two products, flyers and postcards, are specified as products to be created. The category is specified in the category specification area 611. In the example in Figure 6, it is shown that food and beverages are obtained as the category.

[0127] The text specification unit 202 retrieves the text information specified in the title box 602, subtitle box 603, and body box 604. In the example in Figure 6, the text information "Tittle2Tittle2Tittle2" is retrieved as the title, and no text information is retrieved for the subtitle because it is blank. It also shows that the text information for four lines of "TextTextText" is retrieved as the body text.

[0128] The image selection unit 203 retrieves the image specified in the image selection area 605. In the example in Figure 6, it is shown that the image corresponding to the thumbnail 606 is retrieved.

[0129] The design element specification section 204 obtains the color scheme, background pattern, logo, font, and target impression, as well as the degree to which these design elements are reflected.

[0130] The design element specification unit 204 retrieves the color specified in the color scheme specification box 617 as the color scheme. Furthermore, it retrieves whether to use the specified color scheme for product generation based on the on / off status of the radio button 623. In the example in Figure 6, the color represented by color 618 is retrieved and will be used for product generation. The design element specification unit 204 retrieves the background pattern specified in the background pattern specification box 620 as the background pattern. Furthermore, it retrieves whether to use the specified background pattern for product generation based on the on / off status of the radio button 623. In the example in Figure 6, a checkered pattern is specified as the background pattern and will be used for product generation.

[0131] The design element specification unit 204 retrieves the logo specified in the logo specification box 621 as the logo. Furthermore, it determines whether the specified logo will be used for product generation based on the on / off status of the radio button 623. In the example in Figure 6, no logo is specified, indicating that it will not be used for product generation. The design element specification unit 204 retrieves the font specified in the font specification box 622 as the font. Furthermore, it determines whether the specified font will be used for product generation based on the on / off status of the radio button 623. In the example in Figure 6, a Gothic font is specified, but it indicates that it will not be used for product generation.

[0132] The design element specification unit 204 obtains the values ​​of each impression factor from the impression sliders 624 to 627 as the target impression. Furthermore, it obtains whether or not to use the specified impression factors in product generation based on the on / off status of the radio button 628. In the example in Figure 6, luxury is set to -1, approachability to +1, dynamism to -0.8, and solidity to 0. Furthermore, luxury and approachability are used as target impressions in product generation, while dynamism and solidity are not used in product generation. The design element specification unit 204 obtains the reflection level specified by the reflection level slider bar 629 as the reflection level. In the example in Figure 6, 40% is specified.

[0133] The key design specification unit 205 retrieves the key design specified in the key design specification area 608. In the example in Figure 6, the design data (first product data) corresponding to the thumbnail 609 displayed in the key design specification area 608 is retrieved as the key design that will serve as the basis for the created product.

[0134] In S1003, the image acquisition unit 211 acquires image data. Specifically, the image acquisition unit 211 identifies the image file corresponding to the thumbnail 606 acquired by the image selection unit 203 in S1002, and reads the identified image file from the HDD 104 into the RAM 103.

[0135] In S1004, the image analysis unit 212 performs analysis on the image data acquired in S1003 and obtains information indicating feature quantities. Examples of information indicating feature quantities include metadata stored in the image and image features that can be obtained by analyzing the image. This information is used in the analysis processes of object recognition and major color extraction. In this embodiment, object recognition and major color extraction are performed as analysis processes, but the system is not limited to these, and other analysis processes may be performed. Furthermore, the process in S1004 may be omitted. The details of the process performed by the image analysis unit 212 in S1004 will be described below.

[0136] The image analysis unit 212 performs object recognition processing on the image acquired in S1003. Here, known methods can be used for object recognition processing. In this embodiment, objects are recognized by a discriminator created by Deep Learning as the object recognition processing. The discriminator outputs a likelihood of 0 to 1 that a certain pixel in the image is a pixel that constitutes each object, and recognizes that an object exceeding a certain threshold is present in the image. By recognizing the object image, the image analysis unit 212 can obtain the type and location of objects such as faces, pets such as dogs or cats, flowers, food, buildings, ornaments, and landmarks.

[0137] Furthermore, the image analysis unit 212 performs a primary color extraction process on the image acquired in S1003 to determine the primary color of the image. Here, known methods can be used for the primary color extraction process. In this embodiment, the number of pixels for each pixel value is counted, and the mode is determined as the primary color. However, when determining the primary color by pixel value, similar colors such as gradients may be processed as different colors, and the primary color may not be optimal. In that case, as an alternative processing example, the primary color may be determined by grouping a predetermined range of color differences as the same color and counting the number of pixels in the group. This makes it possible to determine a primary color that takes similar colors into account, and to determine a primary color that is closer to human perception. Furthermore, the degree of primary color may be determined not only by the number of pixels, but also by the saturation and brightness of the determined pixel value, and the color difference with surrounding pixels. Colors with high saturation, bright colors, and dark colors stand out and are therefore more likely to be recognized as primary colors. Also, colors with a large color difference with surrounding pixels stand out and are therefore more likely to be recognized as primary colors. Specifically, the calculation of the primary chromaticity is performed such that the higher the saturation, the further the brightness is from gray, and the greater the color difference with surrounding pixels. An example of the calculation formula is shown in equation (5) below. In equation (5), Mc is the primary chromaticity, Pn is the number of pixels, S is the saturation, L is the brightness, and ΔE is the color difference with surrounding pixels.

[0138] Principal chromaticity Mc=Pn×(S+|50-L|+ΔE) ···(5)

[0139] The image analysis unit 212 determines the color with the highest primary chromaticity Mc determined by equation (5) as the primary color. This allows for the determination of a primary color that closely matches human perception. Furthermore, the primary color extraction process is not limited to the process described above, and any method may be used. For example, a discriminator created using deep learning may be used to extract the primary color, or a clustering method such as K-means may be used to extract the primary color. Also, there may be more than one primary color.

[0140] In S1005, the product generation unit 210 determines the products to be created. Specifically, the product generation unit 210 determines the types of products to be created and the number of each type to be created. The product types are obtained in S1002. In the example in Figure 6, two product types are specified. Furthermore, the predetermined number to be created for each product type is obtained, and the total number to be created for all types is determined. In this embodiment, the predetermined number for each product type is 5. In other words, in the example in Figure 6, the number to be created is 5 postcards and 5 flyers, for a total of 10. That is, it is decided to create 10 products. In this embodiment, a product generation result that includes all the specified product types and in which one design is selected for each of those product types is called a product set. In the example in Figure 6, a poster is specified as the key design, and postcards and flyers are specified as products to be created. In this case, the combination of postcards and flyers, which are the products to be created, constitutes one product set.

[0141] In this embodiment, five items were created for each product type, but the number of items is not limited to five and may be any other number. A larger number of items makes it easier to generate various variations of product sets that closely match the target impression and have similar designs. Here, "similar designs" means that they contain common design elements.

[0142] Furthermore, the processes from S1008 to S1016, described later, are repeated for the number of product types to be created, and S1009 to S1015 are repeated for the number of times each product type is created. In other words, S1009 to S1015 are repeated for the number of product types multiplied by the number of products created. In this embodiment, there are 2 product types and 5 products to be created, so S1009 to S1015 are repeated 10 times.

[0143] In S1006, the design element acquisition unit 214 acquires design elements to be used in creating the product. Specifically, the design element acquisition unit 214 first extracts design elements from each key design specified by the key design specification unit 205.

[0144] If the key design is vector data, the design element acquisition unit 214 extracts the color scheme, background image, logo, and font from the tag information. For example, if it is stored as SVG data, the design element acquisition unit 214 extracts elements with a wide drawing area as the background image and elements with a narrow drawing area as the logo. Alternatively, the design element acquisition unit 214 may acquire the graphic elements on the lowest layer as the background image and the graphic elements on the upper layers as the logo. If there is a font name in the text tag area, the design element acquisition unit 214 extracts the font name as the font. The design element acquisition unit 214 extracts the color scheme list from the color values ​​contained in the tag information. Note that the acquisition of the color scheme list is not limited to this method. The design element acquisition unit 214 may rasterize the key design and then perform the main color extraction process described above to acquire the color scheme information.

[0145] If the key design is raster data, the design element acquisition unit 214 uses a machine learning model such as Deep Learning to perform area separation and acquire the design elements. The design element acquisition unit 214 stores the extracted color scheme from the key design as a color scheme list, the background pattern as a background pattern list, the logo as a logo list, and the font as a font list in the RAM 103.

[0146] Next, the design element acquisition unit 214 acquires the design elements specified by the design element specification unit 204 and stores the acquired color schemes, background patterns, logos, and fonts in their respective lists. If the actual data of the background patterns, logos, and fonts stored in the lists is already saved in the HDD 104, the design element acquisition unit 214 sequentially reads that actual data into the RAM 103. In the case of raster data, the design element acquisition unit 214 cuts out the images from the results of image area separation and sequentially stores them in the RAM 103. For the color scheme list, the color scheme extracted from the key design and the color scheme specified by the design element specification unit 204 are kept in a way that allows for distinction.

[0147] Figure 11 shows examples of color scheme lists. Figure 11(a) shows an example of a color scheme list 1101 specified by the design element specification unit 204, and Figure 11(b) shows a color scheme list 1102 obtained from the specified image as a result of analysis by the image analysis unit 212. Figure 11(c) shows an example of a color scheme list 1103 extracted from the key design, and Figure 11(d) shows an example of a color scheme list 1104 newly generated by the color scheme pattern acquisition process (S1007) described later. Each list stores information indicating the color and information indicating the type of data from which it was acquired, linked together. The color values ​​are, for example, represented by the RGB color values ​​in that order from 0 to 255.

[0148] In S1007, the color scheme pattern acquisition unit 215 acquires a color scheme pattern to be used in creating the product. Figure 10(b) is a flowchart illustrating the color scheme pattern acquisition process performed by the color scheme pattern acquisition unit 215 in S1007. The acquisition of the color scheme pattern will be explained in detail using Figure 10(b).

[0149] In S1021, the color scheme pattern acquisition unit 215 acquires color scheme lists from the image analysis unit 212 and the design element acquisition unit 214, respectively. Specifically, the color scheme list 1102 shown in Figure 11(b) is acquired from the image analysis unit 212, and the color scheme list 1101 shown in Figure 11(a) and the color scheme list 1103 shown in Figure 11(c) are acquired from the design element acquisition unit 214. The color scheme pattern acquisition unit 215 registers the color scheme list 1102 acquired from the image analysis unit 212 into a new color scheme list 1104 and stores it in RAM 103. In addition, the color scheme pattern acquisition unit 215 registers the color scheme list 1101, which was specified by the design element specification unit 204, from the color scheme lists 1101 and 1103 acquired from the design element acquisition unit 214, into the new color scheme list 1104. The color scheme pattern acquisition unit 215 passes the color scheme list 1103 extracted from the key design from the color scheme list acquired in S1006 to S1022.

[0150] In S1022, the color scheme acquisition unit 215 selects one of the main colors (primary colors) in a key design from the color scheme list 1103 received from S1021, and determines the hue of the selected color scheme. In the following explanation, the selected color scheme will be referred to as the selected color. The specific determination method will be explained using Figure 12(a).

[0151] Figure 12(a) shows a color wheel, illustrating that the hue angle from 0 to 360 degrees is divided into 12 representative colors. For example, clockwise from color 1201, these represent yellow, yellow-green, green, blue-green, blue, blue-violet, violet, red-violet, red, red-orange, orange, and yellow-orange. The color scheme pattern acquisition unit 215 determines which hue the selected color belongs to. The color scheme pattern acquisition unit 215 calculates the hue angle between each hue and determines the representative color that is closest to the hue angle of the selected color. This closest representative color is then used as the hue of the selected color.

[0152] In S1023, the color scheme pattern acquisition unit 215 determines the hues of candidate new color schemes from the hues of the selected color. The new color scheme will have hues different from the hues of the selected color scheme. Specifically, hues that are complementary, contrasting, or analogous to the hues of the selected color scheme are determined as candidate hues. In Figure 12(a), if color 1201 is the hue of the selected color, then the complementary color is color 1202, the contrasting colors are the range of colors including colors 1205 to 1203 and colors 1204 to 1206, and the analogous colors are colors 1207 to 1208. The color scheme pattern acquisition unit 215 determines one hue for the new color scheme from the candidate hues from colors 1202 to 1208. The determination method may be random. Preferably, the hue of the new color scheme is determined according to the target impression acquired by the design element acquisition unit 214. Specifically, the color scheme pattern acquisition unit 215 determines the hue that has the impression value closest to the target impression from among the impression values ​​that have been pre-associated with the hue candidates.

[0153] For example, if the "dynamism" value of the specified target impression is high, red, red-orange, yellow, and yellow-orange, which have high "dynamism" impression values, will be determined as the new color scheme hues. Conversely, if the "dynamism" value of the specified target impression is low, blue-violet, blue, and blue-green, which have low "dynamism" impression values, will be determined as the new color scheme hues. Also, for example, if the "affinity" value of the specified target impression is high, orange, yellow-orange, yellow, yellow-green, and green, which have high "affinity" impression values, will be determined as the new color scheme hues. Conversely, if the "affinity" value of the specified target impression is low, red-violet, purple, blue-violet, and blue, which have low "affinity" impression values, will be determined as the new color scheme hues.

[0154] Furthermore, the hues of a new color scheme may be determined in accordance with the target impression and its hue relationship. For example, hues that are complementary or contrasting to the selected color will be determined as the hues of a new color scheme when the "dynamism" value of the target impression is high, while hues that are analogous to the selected color will be determined as the hues of a new color scheme when the "dynamism" value of the target impression is low. This allows for the determination of a color scheme pattern that matches the target impression specified by the user, and as a result, it becomes possible to generate products that match the target impression.

[0155] In S1024, the color scheme acquisition unit 215 determines the saturation and brightness of the new color scheme based on the selected color and the hue of the new color scheme determined in S1023. Specifically, the color scheme acquisition unit 215 sets the brightness and saturation of the new color scheme to the brightness and saturation of the selected color. The color scheme acquisition unit 215 also sets the hue of the new color scheme to the hue of the color scheme determined in S1023. This allows a new color scheme to be determined that has a different hue but the same brightness and saturation as the selected color. The color scheme determined in S1024 is stored in the new color scheme list 1104. Here, "same color" refers to a color whose brightness difference and saturation difference are less than or equal to a predetermined difference. A predetermined difference means, for example, that the spatial distance ΔE in the CIEL*a*b* color space is 6.5 or less. Note that in the sRGB color gamut, the same color brightness and saturation may not be representable. In that case, the lightness and saturation of the new color scheme may be determined by the relationship between the hue of the selected color and the hue of the new color scheme. This will be explained in detail with reference to Figure 12(b).

[0156] Figure 12(b) shows the lightness-saturation plane for a certain hue angle in the CIEL*a*b* color space, where color gamut 1209 represents the sRGB color gamut. When the selected color is at position 1210, the color with the highest saturation in the sRGB color gamut at the hue of the selected color is represented by color 1211. Furthermore, the color with the highest saturation in the sRGB color gamut at the hue of the new color scheme is represented as color 1212. Color 1213 represents an intermediate color (L*=50) on the gray axis (a*=0, b*=0) of the sRGB color gamut. The method for determining the new color scheme in this case will be explained using Figure 12(b).

[0157] First, let's explain how the brightness is determined. The brightness of the new color scheme 1214 is determined by the ratio of the brightness difference between color 1213 and color 1211, and the brightness difference between color 1213 and color 1210. Specifically, if the brightness of the selected color 1210 is Ls, the brightness of color 1211, which is the most saturated color in the selected color scheme's hue, is Lsh, and the brightness of color 1212, which is the most saturated color in the new color scheme, is Lnh, then it is determined by the following equation (6).

[0158] The brightness of the new color scheme = 50 + ((Ls-50)÷(Lsh-50)×(Lnh-50))···(6) For example, if the lightness of color 1211 is 70, the lightness of color 1210 is 60, and the lightness of color 1212 is 30, then the lightness of color 1214 will be 40.

[0159] Next, we will explain how to determine the saturation. The saturation of the new color scheme 1214 is determined from the saturation ratio of color 1210 and color 1211 and the saturation of color 1212. Specifically, if the saturation of the selected color 1210 is Hs, the saturation of color 1211, which is the most saturated color in the hue of the selected color, is Ssh, and the saturation of color 1212, which is the most saturated color in the new color scheme, is Snh, then it is determined by the following equation (7).

[0160] Saturation of the new color scheme = Snh × (Hs ÷ Hsh) ... (7)

[0161] For example, if the saturation of color 1211 is 80, the saturation of color 1210 is 50, and the saturation of color 1212 is 60, then the saturation of color 1214 will be 37.5.

[0162] By performing the calculations described above, it is possible to determine the brightness and saturation of a new color scheme even if the hues do not have the same saturation and brightness. In this embodiment, it is sufficient if the new color scheme is the same as the one determined by the method described above. The same color is a color in which the difference in brightness and the difference in saturation are less than or equal to a predetermined difference. Less than or equal to a predetermined difference means, for example, that the spatial distance ΔE in the CIEL*a*b* color space is 6.5 or less.

[0163] More preferably, subjective evaluations by individuals may be performed in advance for each hue to determine the corresponding saturation and brightness. The method of subjective evaluation will be explained with reference to Figure 13. Figure 13 shows a subjective evaluation in which a color that gives the same impression as color 1301 is selected from the 3x3 patch 1302 on the right. The central patch of the 3x3 patch 1302 represents the brightness and saturation calculated by the above formula. Nine patches are displayed in patch 1302, with +5 and -5 in the brightness direction and +5 and -5 in the saturation direction from the central patch. Subjective evaluations are performed for each of the 12 hue circles for a total of 144 colors: 12 colors with different hue angles for color 1301 and 12 combinations of brightness and saturation. This allows the brightness and saturation of a new color scheme to be determined by subjective evaluation so that they are perceived as the same by people. This makes it possible to determine new color schemes that do not feel unnatural to people.

[0164] Return to Figure 10(b). In S1025, the color scheme pattern acquisition unit 215 determines whether a new color scheme has been determined for all color schemes in the color scheme list acquired in S1021. If a new color scheme has been determined for all color schemes, the process proceeds to S1026. Otherwise, the process proceeds to S1022.

[0165] In S1026, the color scheme pattern acquisition unit 215 acquires a color scheme pattern from the colors in the newly determined color scheme list 1104. If there are identical colors in the new color scheme list 1104, they are merged to create a unique color. The basis of the color scheme pattern is the color scheme pattern extracted from the key design. The colors stored in the new color scheme list 1104 are replaced with the main colors of the color scheme pattern and stored in the color scheme pattern list as a new color scheme pattern. This makes it possible to acquire a color scheme pattern with a different hue from the key color. In addition to replacing the main color, colors other than the main color included in the color scheme pattern may also be changed according to the new color scheme. For example, the hue angle of colors other than the main color may be changed according to the hue angle difference between the main color before replacement and the new color scheme. The color scheme pattern acquisition unit 215 stores the modified color scheme pattern in the color scheme pattern list. Note that both the replacement method and the method of changing colors other than the main color may be performed together. This makes it possible to increase the number of new color scheme patterns. Furthermore, the color scheme pattern acquisition unit 215 may add a new color scheme pattern from a pre-stored group of preferred color scheme patterns according to the new color scheme. If there is a new color scheme among the colors included in the pre-stored group of color scheme patterns, that color scheme pattern is stored in the color scheme pattern list as a new color scheme pattern. In this way, a wider variety of products can be generated. The above is an explanation of the color scheme pattern acquisition process in S1007. Returning to the explanation of Figure 10(a).

[0166] In S1008, the skeleton acquisition unit 213 acquires skeletons that meet various setting conditions for the type of product to be processed. In this embodiment, it is assumed that each skeleton is written to a separate file and stored in the HDD 104. The skeleton acquisition unit 213 sequentially reads the skeleton files from the HDD 104 into the RAM 103, keeps the skeletons that meet the setting conditions in the RAM 103, and deletes the skeletons that do not meet the conditions from the RAM 103.

[0167] Figure 10(c) is a flowchart illustrating the conditional judgment process executed by the skeleton acquisition unit 213 in S1008. The conditional judgment process will be explained in detail with reference to Figure 10(c).

[0168] In S1031, the skeleton acquisition unit 213 determines whether the size of the skeleton loaded into RAM 103 matches the size pre-set for the type of product to be processed. For example, suppose the size of the flyer specified as a product to be created on the app startup screen 601 is A4, and the size of the postcard is 100mm x 148mm. Note that here, the size must match, but it is also sufficient if the aspect ratio matches. In that case, the skeleton acquisition unit 213 acquires a skeleton that matches the size of the product to be processed by scaling the coordinate system of the loaded skeleton.

[0169] In S1032, the skeleton acquisition unit 213 determines whether the category of the skeleton matches the usage category specified in the creation condition specification unit 201. For skeletons used only for specific purposes, the usage category is recorded in the skeleton file so that it is not acquired unless the corresponding usage category is selected. This prevents the skeleton from being used in other usage categories if it is designed specifically for a particular purpose, such as if it has a graphic representation of a school or a sports equipment design. Note that if no usage category is set on the application startup screen 601, S1032 is skipped.

[0170] In S1033, the skeleton acquisition unit 213 determines whether the number of image objects placed in the loaded skeleton matches the number of images acquired by the image acquisition unit 211. In S1034, the skeleton acquisition unit 213 determines whether the character objects placed in the loaded skeleton match the character information specified in the text specification unit 202. More specifically, it determines whether the type of character information specified in the text specification unit 202 exists in the skeleton. For example, suppose that in the app startup screen 601, strings are specified in the title box 602 and the body box 604, and a blank is specified in the subtitle box 603. In this case, all character objects placed in the skeleton are searched, and if both character objects with "Title" set as the type of character information in the metadata and character objects with "Body" set are found, it is considered a match; otherwise, it is considered an ineligible match.

[0171] In S1035, the skeleton acquisition unit 213 determines whether a background graphic object exists in the loaded skeleton. A background graphic object must exist in the skeleton in order to draw the background image acquired from the key design. The determination of whether or not an object is a background graphic object is made based on the metadata information, which is pre-set as metadata for the graphic object indicating whether or not it is a background. Alternatively, the graphic object with the largest area among those used as the base for other design elements may be determined as the background object.

[0172] In S1036, the skeleton acquisition unit 213 determines whether a logo object exists in the loaded skeleton. If a logo is specified in the creation condition specification unit 201, an object must exist in the skeleton in order to draw the logo. If a logo is not specified in the creation condition specification unit 201, S1036 is skipped.

[0173] As a result of the above processing, a skeleton is stored in RAM103 in which all of the following conditions match the set conditions: skeleton size, usage category, number of image objects, type of text objects, presence or absence of background graphic objects, and presence or absence of logo objects.

[0174] In S1037, the skeleton acquisition unit 213 selects the skeleton remaining in RAM 103 as the skeleton to be used for poster generation.

[0175] In this embodiment, the skeleton acquisition unit 213 determined all skeleton files stored in the HDD 104, but this is not limited to this. For example, the product creation application may store a database in the HDD 104 that associates the file paths of skeleton files with search conditions in advance. In that case, the skeleton acquisition unit 213 can acquire skeleton files at high speed by searching the database and reading only the matching skeleton files from the HDD 104 to the RAM 103. The search conditions include, for example, the skeleton size, the number of image objects, the type of text object, the presence or absence of a background graphic object, and the presence or absence of a logo object. This concludes the explanation of S1008. Returning to the explanation of Figure 10(a).

[0176] In S1009, the skeleton selection unit 216 selects a skeleton from those acquired in S1008 that matches the target impression acquired by the design element acquisition unit 214. Figure 14 is a diagram illustrating the skeleton selection for a poster, which is an example of a commercial product. The same procedure applies to other types of commercial products besides posters.

[0177] Figure 14(a) shows an example of a skeleton impression table that links skeletons with the impressions associated with them. In the table shown in Figure 14(a), the column for skeleton name contains the file name of the skeleton, and the columns for luxury, approachability, dynamism, and solidity represent numbers (numerical values) that indicate how much each skeleton influences each impression. These numbers indicate the impression as follows: -2 is low, -1 is somewhat low, 0 is neutral, +1 is somewhat high, and +2 is high. First, the skeleton selection unit 216 calculates the distance between the target impression obtained from the design element acquisition unit 214 and the impressions of each skeleton shown in the skeleton impression table in Figure 14(a). Hereinafter, the distance between the two impression values ​​will be called the impression distance. For example, if the target impression is "luxury +1, approachability -1, dynamism -2, solidity +2", the impression distance calculated by the skeleton selection unit 216 will be as shown in Figure 10(b). In this embodiment, Euclidean distance is used as the distance. Hereafter, unless otherwise specified, the impression distance will be the Euclidean distance. The smaller the value indicated by the Euclidean distance, the closer the target impression is to the impression of the skeleton. Next, the skeleton selection unit 216 selects the top N skeletons with small impression distance values ​​in Figure 14(b). In this embodiment, the skeleton selection unit 216 selects the top two skeletons with small impression distance values. In the example in Figure 14(b), skeletons 1 and 4 are selected. The value of N can be any integer greater than or equal to 1.

[0178] Note that the value ranges for each impression in the skeleton impression table in Figure 14(a) do not necessarily have to be the same as the value ranges for the impressions specified in the design element specification unit 204. In this embodiment, the value range for the impressions specified in the design element specification unit 204 is -2 to +2, but the value ranges for the impressions in the skeleton impression table may be different. If they are different, the value ranges in the skeleton impression table are scaled to match the value range of the target impression before calculating the impression distance described above. Also, the distance calculated by the skeleton selection unit 216 is not limited to Euclidean distance; it is acceptable as long as it can calculate the distance between vectors, such as Manhattan distance or cosine similarity. Furthermore, impression factors for which the radio button 628 on the app startup screen 601 is set to off are excluded from the distance calculation.

[0179] The skeleton impression table in Figure 14(a) is created in advance and stored on HDD104. The impression of each skeleton is estimated based on the generated poster image, after fixing, for example, the color scheme, font, and the images and text data placed on the skeleton. That is, the impression of each poster image is estimated, even though the colors of the text and images used are the same, but the arrangement of the text and images is different. This creates a table of the relative characteristics of each skeleton. In this case, it is desirable to perform a process to cancel out the impression caused by the color scheme and images used, such as standardizing the estimated impressions as a whole or averaging the impressions of multiple poster images generated from a single skeleton using multiple color schemes and images. This allows the influence of the arrangement itself on the impression to be recorded in the table, for example, that the impression of a skeleton with a small image is determined by design elements such as graphics and text rather than the image, or that a tilted arrangement of images and text gives a strong impression of dynamism.

[0180] Figure 14(c) shows examples of skeletons corresponding to Skeletons 1 to 4 in Figure 14(a). For example, Skeleton 1 has image and text objects arranged regularly, and the image area is small, resulting in a low sense of dynamism. Skeleton 2 has a high sense of familiarity and a low sense of weight due to the circular shape of the graphic and image objects. Skeleton 3 has a high sense of dynamism by placing large image objects and tilted graphic objects on top of them. Skeleton 4 has a high sense of weight and a low sense of dynamism by placing images throughout the entire skeleton and minimizing the number of text objects. In this way, when a poster image includes text or images, different poster images with different target impressions are generated depending on how the text or images are arranged. Note that the method of creating the skeleton impression table is not limited to these; it may also be estimated from the characteristics of the arrangement information itself, such as the area and coordinates of images and title strings, or it may be adjusted manually. The skeleton impression table is stored in the HDD 104, and the skeleton selection unit 216 reads the skeleton impression table from the HDD 104 into the RAM 103 for reference.

[0181] In S1010, the color scheme pattern selection unit 217 selects a color scheme pattern from the color scheme patterns acquired in S1007 that matches the target impression acquired by the design element acquisition unit 214. The method for selecting a color scheme pattern that matches the target impression is the same as the method for selecting the skeleton in S1009. That is, the color scheme pattern selection unit 217 refers to the impression table corresponding to the color scheme pattern and selects a color scheme pattern that is close to the target impression.

[0182] Figure 15 shows an example of a color scheme impression table that links color schemes with impressions. The color scheme selection unit 217 calculates the impression distance between the values ​​of each impression factor shown in the columns for "luxury" to "heaviness" in Figure 15 and the values ​​of each impression factor of the target impression acquired by the design element acquisition unit 224. Then, it selects the top N color schemes with the smallest impression distance values. In this embodiment, the top two color schemes are selected. The color scheme impression table in Figure 15 is created in advance using the same method as the skeleton impression table and stored in the HDD 104. That is, the impression of each color scheme is estimated based on the generated poster image, with the design elements other than the color scheme, such as the skeleton, font, and image, fixed. This makes it possible to create a table of the impression trends of the color schemes. The method for selecting a color scheme that matches a specified color is to select a color that is close to the color included in the color scheme. For example, if a color with a distance ΔE of 2.0 or less in the CIE L*a*b* color space is included, that color is selected. Alternatively, a criterion of a distance ΔRGB within the RGB color space of 1.0 or less may be used. Specifically, if the specified color is (R,G,B)=(0,67,69), then in the color scheme pattern impression table in Figure 15, the distance ΔRGB between color 1 (color ID "1") and the specified color is 1.0 or less, so color 1 is selected.

[0183] In S1011, the background image scaling unit 218 enlarges or reduces the background image acquired by the design element acquisition unit 214 for each product, changing its size. The background image scaling unit 218 then links the scaled background image to the product and outputs it to the layout unit 221.

[0184] Figure 10(d) is a flowchart illustrating the scaling process performed by the background image scaling unit 218 in S1011. The scaling process of the background image will be explained in detail with reference to Figure 10(d).

[0185] In S1041, the background pattern scaling unit 218 determines the scaling ratio for scaling the background pattern. Figure 16 illustrates an example of scaling the background pattern. Figure 16 illustrates an example where a flyer is specified as the key design and a postcard is specified as the product to be created. Figure 16(a) shows the flyer specified as the key design. Flyer 1601 uses a checkered pattern as the background pattern 1602. Figure 16(b) is an example of a postcard generated by the product generation unit 210. Postcard 1603 has a background pattern 1604 that has been reduced by the background pattern scaling unit 218. Note that the reduction is based on the background pattern 1602 of the key design. Figures 16(c), 16(d), and 16(e) illustrate the differences in the scaling ratio of the background pattern, and each shows a postcard with a different scaling ratio of the background pattern. The scaling ratio refers to the scaling ratio of background images 1606, 1608, and 1610 on postcards 1605, 1607, and 1609 relative to background image 1602 on flyer 1601, which is the key design.

[0186] The scaling ratio of the background image 1606 of postcard 1605 shown in Figure 16(c) corresponds to the size ratio of postcard 1605 to flyer 1601, in this case, the area ratio.

[0187] Furthermore, the scaling ratio of the background image of postcard 1607 shown in Figure 16(d) is 1, which is the same size as the background image 1602 of flyer 1601. In other words, postcard 1607 has the background image 1608 which is not scaled.

[0188] The scaling ratio of the background image 1604 shown in Figure 16(b) above, which has been reduced by the background image scaling unit 218, is set to a value between the scaling ratio of the background image 1606 in Figure 16(c) and the scaling ratio of the background image 1608 in Figure 16(d) (=1). The scaling ratio R is determined by the background image scaling unit 218, for example, according to the following formula (8).

[0189] R = D1 ÷ D2 × C2 ÷ C1 ... (8)

[0190] Here, let C1 be the length of the diagonal line of the key design, and C2 be the length of the diagonal line of the commercial material to be created. Also, let D1 be the assumed observation distance of the key design, and D2 be the assumed observation distance of the commercial material to be created. Note that Equation (8) corresponds to the above-mentioned Equations (2) and (3).

[0191] In Equation (8), for two commercial materials of different sizes, the scaling factor R corresponding to their respective assumed observation distances and the size of the commercial material itself is determined. Therefore, the visual similarity between the two commercial materials increases, and it becomes easier to represent a common brand compared to FIGS. 16(c) and (d). For example, in the creation condition specifying unit 201, assume that the assumed observation distance of an A4-size flyer (297 mm × 210 mm) is specified as 40 cm, and the assumed observation distance of a postcard of the size of a postcard (148 mm × 100 mm) is specified as 30 cm. In this case, the scaling factor R calculated by Equation (8) is 0.65. Note that the scaling factor Re of the background pattern 1608 in FIG. 16(d) is equal magnification (= 1), and the scaling factor Rs of the background pattern 1606 in FIG. 16(c) is the commercial material size ratio (the ratio of the lengths of the diagonal lines of each commercial material) Rs. These are represented by the following Equations (9) and (10).

[0192] Re = 1 …(9) Rs = ((148×100)÷(297×210))^0.5 = 0.49 …(10)

[0193] The scaling factor R in FIG. 16(b) is larger than the commercial material size ratio Rs (the ratio of the lengths of the diagonal lines of each commercial material) and smaller than the same ratio Re (= 1) as the size of the background pattern of the key design. That is, the relationship is Rs < R < Re.

[0194] Thus, when the size of the created commercial material is smaller than the size of the reference commercial material (key design), the scaling factor R is larger than the commercial material size ratio Rs (the ratio of the lengths of the diagonal lines of each commercial material) and smaller than the same ratio Re (= 1) as the size of the background pattern of the key design. That is, the relationship is Rs < R < Re. As a result, the size of the background pattern of the created commercial material (second production data) is smaller than the size of the background pattern of the key design (first production data), and the ratio C of the size of the created commercial material to the key design12 It becomes larger than the size obtained by reducing the background pattern of the key design.

[0195] In general, when the size of the commercial material is small, the observation distance is assumed to be short, and when the size of the commercial material is large, the observation distance is assumed to be long. For example, the observation distance of a business card is shorter compared to a poster of a large size. However, there are many cases where the ratio of the observation distance and the ratio of the commercial material size do not match. In such a case, when the background pattern 1606 equivalent to the commercial material size ratio (Fig. 16(c)) is arranged on the commercial material, the pattern is recognized as finer compared to the key design from the user who visually recognizes the commercial material. Conversely, when the background pattern 1608 equivalent to the same magnification (Fig. 16(d)) is arranged on the commercial material, the pattern is recognized as coarser compared to the key design from the user who visually recognizes the commercial material. With respect to such a sense of incongruity in appearance, by performing the magnification and reduction by the background pattern magnification / reduction unit 218 of the present embodiment, it is possible to set a magnification / reduction rate considering both the observation distance and the commercial material size, and the appearance of the background pattern of the commercial material can be made closer to the appearance of the key design.

[0196] In the example of Fig. 16, the method of magnifying and reducing the background in the case where the commercial material to be created is smaller than the commercial material of the key design has been described, but the same method can be applied when the commercial material to be created is larger than the commercial material of the key design. For example, assume that the key design is a flyer of A4 size and its observation distance is 40 cm, and the commercial material to be created is A1 size (594 mm × 841 mm) and its observation distance is 100 cm. In this case, the magnification / reduction rate R calculated by Equation (8) is 2.5. Comparing this with the commercial material size ratio Rs = 2.83 between A4 and A1, the relationship of Re < R < Rs holds.

[0197] Thus, when the size of the created commercial material is larger than the size of the reference commercial material (key design), the magnification / reduction rate R is smaller than the commercial material size ratio Rs (the ratio of the diagonal lengths of each commercial material) and larger than the same ratio Re (= 1) as the size of the background pattern of the key design. That is, the relationship of Re < R < Rs is established. As a result, the size of the background pattern of the created commercial material (second production object data) is larger than the size of the background pattern of the key design (first production object data), and the ratio C of the size of the created commercial material to the key design12 This results in a size smaller than the enlarged background image of the key design.

[0198] Note that in equation (8), C1 and C2 are the lengths of the diagonals, but they do not have to be diagonal lengths as long as they represent the size of the product. For example, C1 and C2 could be the lengths of the sides of the corresponding products, the square root of the area ratio, or the area ratio itself. The same applies to the equations shown below.

[0199] Furthermore, the method for determining the scaling factor of the background image is not limited to this. For example, the scaling factor R of the background image may be determined according to the following equation (11).

[0200] R = D1 ÷ D2 × F2 ÷ F1 ... (11)

[0201] Here, D1 is the assumed viewing distance for the key design, and D2 is the assumed viewing distance for the product being created. F1 is the maximum font size used in the key design, and F2 is the maximum font size for the product being created. By using font sizes regardless of the product size, the relative size ratio between the background image and the text can be considered. Therefore, for products where text is eye-catching, such as shop names or discount rates, the sense of consistency in the size of the text and background image can be improved. Note that F1 and F2 are set as the maximum font sizes, but they are not limited to these; the size of a representative font, such as the title font, may also be used.

[0202] Alternatively, the background image scaling unit 218 may determine the scaling ratio R of the background image according to the following equation (12).

[0203] R = D1 ÷ D2 ... (12)

[0204] Here, D1 is the assumed viewing distance for the key design, and D2 is the assumed viewing distance for the product being created. In this case, the scaling factor is determined solely by the viewing distance, so when a user views the product at their assumed viewing distance, the background image will appear to be the same size as the key design.

[0205] Furthermore, if the value of R calculated using the above-mentioned formulas (11) and (12) is not between the product size ratio and 1, it is excluded from the list of candidates for the scaling ratio R. The scaling ratio of the background image scaling section 218 is determined by a different method.

[0206] As yet another example, the background image scaling unit 218 may determine the scaling ratio R of the background image according to the following equation (13).

[0207] R=(1+C2÷C1)÷2...Equation (13)

[0208] Here, C1 is, for example, the length of the diagonal of the key design, and C2 is, for example, the length of the diagonal of the product to be created. In this case, the scaling factor R is determined as an intermediate value between the product size ratio and 1, regardless of the observation distance. For example, even if the assumed observation distance cannot be set to a specific value, a relatively stable design can be generated.

[0209] Furthermore, the background pattern scaling unit 218 may dynamically adjust the scaling ratio according to the arrangement of the background pattern. In Figure 16(e), the horizontal lines of the checkered pattern of the background pattern 1610 touch the upper and lower edges of the product, resulting in a different appearance from the design of the key design, the flyer 1601. If the background pattern scaling unit 218 detects such a design inconsistency, it may change the scaling ratio by a small amount ΔR increments to search for a scaling ratio at which the design inconsistency is no longer detected.

[0210] Furthermore, the background pattern scaling unit 218 may output the background pattern scaled at a plurality of scaling ratios R to the layout unit 221. For example, the background pattern scaling unit 218 outputs all the background patterns scaled at a scaling ratio R that satisfies the relationship of Re < R < Rs when enlarging and Rs < R < Re when reducing, among the scaling ratios R calculated by Expressions (8) to (12), to the layout unit 221. Thereafter, since the impression distance between the impression after layout and the target impression is evaluated in the merchandise selection unit 223, it becomes possible to select a background pattern closer to the target impression while maintaining the balance between the merchandise size and the background pattern size. The above is the explanation of the background pattern scaling ratio determination process performed in S1041. Return to the explanation of FIG. 10(d).

[0211] In S1042, the background pattern scaling unit 218 scales the background pattern acquired from the design element acquisition unit 214 at the scaling ratio determined in S1041. In S1043, the background pattern scaling unit 218 trims the background pattern scaled in S1042 to match the merchandise size. In S1044, the background pattern scaling unit 218 complements the portion of the background pattern trimmed in S1043 that is insufficient for the size of the background area of the merchandise.

[0212] FIG. 17 is a diagram for explaining the processes of S1042 to S1044. FIG. 17(a) shows the background pattern 1701 acquired from the design element acquisition unit 214. FIG. 17(b) is an example of the background pattern scaled in S1042, and the size of the background pattern 1703 is larger than the size of the merchandise 1702. FIG. 17(c) shows the background pattern 1704 created by trimming the background pattern 1703 in S1043, and the trimmed background pattern 1704 has the same size as the merchandise 1702. Since there is no shortage of the background pattern for the background area of the merchandise in FIG. 17(c), no processing is performed in S1044.

[0213] Figure 17(d) is an example of a background image scaled for a different product 1705 in S1041 than in Figure 17(b). The size of background image 1706 is larger vertically but smaller horizontally than the size of product 1705. Figure 17(e) shows background image 1707 created by cropping background image 1706 in S1042. Figure 17(f) shows background image 1708 created by complementing background image 1707 in S1044. Background image 1708 can be created by repeating background image 1707 horizontally. Alternatively, background image 1708 can be created by generating the area outside background image 1707 using the Generative AI's outpainting function. In that case, a complementary image can be obtained by masking the left and right edges of the product image 1705 shown in Figure 17(e) where background image 1707 is not assigned, and then performing outpainting.

[0214] In S1045, the background image scaling unit 218 determines whether the scaling processes in S1041 to S1044 have been completed for all the products to be created. If all processes are not completed, it returns to S1041. If all processes are completed, this flowchart is terminated and the process proceeds to S1012 in Figure 10(a). This concludes the explanation of the background image scaling process in S1011. Return to the explanation of Figure 10(a).

[0215] In S1012, the logo selection unit 219 selects a combination of logos that matches the target impression from the logo list acquired by the design element acquisition unit 214. The method for selecting a combination of logos that matches the target impression is the same as in S1009. That is, the logo selection unit 219 refers to a logo impression table that shows the impression value corresponding to each logo and selects a logo according to the target impression.

[0216] Furthermore, in S1012, the font selection unit 220 selects a font combination that matches the target impression acquired by the design element acquisition unit 214. The method for selecting a font combination that matches the target impression is the same as in S1009, by referring to the impression table corresponding to the fonts and selecting a font according to the target impression.

[0217] Figure 18 shows an example of a font impression table that links fonts to impressions. The font selection unit 220 calculates the impression distance from the values ​​of each impression factor shown in the columns from "luxury" to "gravitas" in Figure 18, and the values ​​of each impression factor of the target impression. Then, it selects the top N fonts with the smallest impression distance values. The font impression table in Figure 18 is created in advance using the same method as the skeleton impression table and stored in the HDD 104. That is, the impression of each font is estimated based on poster images generated by changing the font while fixing design elements other than the font, such as the skeleton, color scheme, and image. This makes it possible to create a table of font impression trends. The method for selecting a font combination that matches a specified font is to select a font combination that contains the specified font. Furthermore, it is also possible to select a font combination that is close to the impression value of the font in the font combination that contains the specified font. This makes it possible to select fonts that are close to the impression of the specified font.

[0218] In S1013, the layout unit 221 sets text information, images, color schemes, fonts, background patterns, and logos for the skeleton selected by the skeleton selection unit 216, and generates a design for the product.

[0219] The processing of S1013 and the software configuration of the layout unit 221 will be explained in detail using Figures 19, 20, 21, and 22.

[0220] Figure 19 is an example of a software block diagram that explains the layout unit 221 in detail. The layout unit 221 includes a color assignment unit 1901, an image placement unit 1902, an image correction unit 1903, a font setting unit 1904, a text placement unit 1905, a text decoration unit 1906, a background pattern setting unit 1907, a logo placement unit 1908, and a background illustration setting unit 1909.

[0221] Figure 20 is a flowchart illustrating the layout processing of S1013 in detail. Figure 21 is a diagram illustrating the information input to the layout unit 221. Figure 21(a) is a table summarizing the input character information, image data 2101, background pattern, and logo. Figure 21(b) is an example of a table showing the color scheme obtained from the color scheme pattern selection unit 217, and Figure 21(c) is an example of a table showing the font obtained from the font selection unit 220. Figure 22 is a diagram illustrating the processing steps of the layout unit 221.

[0222] First, the layout process will be explained in detail using Figure 20. Steps S2001 through S2011 are executed in step S1013.

[0223] In S2001, the layout unit 221 combines and enumerates all the skeletons obtained from the skeleton selection unit 216, the color patterns obtained from the color pattern selection unit 217, the background patterns obtained from the background pattern scaling unit 218, and the fonts obtained from the font selection unit 220. The layout unit 221 generates design data for the product by sequentially performing subsequent layout processing for each combination. For example, if 3 skeletons are obtained, 2 color patterns are obtained, 3 background patterns are obtained, and 2 fonts are obtained, then 3 × 2 × 3 × 2 = 36 combinations are enumerated. The layout unit 221 then executes the processing in S2002 to S2011 on one of the enumerated combinations.

[0224] In S2002, the color assignment unit 1901 assigns a color pattern obtained from the color pattern selection unit 217 to a skeleton obtained from the skeleton selection unit 216. Figure 22(a) shows an example of a skeleton. In this embodiment, an example of assigning the color pattern with color ID "1" in Figure 21(b) to the skeleton 2201 in Figure 22(a) will be described. The skeleton 2201 in Figure 22(a) consists of two graphic objects 2202 and 2203, one image object 2204, four character objects 2205, 2206, and 2207, and a logo object 2210. First, the color assignment unit 1901 assigns colors to the graphic objects 2202 and 2203. Specifically, it assigns the corresponding colors from the color pattern based on the color number, which is metadata described in the graphic object. The graphic object 2202 is assigned color scheme "1", and the graphic object 2203 is assigned color scheme "2". Next, the color scheme assignment unit 1901 assigns a color scheme to a text object whose metadata is type and whose attribute is "Title".<type=Title> For the characters in the color scheme, for example, the last color of the color scheme pattern is assigned. That is, in this embodiment, color 4 is assigned to the characters placed in character object 2205. Next, for the characters placed in character objects 2206 and 2207, whose metadata is type and whose attribute is not "title", the character color is set based on the brightness of the background of the character object. In this embodiment, if the brightness of the background of the character object is below a threshold, the character color is set to white, and otherwise the character color is set to black. Figure 22(b) shows the state of skeleton 2208 after the above color scheme assignment process has been performed. The color scheme assignment unit 1901 outputs the colored skeleton data to the image placement unit 1902.

[0225] In S2003, the image placement unit 1902 places the image data obtained from the image analysis unit 212 onto the skeleton 2208 obtained from the color assignment unit 1901, based on the analysis information associated with the image data. In this embodiment, the image placement unit 1902 assigns the image data 2101 to the image object 2204 within the skeleton. Furthermore, if the aspect ratio of the image object 2204 and the image data 2101 are different, the image placement unit 1902 performs cropping to match the aspect ratio of the image data 2101 to that of the image object 2204. More specifically, based on the object position obtained by the image analysis unit 212 analyzing the image data 2101, the image placement unit 1902 performs cropping to minimize the object area reduced by cropping. Note that the cropping method is not limited to this, and other cropping methods may be used, such as cropping the center of the image or devising a composition so that the face position forms a triangular composition. The image placement unit 1902 outputs the skeleton data with image assignments to the image correction unit 1903.

[0226] In S2004, the image correction unit 1903 obtains the image-assigned skeleton data from the image placement unit 1902 and corrects the images placed on the skeleton. In this embodiment, if the image resolution is insufficient, upsampling processing is performed using super-resolution processing. First, the image correction unit 1903 determines whether the images placed on the skeleton meet a certain resolution. For example, suppose a 1600px × 1200px image is assigned to a 200mm × 150mm area on the skeleton. In this case, the print resolution of the image can be calculated using equation (14).

[0227]

number

[0228] Next, the image correction unit 1903 increases the resolution by super-resolution processing if it determines that the print resolution of the image is below a threshold. On the other hand, if the print resolution of the image is above the threshold and it is determined that there is sufficient resolution, no image correction is performed. In this embodiment, super-resolution processing is performed when the print resolution of the image is less than 300 dpi.

[0229] In S2005, the font setting unit 1904 sets the font obtained from the font selection unit 220 to the image-corrected skeleton data obtained from the image correction unit 1903. Figure 21(c) shows an example of a font combination selected by the font selection unit 220. In this embodiment, an example of assigning a font when the font to be assigned to the image-corrected skeleton data is font ID "2" in Figure 21(c) will be described. In this embodiment, the font is set for the character objects 2205, 2206, and 2207 of the skeleton 2209. When the product is a poster, it is common to set a font that stands out for the title from the perspective of attracting attention, and a font that is easy to read from the perspective of visibility for the other characters. Therefore, when the product to be created is a poster, the font selection unit 220 selects two types of fonts: a title font and a body font. The font setting unit 1904 sets the title font for the character object 2205 whose attribute is "title", and the body font for the other character objects 2206 and 2207. The font setting unit 1904 outputs the skeleton data with the font settings already configured to the text placement unit 1905. In this embodiment, the font selection unit 220 selected two types of fonts, but this is not limited to this; for example, only the title font could be selected. In that case, the font setting unit 1904 uses the font corresponding to the title font as the body font. That is, if the title is a Gothic font, then a representative Gothic font with high readability should be selected for the other text as well; if the title is a Mincho font, then a representative Mincho font should be selected for the other text as well, and so on. The body font should be set to match the type of title font. Of course, the title font and body font can be the same. Furthermore, fonts can be used differently depending on the degree to which you want to make something stand out, such as using the title font for the title and subtitle character objects and the body font for other character objects, or using the title font for fonts above a certain font size.

[0230] In S2006, the text placement unit 1905 places the text specified by the text specification unit 202 onto the font-configured skeleton 2209 obtained from the font-configured unit 1904. In this embodiment, each text shown in Figure 21(a) is assigned by referring to the metadata attributes of the character objects in the skeleton 2209. That is, "Summer Grand Sale," whose attribute is "Title," is assigned to character object 2205, and "Beat the Summer Heat," whose attribute is "Subtitle," is assigned to character object 2206. Since no body text is set, nothing is assigned to character object 2207. Figure 22(c) shows a skeleton 2209, which is an example of skeleton data after processing by the text placement unit 1905. The text placement unit 1905 outputs the text-configured skeleton 2209 to the text decoration unit 1906.

[0231] In S2007, the text decoration unit 1906 applies decorations to the character objects within the pre-placed text skeleton 2209 obtained from the text placement unit 1905. In this embodiment, if the color difference between the title character and its background area is below a threshold, the title character is given an outline. This improves the readability of the title. The text decoration unit 1906 outputs the decorated skeleton to the background pattern setting unit 1907.

[0232] In S2008, the background pattern setting unit 1907 sets a background pattern on graphic objects determined to be background areas within the text-decorated skeleton 2209 obtained from the text decoration unit 1906. Graphic objects determined to be background areas are those with a color scheme number of "1". In Figure 22, graphic object 2202 is determined to be a background area, and the background pattern setting unit 1907 sets a background pattern on graphic object 2202. When setting a background pattern, the unit changes some of the colors within the background pattern to the color scheme assigned by the color scheme assignment unit 1901, and then sets the background pattern 2212 on graphic object 2202, which is the background area. Figure 22(d) is an example of a skeleton 2211 with a background pattern set. The background pattern setting unit 1907 outputs the skeleton 2211 with the background pattern 2212 set to the logo placement unit 1908.

[0233] In S2009, the logo placement unit 1908 places the logo on the logo object 2210 within the skeleton 2211, which has a background pattern already set, obtained from the background pattern setting unit 1907. Figure 22(d) shows an example of a logo already placed. If no logo is set in the creation condition specification unit 201, S2009 is skipped. The logo placement unit 1908 outputs the data of the skeleton 2211 with the placed logo as product data to the impression estimation unit 222.

[0234] In S2010, the background illustration setting unit 1909 places a background illustration in the background area of ​​the skeleton 2211 with the logo already placed, which is obtained from the logo placement unit 1908. The background illustration is, for example, the illustration specified in the image specification area 605 of the application startup screen 601, and is obtained by the image acquisition unit 211. If no background illustration is set, S2010 skips this step. The background illustration setting unit 1909 outputs the data of the skeleton with the background illustration placed, that is, the data of the skeleton with the layout completed, as product data to the impression estimation unit 222.

[0235] In S2011, the layout unit 221 determines whether product data has been generated for all combinations of the type of product being processed. If the layout unit 221 determines that product data with background images, logos, and background illustrations has been generated for all combinations of skeletons, color schemes, and fonts, it terminates the layout process and proceeds to S1011. If it determines that product data has not been generated for all created products, it returns to S2001 and generates product data for the combinations for which data has not yet been generated. This concludes the explanation of S1013. Return to the explanation of Figure 10(a).

[0236] In S1014, the impression estimation unit 222 performs rendering processing on each of the multiple product data generated in S1013 and acquired from the layout unit 221, and estimates the impression of the rendered product image. It then links the estimated impression with the product data. Rendering processing is the process of converting product data into image data. For example, even with the same color scheme, the arrangement changes if the skeleton is different, so the actual area occupied by each color will differ. Therefore, it is necessary to evaluate not only the impression tendencies of individual color schemes and skeletons, but also the final impression of the product, and this process is executed at this time. This makes it possible to evaluate not only the impression of individual elements of the product such as color scheme and arrangement, but also the final impression of the product laid out including images and text.

[0237] In S1015, it is determined whether the specified number of product data entries for a single product type have been generated. The number of entries is determined in S1005. In this embodiment, it is 5. If the required number of product data entries has not been generated, the process returns to S1009 to generate new product data. If the required number of product data entries have been generated, the process proceeds to S1016.

[0238] In S1016, it is determined whether the generation of product data has been completed for all product types. In this embodiment, it is set to generate 5 product data entries each for the two product types, postcards and flyers. In other words, 10 product data entries are generated. If the generation of product data for the required number of items has not been completed for all product types, the product creation application changes the product type to be generated, returns to S1008, and executes the product data generation process for the newly set product type. If the generation of product data for the required number of items has been completed for all product types, the process proceeds to S1017.

[0239] In S1017, the product selection unit 223 selects a set of products to output to the display 105 from the estimated impression of each product image acquired from the impression estimation unit 222. Specifically, the product selection unit 223 first calculates the distance (impression distance) between the estimated impression associated with the product data and the target impression. The calculated impression distance is stored in association with the product data. In this embodiment, since 10 product data are generated, the product selection unit 223 calculates the impression distance for each of the 10 product data. The impression distance uses Euclidean distance. The smaller the value indicated by the Euclidean distance, the closer the target impression and the estimated impression are. The impression distance calculated by the product selection unit 223 is not limited to Euclidean distance; it is acceptable to use any method that can calculate the distance between vectors, such as Manhattan distance or cosine similarity.

[0240] Next, the product selection unit 223 calculates the total impression distance for each product set. The total impression distance is the sum of the impression distances of all product data included in the product set. For example, if a product set contains one of each of two types of product data, a flyer and a postcard, the total impression distance is calculated by adding up the estimated impression distances for each of these two types of product data. The calculated total impression distance is linked to the product set and stored in the HDD 104. By evaluating the total impression distance of a product set, the overall impression of products that are spread across multiple types can be brought closer to the intended target impression. Next, the product selection unit 223 selects a product set from the total impression distances linked to the product set. The number to be selected is determined, for example, based on the number of items displayed on the preview screen 701. Specifically, if the setting is to display one product set, the product selection unit 223 selects the product set with the smallest total impression distance. Furthermore, as shown in the preview screen 701C, if the system is configured to display multiple product sets, the product selection unit 223 selects product sets equal to the number of sets to be displayed, in order of increasing total impression distance. Alternatively, the product selection unit 223 may limit the product sets displayed on the preview screen to those whose total impression distance is less than or equal to a predetermined threshold. In this case, if the number of product sets is less than the number of sets to be displayed on the preview screen, the system may return to S1008 and generate further different product data.

[0241] In S1018, the generated product display unit 206 displays the preview screen 701 shown in Figure 7. Specifically, it renders the product data included in the product set selected by the product selection unit 223 in S1017 and outputs the product image to the display 105. The above is a description of the product generation process that generates other products from a key design specified by the user.

[0242] As described above, according to this embodiment, the product generation device 100 can generate product data that has a sense of unity with the key design and appropriately expresses the brand's impression. In particular, in this embodiment, when assigning the background pattern used in the key design to a product of a different size than the key design, the scaling ratio is limited to the specific range described above. This suppresses the inconsistency in the appearance of the background pattern due to changes in product size, making it possible to create products with a sense of unity across all product types. Furthermore, the product generation device 100 selects a product set from among multiple candidate product sets in which the total impression distance is smaller than a threshold, i.e., the product set is close to the target impression. Therefore, it is possible to generate a product set in which not only the impression of each individual design element, but also the impression of the product set as a whole is close to the target impression.

[0243] <<Modification of the First Embodiment>> As a variation of the first embodiment, another example of the preview screen will be described. In this example, the preview screen displays the generated product image and accepts settings for design elements and key design.

[0244] Figure 23 shows a preview screen 2301 in a modified example of the first embodiment. The preview screen 2301 shown in Figure 23 has a key design specification area 608, a design element specification area 615, a preview area 2302, an edit button 702, a print button 703, a save button 704, and an apply button 2310. Each part with the same reference numerals as the application startup screen 601 in Figure 6 operates in the same way as in Figure 6. Similarly, each part with the same reference numerals as the preview screen 701 in Figure 7 operates in the same way as in Figure 7. The operation of the key design specification area 608 and the design element specification area 615 in the preview screen 2301 is the same as the key design specification area 608 and the design element specification area 615 shown in Figure 6. Furthermore, the operation of the edit button 702, print button 703, save button 704, previous candidate display button 705, and next candidate display button 706 is the same as the preview screen 701 in Figure 7. The preview screen 2301 is displayed by the generated product display unit 206 in S1018.

[0245] The key design selection area 608 is a UI that accepts the selection of key designs to be used in creating merchandise. The thumbnail 609 of the selected key design is displayed in the key design selection area 608. The "Add Key Design" button 610 is displayed.

[0246] The design element specification area 615 is a UI that accepts user specification of design elements for each design element. The design element specification area 615 includes a color scheme specification box 617, a background pattern specification box 620, a logo specification box 621, impression sliders (impression slider bars, or impression setting sliders) 624-627, and an influence level slider bar 629. It also includes radio buttons 623 to control the enabling / disabling of the settings for each design element, radio buttons 628 to control the enabling / disabling of each impression factor, and a checkbox 632 to enable the settings in the design element specification area 615.

[0247] The Reflect button 2310 is operated to reflect the information specified in the key design specification area 608 and the design element specification area 615 of the preview screen 2301.

[0248] The product generation process shown in Figure 10(a) is executed, and in S1018, the generated product images 2307 and 2308 are displayed in the preview area 2302 of the preview screen 2301 in Figure 23. In this modified example, the preview screen 2301 accepts changes to design elements and key designs while the product images 2307 and 2308 are displayed. The user changes the design elements and key designs by operating the key design specification area 608 and the design element specification area 615 while checking the product images 2307 and 2308 displayed in the preview area 2302. When the reflect button 2310 is pressed, the process moves to S1002. The product generation unit 210 executes the processes from S1002 to S1018 using the changed design elements and key designs. As a result, product images reflecting the key design and design element settings set on the preview screen 2301 are created, and the generated product display unit 206 displays the product images on the preview screen 2301.

[0249] This allows users to configure key designs and design elements while reviewing the generated product data. This eliminates the need to return to the app launch screen 601 after previewing and recreate the product with different settings, a cumbersome process. Users can efficiently obtain multiple product data sets with a consistent design.

[0250] <<Second Embodiment>> In the first embodiment, an example was described in which a product creation application determines a skeleton, color scheme, background pattern, and font that match a target impression specified by the user, based on design elements, content, key design, etc., that include the target impression, and generates a product. The product creation application of the second embodiment has a combination generation unit 2402. The combination generation unit 2402 searches for combinations of components that make the overall impression of the product close to the target impression based on a genetic algorithm. This allows for the more flexible generation of the optimal product that matches the target impression without the need for pre-processing such as creating a skeleton impression table, a color scheme impression table, and a font impression table.

[0251] <Software Block Diagram> Figure 24 is a software block diagram of the product creation application in the second embodiment. As shown in Figure 24, the product creation application in the second embodiment has a creation condition specification unit 201, a text specification unit 202, an image specification unit 203, a design element specification unit 204, a key design specification unit 205, a generated product display unit 206, and a product generation unit 210. The product generation unit 210 has an image acquisition unit 211, an image analysis unit 212, a multiple skeleton acquisition unit 2401, a design element acquisition unit 214, a color scheme pattern acquisition unit 215, a background pattern scaling unit 218, a combination generation unit 2402, a layout unit 221, an impression estimation unit 222, a combination selection unit 2403, and a product selection unit 223.

[0252] The difference from the product generation unit 210 of the first embodiment is that, instead of the skeleton selection unit 216, color scheme pattern selection unit 217, logo selection unit 219, and font selection unit 220 in Figure 2, it has a combination generation unit 2402 and a combination selection unit 2403. Also, instead of the skeleton acquisition unit 213, it has a multiple skeleton acquisition unit 2401. Components in Figure 24 that are denoted by the same reference numerals as in Figure 2 are the same as in the first embodiment, so their explanation is omitted.

[0253] The multiple skeleton acquisition unit 2401 acquires skeletons for the number of product types specified in the creation condition specification unit 201. In order to use the search algorithm, all skeletons for the specified product types are acquired and stored in RAM 103.

[0254] The combination generation unit 2402 acquires one or more skeletons for each product type from the multiple skeleton acquisition unit 2401. The combination generation unit 2402 also acquires the estimated product impression linked to the product data generated from the impression estimation unit 222. Furthermore, the combination generation unit 2402 acquires the target impression from the design element acquisition unit 214 and acquires the scaled background image from the background image scaling unit 218. In addition, the combination generation unit 2402 acquires the color scheme pattern list, font list, and logo list from the HDD 104. The combination generation unit 2402 generates combinations of product components (skeleton, color scheme pattern, font) and design elements (logo, background image) for each product type used to generate the product. As an example, in this embodiment, combinations of two types of products, postcards and flyers, are generated. These combinations of product types are defined as product sets. The combination generation unit 2402 outputs the generated product sets to the layout unit 221 for the number of sets created. The generation of combinations will be discussed later.

[0255] The combination selection unit 2403 selects a set of merchandise that is closest to the target impression from among the multiple merchandise sets generated by the combination generation unit 2402 and saves it to the RAM 103. That is, for the multiple generated merchandise sets, it selects a set of merchandise for which the impression distance value between the impression estimated by the impression estimation unit 222 and the target impression obtained from the design element acquisition unit 214 is less than or equal to a threshold value and saves it to the RAM 103. The combination selection unit 2403 determines whether the number of selected and saved merchandise sets has reached a predetermined number of creations, and repeats the generation of combinations and the selection of merchandise sets until the predetermined number of creations is reached. The predetermined number of creations may be a number set in advance or a number specified by the user. The merchandise selection unit 223 outputs the data of the selected merchandise set to the generated merchandise display unit 206.

[0256] <Product Generation Process> Figure 25 is a flowchart showing the product generation process executed by the product generation unit 210 of the product creation application in the second embodiment. The process shown in this flowchart starts when the user presses the launch icon of the product creation application, similar to the first embodiment. In Figure 25, processes that are denoted by the same reference numerals as in Figure 10 (product generation process of the first embodiment) are the same as in the first embodiment, so their explanation is omitted. In addition, in the product generation process of the second embodiment shown in this flowchart, S1010 (color scheme pattern selection) and S1012 (font / logo selection) shown in Figure 10 are omitted. After S1007 (skeleton acquisition), S2502 (product set table generation) and S2503 (combination generation) are added. Also, after the impression estimation process in S1014, evaluation value determination and termination determination (S2504~S2507) are added. Furthermore, the background pattern scaling process (S2501) is performed after the acquisition of the color scheme pattern in S1007. The following explanation will focus on the differences from the first embodiment.

[0257] In steps S1001-S1007, the product creation app retrieves settings from the app startup screen 601. Specifically, it retrieves the type of product to be created, category, content (text information, images), design elements (color scheme, background pattern, logo, font, and target impression), degree of reflection of design elements, and key design settings. The image acquisition unit 211 acquires the specified image data. The image analysis unit 212 analyzes the acquired image data to obtain image features and the main color of the image. The product generation unit 210 determines the type of product to be created and the number of each product type to be created. The design element acquisition unit 214 extracts design elements from the key design specified in the key design specification unit 205 as design elements to be used in creating the product. The color scheme pattern acquisition unit 215 acquires the color scheme pattern to be used in creating the product.

[0258] In S2501, the background image scaling unit 218 enlarges or reduces the background image acquired by the design element acquisition unit 214 according to the size or type of the product being created. The processing of the background image scaling unit 218 is the same as in the first embodiment. The scaled background image is linked to the product and output to the layout unit 221.

[0259] Next, the process proceeds to S1008. The multiple skeleton acquisition unit 2401 acquires skeletons for the number of product types specified in the creation condition specification unit 201. Specifically, the process from S1021 to S1026 in Figure 10(b) is repeated for the number of product types specified. All skeletons for the specified product types are acquired in order to use the search algorithm. For example, if postcards and flyers are set as products to be created on the application startup screen 601 in Figure 6, the multiple skeleton acquisition unit 2401 acquires all skeletons created for postcards and flyers.

[0260] Processes S2502, S2503, S1013, S1014, and S2504-S2506 are loop processes that are repeated for each product created. The following explanation will describe the initial operation and the operation from the second loop onward separately.

[0261] In the initial operation S2502, the combination generation unit 2402 acquires tables for skeleton, color scheme, font, logo, and background image to be used in product generation.

[0262] The tables used by the combination generation unit 2402 will be explained using Figure 26. Figure 26(a) shows a list of skeletons acquired by the combination generation unit 2402 from the multiple skeleton acquisition unit 2401. Figure 26(b) shows a list of fonts acquired by the combination generation unit 2402 from the design element acquisition unit 214. Figure 26(c) shows a list of logos acquired by the combination generation unit 2402 from the design element acquisition unit 214. Figure 26(d) shows a list of background patterns acquired by the combination generation unit 2402 from the background pattern scaling unit 218. The combination generation unit 2402 also acquires color patterns (Figure 5) acquired by the color pattern acquisition unit 215. In Figures 5 and 26(a) to (d), each table shows four design elements and four IDs, but the number is arbitrary. In the following explanation, we will assume, as an example, that we have obtained 115 skeleton IDs, 20 color scheme IDs, 21 font IDs, 35 logo IDs, and 8 background pattern IDs.

[0263] In S2503, the combination generation unit 2402 randomly generates combinations from the above five tables. These combinations are generated for each specified product material type. In this embodiment, 100 combinations are generated for each of the product material sets of product material 1 and product material 2. FIG. 26(e) shows the product material set table generated in this embodiment.

[0264] In the design element specifying unit 204, when a design element common to product materials is specified, the combination generation unit 2402 generates combinations so that the same ID is assigned to the product materials. Also, since product material 1 and product material 2 are of different types, and the product material sizes and estimated observation distances are different, different magnification ratios are determined by the background pattern scaling unit. That is, the magnification ratios of the background patterns of product material 1 and product material 2 are different from each other. In FIG. 26(e), background pattern IDs 1 to 4 are set for product material 1, and background pattern IDs 5 to 8 are set for product material 2. Further, when the reflectance is specified in the design element specifying unit 204, the ratio of having the same ID in product material 1 and product material 2 is determined based on the set value. Specifically, when the reflectance of the design element is specified as 0.3 (30%), among 100 combinations, 30 have the same ID between different types, and the remaining 70 are random combinations.

[0265] Thereafter, the combination generation unit 2402 executes layout (S1013), impression estimation (S1014), and impression value determination (S2504) for each of the generated product material sets of all combinations. The impression estimation can be performed by operating the deep learning model for impression estimation developed in the RAM 103 with the CPU 101 or the GPU 109, similar to the first embodiment. Note that not limited to deep learning, for example, when using a machine learning method such as a decision tree, a machine learning model for estimating an impression may be created by extracting feature amounts such as the luminance average value and the edge amount of the product material image by image analysis and estimating the impression based on the feature amounts.

[0266] In S2504, the combination selection unit 2403 calculates an evaluation value for each product set from the estimated impression obtained from the impression estimation unit 222 and associates it with each product set in the product set table. The evaluation value is the total impression distance as described in the first embodiment. Figure 27(a) is a table in which the evaluation value is associated with each row of the product set table shown in Figure 26(e). The evaluation value column in Figure 27(a) represents the evaluation value of the product set generated by the combination in the corresponding row. In S2503, the combination selection unit 2403 determines whether the evaluation value of the total impression distance is below a predetermined threshold. For product sets whose evaluation value of the total impression distance is below the predetermined threshold, the product set is saved to RAM 103 in S2505 and the process proceeds to S2505. Otherwise, S2505 is skipped and the process proceeds to S2506.

[0267] In S2506, the combination selection unit 2403 determines whether the determination of the total impression distance has been completed for all product sets in the product set table. If the determination of the total impression distance has been completed for all product sets, proceed to S2507. Otherwise, return to S2504.

[0268] In S2507, the combination selection unit 2403 determines whether the number of product sets stored in RAM 103 has reached a predetermined number. If the predetermined number has been reached, the process proceeds to S1017; otherwise, it returns to S2503.

[0269] In S2503 from the second loop onward, the combination generation unit 2402 generates a new product set table from the product set table in Figure 27(a) using new combinations. Figure 27(b) shows the newly generated product set table. In this embodiment, new combinations are generated using tournament selection and uniform crossover in a genetic algorithm. First, the combination generation unit 2402 randomly selects N combinations from the table in Figure 27(a). Here, for example, let's say N=3. Next, from the selected combinations, it selects the top two combinations with small evaluation values ​​(= close to the target impression and with high design similarity). Finally, in the two selected combinations, it randomly swaps each element of the combination (skeleton ID, color scheme ID, font ID, logo ID, background pattern ID) to generate new combinations. For example, combination IDs 1 and 2 in Figure 27(b) show the results generated from combination IDs 1 and 3 in Figure 27(a), where the color scheme ID has been swapped. By repeating the above procedure, 100 new combinations are generated, as shown in Figure 27(b).

[0270] This allows for efficient combination searching based on an evaluation value, which is the sum of the distances between the estimated impression of each product in the product set and the target impression. In this embodiment, 100 combinations were generated, but the number of combinations to be generated is not limited to this. Furthermore, while tournament selection and uniform crossover were used, other methods such as ranking selection, roulette selection, or single-point crossover may also be used. In addition, mutation may be incorporated to make it less likely to fall into local optima. Furthermore, while skeleton (arrangement), color scheme, font, logo, and background image were used as components of the product to be searched, other components may also be used. By increasing the number of components to be searched, a wider variety of products can be generated, and the range of impression expression can be increased.

[0271] In S2504 for the second loop and beyond, similar to S2503 in the first loop, the combination selection unit 2403 calculates an evaluation value for the total impression distance for each product set in the new product set table and associates it with each product set in the product set table. In S2505, the combination selection unit 2403 saves the product sets whose evaluation value for the total impression distance is below a predetermined threshold to RAM 103.

[0272] S2506~S2507 are the same as the first loop.

[0273] In S2507, the combination selection unit 2403 determines whether the number of product sets stored in RAM 103 has reached a predetermined number. If the predetermined number has been reached, the process proceeds to S1017.

[0274] If a predetermined number of product sets with evaluation values ​​below a threshold are stored, the combination selection unit 2403 may compare the evaluation values ​​of each stored product set and leave the product set with the smaller evaluation value in RAM 103 as the final evaluation result. In this case, product sets that are judged to have a larger evaluation value based on the comparison result may be deleted from RAM 103. Furthermore, although a combination search using a genetic algorithm was performed in this embodiment, the search method is not limited to this, and other search methods such as nearest neighbor search or tab search may be used.

[0275] In S1017, similar to the first embodiment, the product selection unit 223 selects a set of products to output to the display 105 (present to the user) from the estimated impression of each product image acquired from the impression estimation unit 222. In S1018, the generated product display unit 206 displays the preview screen 701. The preview screen 701 may include a key design specification area 608 and a design element specification area 615, as shown in Figure 23.

[0276] As described above, according to the second embodiment, the product creation application automatically generates combinations of constituent elements and design elements to be used in the product, and generates a product set by evaluating the products generated based on the combinations based on the target impression. This makes it possible to generate a product set with a design that is close to the target impression and has a sense of unity. This method is particularly effective when generating a product set in accordance with images and text information entered by the user. As an example, consider a case where the image specified by the user has a dynamic impression, but the user wants to generate a product set with a calm impression as a whole. According to the processing of the second embodiment, the overall impression of the product is estimated, and combinations of skeletons, color schemes, fonts, logos, and background patterns that are close to the target impression and have a high degree of design similarity are searched for. Therefore, the combination selection unit 2403 of the product creation application selects products that use skeletons with a small image area or products with fonts and color schemes that have a calmer impression in order to suppress the impression that the image has. In other words, the constituent elements used in the product are controlled in accordance with the image specified by the user. Furthermore, even if the impression intended by the user differs from the impression conveyed by the key design, the combination selection unit 2403 of the product creation application selectively uses a skeleton with a small placement area for the key design. Therefore, according to the second embodiment, it is possible to flexibly find the optimal combination of constituent elements and design elements that create a unified impression for the entire product set. In this way, product sets with a unified design can be created in various variations, and as a result, it becomes possible to generate product sets with a design that is close to the target impression and has a unified feel.

[0277] Preferred embodiments of the present disclosure have been described above with reference to the attached drawings, but the present disclosure is not limited to such examples. For example, the scaling of the background pattern by the background pattern scaling unit 218 may be determined not only by the illustrated formula, but also by other methods. For example, the scaling ratio may be predetermined for each combination of the source product (key design) and the destination product. Alternatively, an appropriate scaling ratio may be determined for each background pattern and maintained in association with the background pattern. Furthermore, it is clear to those skilled in the art that various modifications or alterations can be conceived within the scope of the disclosed technical idea, and these will naturally also fall within the technical scope of the present disclosure.

[0278] <<Other Embodiments>> The embodiments described above can also be realized by performing the following process: supplying software (programs) that realize the functions of the embodiments described above to a system or device via a network or various storage media, and having the computer (CPU, MPU, etc.) of that system or device read and execute the program. The program may be executed by one computer or by multiple computers working together. Furthermore, it is not necessary to realize all of the above processes in software; some or all of the processes may be realized by hardware such as an ASIC. Also, the CPU is not limited to one CPU that performs all the processing; multiple CPUs may work together as appropriate to perform the processing. Moreover, the functions of the embodiments described above are not only realized by the execution of the program code read by the computer. It is also possible that the OS running on the computer performs some or all of the actual processing based on the instructions of the program code, and the functions of the embodiments described above are realized through that processing.

[0279] The above-described embodiments include the following configurations.

[0280] (Composition 1) An information processing device that generates production data, Generating means for generating second production data that has background information of the first production data and is at least different in size from the first production data, The generating means, The information processing apparatus is characterized in that the background information of the first production data is enlarged or reduced so that the size ratio between the background information of the first production data and the background information of the second production data is a value between the ratio of the size of the first production data to the size of the second production data and 1 to generate the background information of the second production data.

[0281] (Configuration 2) When the size of the second production data is smaller than the size of the first production data, The size of the background information of the second production data is smaller than the size of the background information of the first production data and larger than the size obtained by reducing the background information of the first production data by the ratio of the size of the second production data to the first production data, When the size of the second production data is larger than the size of the first production data, The size of the background information of the second production data is larger than the size of the background information of the first production data and smaller than the size obtained by enlarging the background information of the first production data by the ratio of the size of the second production data to the first production data The information processing apparatus according to Configuration 1, characterized in that.

[0282] (Configuration 3) The generating means further determines the ratio of enlargement or reduction based on the ratio between the distance at which observation of the first production data is assumed and the distance at which observation of the second production data is assumed. The information processing apparatus according to Configuration 1 or Configuration 2, characterized in that.

[0283] (Configuration 4) The generating means further determines the ratio of enlargement or reduction based on the ratio between the font size of the representative text used for the first production data and the font size of the representative text used for the second production data. The information processing apparatus according to Configuration 1 or Configuration 2, characterized in that.

[0284] (Composition 5) The information processing apparatus according to any one of configurations 1 to 4, characterized in that the generation means determines the ratio of enlargement or reduction so that the impression of the second production data approaches the impression of the first production data.

[0285] (Composition 6) The information processing device according to any one of configurations 1 to 5, characterized in that the second production data is of a different type from the first production data.

[0286] (Composition 7) The information processing apparatus according to any one of configurations 1 to 6, characterized in that the generation means complements the background information which has been enlarged or reduced to match the size of the background area in the second production data.

[0287] (Composition 8) The information processing apparatus according to any one of configurations 1 to 6, characterized in that the generation means adjusts the enlargement or reduction of the background information so that there is no change in the pattern when the background information is placed in the background area of ​​the second production data.

[0288] (Composition 9) The information processing apparatus according to any one of configurations 1 to 8, characterized in that the generation means generates the second production data such that the distance between the impression estimated from the second production data and the target impression is less than a predetermined threshold.

[0289] (Composition 10) If the second production data includes multiple production data of different types, The information processing apparatus according to any one of configurations 1 to 8, characterized in that the generation means generates the second production data such that the sum of the distances between the impressions estimated from each of the multiple production data of different types and the target impression is less than a predetermined threshold.

[0290] (Composition 11) A means for receiving the specification of the first production data from the user, The system further comprises an acquisition means for acquiring design elements from the first production data received by the reception means, The information processing apparatus according to any one of configurations 1 to 10, characterized in that the generation means generates the second production data based on the design elements acquired by the acquisition means.

[0291] (Composition 12) The receiving means further receives the design element specification from the user, The information processing apparatus according to configuration 11, characterized in that the acquisition means mixes the design elements received by the receiving means with the design elements acquired from the first production data.

[0292] (Composition 13) The information processing apparatus according to configuration 11 or configuration 12, characterized in that the design element includes at least a target impression.

[0293] (Composition 14) The information processing device according to configuration 13, wherein the design element further includes at least one of the following: color scheme, background information, logo, and font.

[0294] (Composition 15) The receiving means further receives the user's specification of the degree to which the design elements should be reflected, The information processing apparatus according to any one of configurations 12 to 14, characterized in that the acquisition means mixes the design elements acquired from the first production data with the design elements received from the user based on the degree of reflection.

[0295] (Composition 16) The information processing device according to any one of configurations 11 to 15, characterized in that the receiving means receives a specification of the type of the second production data.

[0296] (Composition 17) The information processing device according to any one of configurations 11 to 16, further comprising a display means for displaying a first screen for receiving the specification of the first production data from the user.

[0297] (Composition 18) The information processing device according to configuration 17, wherein the first screen further receives the specification of the design elements from the user.

[0298] (Composition 19) The information processing device according to configuration 17 or configuration 18, wherein the first screen further receives a specification from the user for the type of the second production data.

[0299] (Composition 20) The aforementioned display means is An information processing device according to any one of configurations 17 to 19, characterized in that it displays a second screen containing the second production data generated by the generation means. (Composition 21) The information processing device according to configuration 20, wherein the second screen further includes an area for receiving the specification of the first production data from the user.

[0300] (Composition 22) The information processing apparatus according to configuration 20 or configuration 21, wherein the second screen further includes an area for receiving the design element specification from the user.

[0301] (Composition 23) The information processing device according to any one of configurations 11 to 22, characterized in that the generation means generates data for the second production by combining the design elements using a genetic algorithm.

[0302] (Composition 24) An information processing method performed by an information processing device, The process includes the step of generating a second production data which has background information of the first production data and is at least different in size from the first production data, An information processing method characterized in that the background information of the first production data is enlarged or reduced so as to generate the background information of the second production data, such that the size ratio between the background information of the first production data and the background information of the second production data is a value between the ratio of the size of the first production data and the size of the second production data and 1.

[0303] (Composition 25) A program that causes a computer to execute an information processing method, The aforementioned information processing method is: The process includes the step of generating a second production data which has background information of the first production data and is at least different in size from the first production data, A program characterized in that the background information of the first production data is enlarged or reduced so that the size ratio of the background information of the first production data to the background information of the second production data is a value between the ratio of the size of the first production data to the size of the second production data and 1, thereby generating the background information of the second production data.

Claims

1. An information processing device that generates production data, The system includes a generation means that generates a second production data which has background information of the first production data and is at least different in size from the first production data, The generating means is An information processing device characterized by generating background information for the second production data by enlarging or reducing the background information for the first production data so that the size ratio between the background information for the first production data and the background information for the second production data is a value between the ratio of the size of the first production data and the size of the second production data and 1.

2. If the size of the second production data is smaller than the size of the first production data, The size of the background information of the second production data is smaller than the size of the background information of the first production data, and is larger than the size obtained by reducing the background information of the first production data in terms of the ratio of the size of the second production data to the size of the first production data. If the size of the second production data is larger than the size of the first production data, The size of the background information in the second production data is larger than the size of the background information in the first production data, and smaller than the size obtained by enlarging the background information of the first production data in terms of the ratio of the size of the second production data to the size of the first production data. The information processing apparatus according to feature 1.

3. The information processing apparatus according to claim 2, wherein the generation means further determines the magnification or reduction ratio based on the ratio of the distance at which the first production data is expected to be observed to the distance at which the second production data is expected to be observed.

4. The information processing apparatus according to claim 2, wherein the generation means further determines an enlargement or reduction ratio based on the ratio of the font size of the representative text used in the first production data to the font size of the representative text used in the second production data.

5. The information processing apparatus according to claim 1, characterized in that the generation means determines the ratio of enlargement or reduction so that the impression of the second production data approaches the impression of the first production data.

6. The information processing device according to claim 1, characterized in that the second production data is of a different type from the first production data.

7. The information processing apparatus according to any one of claims 3 to 6, characterized in that the generation means complements the background information which has been enlarged or reduced to match the size of the background area in the second production data.

8. The information processing apparatus according to any one of claims 3 to 6, characterized in that the generation means adjusts the enlargement or reduction of the background information so that there is no change in the pattern when the background information is placed in the background area of ​​the second production data.

9. The information processing apparatus according to claim 1, characterized in that the generation means generates the second production data such that the distance between the impression estimated from the second production data and the target impression is smaller than a predetermined threshold.

10. If the second production data includes multiple production data of different types, The information processing apparatus according to claim 1, wherein the generation means generates the second production data such that the sum of the distances between the impressions estimated from each of the multiple production data of different types and the target impression is less than a predetermined threshold.

11. A receiving means for receiving the specification of the first production data from the user, The system further comprises an acquisition means for acquiring design elements from the first production data received by the reception means, The information processing apparatus according to claim 1, characterized in that the generation means generates the second production data based on the design elements acquired by the acquisition means.

12. The receiving means further receives the design element specification from the user, The information processing apparatus according to claim 11, characterized in that the acquisition means mixes the design elements received by the receiving means with the design elements acquired from the first production data.

13. The information processing apparatus according to claim 11 or 12, characterized in that the design element includes at least a target impression.

14. The information processing apparatus according to claim 13, characterized in that the design element further includes at least one of the following: color scheme, background information, logo, and font.

15. The receiving means further receives the user's specification of the degree to which the design elements should be reflected, The information processing apparatus according to claim 12, characterized in that the acquisition means mixes the design elements acquired from the first production data with the design elements received from the user based on the degree of reflection.

16. The information processing device according to claim 11, characterized in that the receiving means receives a specification of the type of the second production data.

17. The information processing apparatus according to claim 11, further comprising a display means for displaying a first screen for receiving the specification of the first production data from a user.

18. The information processing apparatus according to claim 17, wherein the first screen further receives the specification of the design elements from the user.

19. The information processing apparatus according to claim 17 or 18, wherein the first screen further receives a specification from the user for the type of the second production data.

20. The aforementioned display means is The information processing apparatus according to claim 17, characterized in that it displays a second screen including the second production data generated by the generation means.

21. The information processing apparatus according to claim 20, wherein the second screen further includes an area for receiving the specification of the first production data from the user.

22. The information processing apparatus according to claim 20 or 21, wherein the second screen further includes an area for receiving the design element specification from the user.

23. The information processing device according to claim 1, characterized in that the generation means generates data for the second manufactured product by combining design elements using a genetic algorithm.

24. An information processing method performed by an information processing device, The process includes the step of generating a second production data which has background information of the first production data and is at least different in size from the first production data, An information processing method characterized in that the background information of the first production data is enlarged or reduced so as to generate the background information of the second production data, such that the size ratio between the background information of the first production data and the background information of the second production data is a value between the ratio of the size of the first production data and the size of the second production data and 1.

25. A program that causes a computer to execute an information processing method, The aforementioned information processing method is The process includes the step of generating a second production data which has background information of the first production data and is at least different in size from the first production data, A program characterized in that the background information of the first production data is enlarged or reduced so that the size ratio of the background information of the first production data to the background information of the second production data is a value between the ratio of the size of the first production data to the size of the second production data and 1, thereby generating the background information of the second production data.

Images