Information processing device, information processing method, and information processing program

Abstract

To allow a user to grasp physical characteristics that suit predetermined clothing.SOLUTION: An information processing apparatus includes: an estimation unit which estimates clothing that suits a predetermined target person, on the basis of a distributed representation space in which an image of the clothing and evaluator information are projected, the distributed representation space being generated based on evaluations on the image of the clothing made from a plurality of evaluators, the evaluations indicating whether the clothing suits an evaluator or not, and the evaluator information on the evaluator; and a providing unit which provides information on the clothing estimated by the estimation unit, to a user.SELECTED DRAWING: Figure 15

Description

Technical Field

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

Background Art

[0002] Conventionally, a technique for generating a distributed representation of input information and determining the relationship between information based on the comparison result between the generated distributed representations is known. As an example of such a technique, based on product information representing features of a product, word information representing the relationship between words, and learning data representing a product that has become an object of action according to a user's preference, a technique for estimating a hidden feature vector representing a position on a map space for each of the user and the product is provided.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, with the above-described technique, a user may not be able to grasp physical characteristics suitable for a predetermined piece of clothing.

[0005] For example, with the above-described technique, the hidden feature vector is merely estimated so that the distance between the user's hidden feature vector and the product's hidden feature vector becomes a distance reflecting the user's preference for the product, and a user may not be able to grasp physical characteristics suitable for a predetermined piece of clothing.

[0006] The present application has been made in view of the above, and an object thereof is to enable a user to grasp physical characteristics suitable for a predetermined piece of clothing.

Means for Solving the Problems

[0007] The information processing device according to the present application is characterized by comprising: an estimation unit that estimates clothing that suits a predetermined target person based on a distributed representation space generated based on evaluations from multiple evaluators indicating whether or not the clothing suits the evaluator, and evaluator information relating to the evaluator, for evaluating an image of clothing, and a provision unit that provides information about the clothing estimated by the estimation unit to a user. [Effects of the Invention]

[0008] According to one embodiment, the system has the effect of allowing users to understand the physical characteristics that suit a particular garment. [Brief explanation of the drawing]

[0009] [Figure 1] Figure 1 shows an example of the configuration of the information processing system 1 according to an embodiment. [Figure 2] Figure 2 is a diagram (1) showing an example of information processing according to the embodiment. [Figure 3] Figure 3 shows an example of a distributed representation space according to the embodiment. [Figure 4] Figure 4 is a diagram (2) showing an example of information processing according to the embodiment. [Figure 5] Figure 5 shows an example of the configuration of the information processing device 10 according to the embodiment. [Figure 6] Figure 6 shows an example of the evaluator information database 31 according to the embodiment. [Figure 7] Figure 7 shows an example of a user information database 32 according to the present invention. [Figure 8] Figure 8 is an example of the screen of user terminal 100, shown in Figure (1). [Figure 9] Figure 9 is Figure (2), which shows an example of the screen of user terminal 100. [Figure 10] Figure 10 is a flowchart (1) showing an example of the information processing procedure according to the embodiment. [Figure 11] FIG. 11 is a flowchart (2) showing an example of the procedure of information processing according to an embodiment. [Figure 12] FIG. 12 is a flowchart (3) showing an example of the procedure of information processing according to an embodiment. [Figure 13] FIG. 13 is a diagram showing a configuration example of an information processing system 1A according to an embodiment. [Figure 14] FIG. 14 is a diagram (3) showing an example of information processing according to an embodiment. [Figure 15] FIG. 15 is a diagram showing a configuration example of an information processing apparatus 10A according to an embodiment. [Figure 16] FIG. 16 is a diagram showing an example of an evaluator information database 31A according to an embodiment. [Figure 17] FIG. 17 is a diagram showing an example of a user information database 32A according to an embodiment. [Figure 18] FIG. 18 is a flowchart (4) showing an example of the procedure of information processing according to an embodiment. [Figure 19] FIG. 19 is a flowchart (5) showing an example of the procedure of information processing according to an embodiment. [Figure 20] FIG. 20 is a hardware configuration diagram showing an example of a computer that realizes the functions of the information processing apparatus 10A.

BEST MODE FOR CARRYING OUT THE INVENTION

[0010] Hereinafter, embodiments for implementing the information processing apparatus, information processing method, and information processing program according to the present application (hereinafter referred to as "embodiments") will be described in detail with reference to the drawings. Note that the information processing apparatus, information processing method, and information processing program according to the present application are not limited by this embodiment. Also, in the following embodiments, the same parts are denoted by the same reference numerals, and redundant explanations are omitted.

[0011] (First Embodiment) [1. Configuration of Information Processing System] First, the information processing system 1 according to the embodiment will be described. FIG. 1 is a diagram showing a configuration example of the information processing system 1 according to the embodiment. As shown in FIG. 1, the information processing system 1 includes an information processing device 10, a user terminal 100, and an evaluator terminal 200. The information processing device 10, the user terminal 100, and the evaluator terminal 200 are communicably connected by wire or wirelessly via a predetermined communication network (network N). Note that the information processing system 1 shown in FIG. 1 may include a plurality of information processing devices 10, a plurality of user terminals 100, and a plurality of evaluator terminals 200.

[0012] The information processing device 10 is an evaluation of an image of a wearer wearing a combination of a plurality of clothing items (also referred to as fashion items, including footwear (also referred to as shoes), hats (such as caps and hats), accessories (also referred to as accessories), etc.), and indicates whether the combination suits the wearer. It receives evaluations from a plurality of evaluators (annotators), and based on the received evaluations, generates a dispersion representation space obtained by projecting the image and the evaluator information indicating the evaluator, and is an information processing device that realizes information processing using the generated dispersion representation space, and is realized by, for example, a server device, a cloud system, or the like.

[0013] Also, for example, the information processing device 10 provides an e-commerce service that provides clothing (searching, selling, etc.). Further, the information processing device 10 provides a coordination service that receives submissions of content (images, videos, articles, etc.) indicating the coordination of clothing from users and provides (searches, distributes, etc.) them to other users.

[0014] Note that the information processing device 10 may have a function as a web server that provides a website related to the service. Further, the information processing device 10 may be a device that distributes information to be displayed on applications related to various services installed on the user terminal 100 to the information processing device 10. Further, the information processing device 10 may be a device that distributes the application data itself.

[0015] Furthermore, the information processing device 10 may function as a distribution device that distributes control information to the user terminal 100. Here, the control information is written using, for example, a scripting language such as JavaScript (registered trademark) or a stylesheet language such as CSS (Cascading Style Sheets). Note that the application itself distributed from the information processing device 10 may also be considered as control information.

[0016] The user terminal 100 is an information processing device used by the user. The user terminal 100 can be implemented as, for example, a smartphone, a tablet, a notebook PC (Personal Computer), a desktop PC, a mobile phone, or a PDA (Personal Digital Assistant). The user terminal 100 displays information distributed by the information processing device 10 and server devices providing predetermined services via a web browser or application. In the example shown in Figure 2, the user terminal 100 is a smartphone.

[0017] The evaluator terminal 200 is an information processing device used by evaluators to evaluate images of a wearer wearing multiple combinations of clothing. The evaluator terminal 200 can be implemented as, for example, a smartphone, tablet, notebook PC, desktop PC, mobile phone, or PDA. The evaluator terminal 200 also displays information distributed by the information processing device 10 and a server device that provides predetermined services, using a web browser or application. In the example shown in Figure 2, the evaluator terminal 200 is a smartphone.

[0018] [2. An example of information processing] Next, an example of information processing realized by the information processing device, etc., according to this embodiment will be described using Figures 2 to 4.

[0019] [2-1. Regarding the first information processing] The first information processing implemented by the information processing device, etc. according to this embodiment will be described below with reference to Figure 2. Figure 2 is Figure (1) showing an example of information processing according to the embodiment. In the following description, it is assumed that the user terminal 100 is used by a user (user U1) identified by the user ID "UID#1". In the following description, the user terminal 100 may be treated as the same as user U1. That is, in the following, user U1 can be read as user terminal 100.

[0020] Furthermore, in the following explanation, evaluator terminals 200-1 to 200-N (where N is any natural number) will be used depending on the evaluator using evaluator terminal 200. For example, evaluator terminal 200-1 is evaluator terminal 200 used by evaluator A1, identified by evaluator ID "AID#1". Also, in the following explanation, evaluator terminals 200-1 to 200-N will be referred to simply as evaluator terminal 200 without any particular distinction. In addition, in the following explanation, evaluator terminal 200 may be treated as the same as the evaluator. That is, in the following explanation, the evaluator can be read as evaluator terminal 200.

[0021] First, the information processing device 10 receives evaluations from the evaluator terminal 200 regarding images of a wearer wearing multiple clothing combinations (coordinates) (step Sa1). For example, the information processing device 10 receives evaluations indicating whether the wearer's coordinate suits the wearer or not. To give a specific example, the information processing device 10 presents the evaluator with a pair of images P1 and P2, compares the two images, and receives evaluations indicating which coordinate in the images suits the wearer and which coordinate in the images does not.

[0022] To give a more specific example, the information processing device 10 receives from evaluator A1 an evaluation that outfit #1 shown in image P1 suits wearer #1 shown in image P1, and an evaluation that outfit #2 shown in image P2 does not suit wearer #2 shown in image P2 (i.e., an evaluation of "suits" for image P1 and an evaluation of "does not suit" for image P2). In both examples, if the outfit suits the person, you can compare the evaluation that outfit #1 shown in image P1 suits wearer #1 shown in image P1 with the evaluation that outfit #2 shown in image P2 suits wearer #2 shown in image P2, and then determine which evaluation is relatively higher or lower. Conversely, if neither example suits the person, you can compare the evaluation that outfit #1 shown in image P1 does not suit wearer #1 shown in image P1 with the evaluation that outfit #2 shown in image P2 does not suit wearer #2 shown in image P2, and then determine which evaluation is relatively higher or lower. In the following examples, we will simply write "suits" including cases where it is judged to suit the person relatively, and similarly, we will simply write "does not suit" including cases where it is judged to not suit the person relatively. Furthermore, the information processing device 10 receives evaluations from evaluator A2 indicating that outfit #1 shown in image P1 does not suit wearer #1 shown in image P1, and that outfit #2 shown in image P2 suits wearer #2 shown in image P2 (i.e., an evaluation of "does not suit" for image P1 and an evaluation of "suits" for image P2). Similarly, the information processing device 10 also receives evaluations from other evaluators regarding the pair of images P1 and P2. In addition, the information processing device 10 receives evaluations from evaluators regarding other pairs such as the pair of images P3 and P4, the pair of images P5 and P6, the pair of images P1 and P3, the pair of images P2 and P4, and so on.

[0023] Next, the information processing device 10 generates a distributed representation space by projecting the image and evaluator information indicating the evaluator, based on the evaluator's evaluation of the image (step Sa2). For example, the information processing device 10 generates the distributed representation space using the VSE (Visual-Semantic Embedding) technique. To give a specific example, the information processing device 10 generates a distributed representation space F1 by projecting image P1 closer to evaluator information D1 indicating evaluator A1 who evaluated image P1 as "suitable" than to evaluator information D2 indicating evaluator A2 who evaluated image P1 as "unsuitable". Similarly, the information processing device 10 generates a distributed representation space F1 by projecting image P2 closer to evaluator information D2 indicating evaluator A2 who evaluated image P2 as "suitable" than to evaluator information D1 indicating evaluator A1 who evaluated image P2 as "unsuitable".

[0024] The information processing device 10 may also generate a distributed representation space onto which various information related to the image and evaluator information indicating various information related to the evaluator are projected. Here, an example of a distributed representation space generated by the information processing device 10 will be explained using Figure 3. Figure 3 is a diagram showing an example of a distributed representation space according to the embodiment.

[0025] For example, it is presumed that the clothing (items) included in the outfit shown in the image contribute to whether or not the evaluator feels that it suits the person. Therefore, the information processing device 10 generates a distributed representation space F2 on which the images of the clothing included in the outfit shown in the image are further projected. To give a specific example, the information processing device 10 generates a distributed representation space F2 on which the image of clothing I1 included in the outfit shown in image P1 is projected closer to the evaluator information D1 indicating evaluator A1 who evaluated image P1 as "suitable" than to the evaluator information D2 indicating evaluator A2 who evaluated image P1 as "unsuitable". Similarly, the information processing device 10 generates a distributed representation space F2 on which the image of clothing I2 included in the outfit shown in image P2 is projected closer to the evaluator information D2 indicating evaluator A2 who evaluated image P2 as "suitable" than to the evaluator information D1 indicating evaluator A1 who evaluated image P2 as "unsuitable".

[0026] Furthermore, for example, the information processing device 10 generates a distributed representation space F3 onto which it further projects information associated with the outfit shown in the image. To give a specific example, the information processing device 10 generates a distributed representation space F3 onto which it further projects categories indicating the purpose of the outfit and the situation in which it will be worn. To give a more specific example, the information processing device 10 generates a distributed representation space F3 onto which it projects the category T1 (casual) of the outfit shown in image P1, closer to the evaluator information D1 indicating evaluator A1 who evaluated image P1 as "suitable" than to the evaluator information D2 indicating evaluator A2 who evaluated image P1 as "unsuitable". Also, the information processing device 10 generates a distributed representation space F3 onto which it projects the category T2 (date) of image P2, closer to the evaluator information D2 indicating evaluator A2 who evaluated image P2 as "suitable" than to the evaluator information D1 indicating evaluator A1 who evaluated image P2 as "unsuitable". Furthermore, when the information processing device 10 generates a dispersed representation space by further projecting images of clothing included in the outfit shown in the image, it may also generate a distributed representation space by further projecting information associated with the clothing.

[0027] Furthermore, for example, the information processing device 10 generates a distributed representation space F4 onto which evaluator information, which represents the attribute information of the evaluators, is projected. Here, suppose that evaluator A1's attribute information is "late 20s, lives in Tokyo" and evaluator A2's attribute information is "early 20s, lives in Osaka". In such a case, the information processing device 10 generates a distributed representation space F4 onto which image P1 is projected closer to the attribute information of evaluator A1 (for example, attribute information B1 (late 20s, lives in Tokyo), attribute information B2 (late 20s), and attribute information B3 (lives in Tokyo)), who evaluated image P1 as "suitable", than to the attribute information of evaluator A2, who evaluated image P1 as "unsuitable". Furthermore, the information processing device 10 generates a distributed representation space F4 by projecting image P2 closer to the attribute information of evaluator A2 who evaluated image P2 as "suitable" (for example, attribute information B4 (early 20s, lives in Osaka), attribute information B5 (lives in Osaka), and attribute information B6 (early 20s)) than to the attribute information of evaluator A1 who evaluated image P2 as "unsuitable".

[0028] Furthermore, for example, the information processing device 10 generates a distributed representation space F5 onto which evaluator information, representing the evaluators' responses to questionnaires such as "Are you confident in your fashion sense?" and "What is your favorite city?", is projected. Here, suppose that evaluator A1 answers "I'm not confident in my fashion sense" and "I like Shibuya" to the above questionnaire, and evaluator A2 answers "I'm confident in my fashion sense" and "I like Harajuku". In such a case, the information processing device 10 generates a distributed representation space F5 onto which image P1 is projected closer to the responses of evaluator A1 who evaluated image P1 as "suitable" (for example, response C1 (I'm not confident in my fashion sense, I like Shibuya), response C2 (I'm not confident in my fashion sense), and response C3 (I like Shibuya)) than to the responses of evaluator A2 who evaluated image P1 as "not suitable". Furthermore, the information processing device 10 generates a distributed representation space F5 in which image P2 is projected closer to the responses of evaluator A2 who evaluated image P2 as "suitable" (for example, response C4 (confident in fashion, likes Harajuku), response C5 (confident in fashion), and response C6 (likes Harajuku)) than to the responses of evaluator A1 who evaluated image P2 as "unsuitable".

[0029] The distributed representation space generated by the information processing device 10 is not limited to the above example. For example, the information processing device 10 may generate a distributed representation space that projects images, images of clothing included in the coordination, the coordination category, the evaluator's attribute information, and the evaluator's responses to the questionnaire.

[0030] Furthermore, the information included in the distributed representation space generated by the information processing device 10 is not limited to the examples described above, and any information may be projected. For example, the information processing device 10 may generate a distributed representation space onto which information indicating the provider of the clothing (e.g., a brand or a shop that sells clothing) is projected. To give a specific example, the information processing device 10 generates a distributed representation space onto which information indicating the provider of the clothing included in the outfit shown in image P1 is projected, closer to the evaluator information D1 indicating evaluator A1 who evaluated image P1 as "suitable" than to the evaluator information D2 indicating evaluator A2 who evaluated image P1 as "unsuitable". Similarly, the information processing device 10 generates a distributed representation space onto which information indicating the provider of the clothing included in the outfit shown in image P2 is projected, closer to the evaluator information D2 indicating evaluator A2 who evaluated image P2 as "suitable" than to the evaluator information D1 indicating evaluator A1 who evaluated image P2 as "unsuitable". Furthermore, for example, the information processing device 10 may generate a distributed representation space onto which the wearer's physical information (e.g., features related to skeleton and physique) is projected, and the method described in Non-Patent Literature 1 (ViBE: Dressing for Diverse Body Shapes. Wei-Lin Hsiao, Kristen Grauman, Proceedings of the IEEE / CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020, pp. 11059-11069.) may be used when extracting physical features.

[0031] Furthermore, for example, the information processing device 10 may generate a distributed representation space that projects various attribute information of the evaluator (e.g., occupation, height, weight, place of residence, personal color, income, rent), responses to the evaluator's questionnaire items (e.g., favorite provider, amount of money spent on fashion, media used for fashion reference (magazines frequently read, web services frequently used, etc.), things they are passionate about (hobbies, etc.), future dreams, people they look to for fashion inspiration (celebrities, influencers, etc.), desired image, items they want to incorporate, information on the social media of people they like (admire), body shape concerns, fashion concerns, etc.), and information linked to the outfit shown in the image (e.g., item type, provider, usage scene, color, season, purpose of wearing, selling point, material, size, price range, etc.).

[0032] Returning to Figure 2, let's continue the explanation. Next, in response to a request for coordination from the user terminal 100 via an e-commerce service or coordination service, the information processing device 10 estimates a coordination that suits user U1 (in other words, a coordination that is estimated to be highly rated by the user, or a coordination that is estimated to be considered suitable by the user) based on the generated distributed representation space and information about user U1 (step Sa3). Here, let's assume that user U1's attribute information is "late 20s". In this case, the information processing device 10 estimates that the coordination shown by the image projected within a predetermined range from the attribute information B2 in the distributed representation space F4 is a coordination that suits user U1.

[0033] Furthermore, the information processing device 10 may estimate the position of the information about user U1 when it is projected onto the distributed representation space, based on the information projected onto the distributed representation space and the information about user U1, and estimate an outfit that suits user U1 based on the estimated position. For example, if user U1's attribute information is "late 20s, lives in Osaka," the information processing device 10 may estimate that the position midway between attribute information B2 and B5 in the distributed representation space F4 is the position when the information about user U1 is projected onto the distributed representation space F4, and estimate that the outfit shown by the image projected within a predetermined range from that position is an outfit that suits user U1.

[0034] Here, in the example in Figure 2, the information processing device 10 estimates that Coordinate #1 shown in image P1, Coordinate #3 shown in image P3, and Coordinate #4 shown in image P4 are coordinates that suit user U1. In such a case, the information processing device 10 provides information about coordinates #1, #3, and #4 to the user terminal 100 (step Sa4). For example, the information processing device 10 estimates that the information about the coordinate shown in the image closer to the attribute information B2 in the distributed representation space F4 is more likely to suit user U1, and provides it with a higher priority. To give a specific example, the information processing device 10 provides information about coordinates #1, #3, and #4, as well as information about the clothing included in coordinates #1, #3, and #4, in an e-commerce service. The information processing device 10 also provides information about coordinates #1, #3, and #4, as well as information about the clothing included in coordinates #1, #3, and #4, in a coordination service. In this case, the information processing device 10 may display information with higher priority in a more prominent position or higher in the rankings.

[0035] The information processing device 10 may also provide information on coordinations #1, #3, and #4 that correspond to information about user U1 (attribute information, physical information (height, weight, body type, bone structure, hairstyle, hair color, skin color, eye color, etc.)) and information related to coordination (for example, the current season or current fashion trends). For example, if user U1 is male, the information processing device 10 will provide information on coordinations #1, #3, and #4 that are suitable for men. For example, if user U1 is male, the information processing device 10 may also provide information on coordinations #1, #3, and #4 that are suitable for male wearers or that include clothing suitable for men. For example, if the current season is autumn, the information processing device 10 will provide information on coordinations #1, #3, and #4 that are suitable for autumn or winter.

[0036] Furthermore, the information processing device 10 may provide coordination information according to the conditions specified by user U1. For example, if user U1 specifies the category "casual" as a condition, the information processing device 10 will provide coordination #1, #3, and #4 with a higher priority given to the coordination indicated by the image projected closer to category T1 in the distributed representation space (e.g., distributed representation space F3).

[0037] Furthermore, the information processing device 10 may provide coordination information using a model that has learned the user U1's clothing preferences. For example, the information processing device 10 may train a model so that when information about clothing (e.g., an image) is input, the higher the degree to which user U1 likes that clothing, the higher the score it outputs. The information processing device 10 then inputs information about coordination #1, #3, and #4 into the model and provides information about coordination #1, #3, and #4 to user U1 in order of the highest output scores.

[0038] Next, the information processing device 10 obtains feedback from the user terminal 100 indicating whether or not user U1 had a positive reaction to the information provided (step Sa5). For example, in the e-commerce service, the information processing device 10 obtains information (feedback) indicating whether or not user U1 purchased clothing included in coordinates #1, #3, and #4, or whether or not they applied for rental. In addition, in the coordination service, the information processing device 10 obtains information (feedback) indicating whether or not user U1 registered coordinates #1, #3, and #4 as favorites. Note that the information indicating whether or not user U1 had a positive reaction to the information provided may also include whether or not user U1 viewed coordinates #1, #3, and #4 in the e-commerce service or coordination service, or whether or not user U1 viewed clothing included in coordinates #1, #3, and #4, and is not limited to the above examples.

[0039] Next, the information processing device 10 updates the distributed representation space based on feedback from user U1 (step Sa6). For example, if user U1 has a positive opinion of coordination #1 (for example, if they purchase clothing included in coordination #1 in an e-commerce service or register coordination #1 as a favorite in a coordination service), the information processing device 10 updates the distributed representation space F4 so that attribute information B2 and image P1 are brought closer together by a predetermined distance. On the other hand, if user U1 has a negative opinion of coordination #3 (for example, if they do not view the information for coordination #3 presented in an e-commerce service or coordination service), the information processing device 10 updates the distributed representation space F4 so that attribute information B2 and image P3 are moved further apart by a predetermined distance. Since the feedback from user U1 can also be interpreted as an evaluation of whether or not the image representing the coordination suits the user, evaluator information indicating the evaluator (user U1) who provided the evaluation (feedback) on the image may be further projected into the distributed representation space.

[0040] As described above, the information processing device 10 according to the embodiment generates a distributed representation space that projects an evaluation of whether or not an image showing a coordination is suitable, and evaluator information indicating the evaluator who performed the evaluation of the image.

[0041] Traditionally, whether an outfit "suits" the wearer has been subjective and difficult to quantify. However, according to the information processing device 10 of this embodiment, by configuring an image representing the outfit and a space that can measure the closeness to the evaluator's personality, it is possible to quantify whether an outfit "suits" the wearer, thus allowing each evaluator to understand whether the clothing suits the wearer based on their own subjective opinion.

[0042] Furthermore, the information processing device 10 according to the embodiment uses the generated distributed representation space to provide information about outfits that are estimated to suit the user, based on evaluations from evaluators similar to the user. As a result, the information processing device 10 according to the embodiment has the effect of enabling the user to understand what clothing suits a predetermined target (such as the user themselves).

[0043] [2-2. Regarding the second type of information processing] Next, a second information processing method implemented by the information processing device, etc., according to this embodiment will be described using Figure 4. Figure 4 is Figure (2), which shows an example of information processing according to this embodiment.

[0044] First, the information processing device 10 receives an evaluation from the evaluator terminal 200 of an image of a wearer wearing multiple combinations of clothing (coordinates) (step Sb1). For example, the information processing device 10 receives an evaluation indicating whether or not the coordinate #1 shown in image P1 suits the wearer #1 shown in image P1.

[0045] Next, the information processing device 10 generates a convolutional neural network (CNN) for generating distributed representations of evaluators and images based on the evaluators' evaluations of the images (step Sb2). For example, the information processing device 10 generates a CNN using training data that includes an image, evaluators' evaluations of the image, and evaluator information about the evaluators. To give a specific example, when an image is input, the information processing device 10 modifies the connection coefficients of the CNN so that it outputs a distributed representation similar to the distributed representation of evaluators who evaluated the image as "suitable" and a distributed representation that is not similar to the distributed representation of evaluators who evaluated the image as "unsuitable".

[0046] Here, if the information processing device 10 repeatedly trains the CNN based on the evaluations of each evaluator, the CNN will estimate the range on which each evaluator made their evaluation of suitability or unsuitability, and generate distributed representations of each evaluator and the image according to the estimation results. For example, when generating a distributed representation from an image, the CNN divides the image into multiple regions and generates a vector representing the features of each divided region (hereinafter sometimes referred to as "region vector"). Then, the CNN integrates the multiple region vectors to generate a distributed representation that represents the features of the entire image. The CNN repeatedly performs the same process for regions and features of various sizes to generate the final distributed representation. The entire image may also be considered as a single region. Furthermore, the distributed representation may be generated based on a single region instead of multiple regions. That is, if necessary, instead of generating a distributed representation based on partial regions on the image, a single feature vector considering the entire image may be calculated. Also, the definition of a region may be arbitrarily determined based on the application being used. For example, feature vectors obtained by feature extraction using a Fully CNN can be considered as each dimension corresponding to a region in an image. However, the decision of whether or not to use a region vector may be made based on the content in the image space corresponding to that region. Specifically, multiple regions in an image where a person exists can be identified using a human body detector, and these region vectors can be merged by calculating the Average Pooling of these region vectors. Another example is to identify regions by performing semantic region partitioning on the image. Specifically, for example, by identifying which parts of a pixel containing a person in an image include the top or bottom, multiple semantic regions can be obtained. By calculating region vectors corresponding to these regions and concatenating multiple region vectors in a specific order, a distributed representation showing the features of the entire image that is relatively robust to changes in the pose of the subject can be calculated.In this case, if there is a non-existent semantic region (for example, if the subject's feet are not imaged, the shoe region does not exist), the region vector in question may be replaced with an arbitrary feature vector. For example, the average vector of the shoe region vectors in the entire training data may be used, or a feature vector filled with zeros may be used.

[0047] When training a CNN in this manner, the information processing device 10 fixes the position of each evaluator and trains the CNN to output a distributed representation similar to the distributed representation of evaluators who evaluated an image as suitable (for example, a distributed representation based on evaluator information showing attribute information, etc.) for each image, and a distributed representation that is not similar to the distributed representation of the evaluator information of evaluators who evaluated an image as unsuitable. When this type of training is performed, the CNN adjusts connection coefficients and other parameters so that it generates region vectors that are relatively similar to the distributed representation of the evaluator information of evaluators who evaluated an image as suitable, and region vectors that are relatively not similar to the distributed representation of the evaluator information of evaluators who evaluated an image as unsuitable. When this training process is repeated, the CNN will eventually generate region vectors for a given region that are similar to the distributed representations of evaluators who are likely to have evaluated that region as suitable, and not similar to the distributed representations of evaluators who are likely to have evaluated that region as unsuitable.

[0048] Therefore, the information processing device 10 uses the region vectors generated by such a trained CNN to estimate the regions that each evaluator judges to be suitable, and provides information indicating the estimated regions, such as a heat map.

[0049] For example, when the information processing device 10 provides an image showing a coordinated outfit to user U1 in an e-commerce service or a coordination service, it determines how well each piece of clothing shown in each region of the image suits user U1 based on the generated CNN and information about user U1 (step Sb3). Here, in the example in Figure 4, let's assume that image P1 is provided to user U1. In this case, the information processing device 10 identifies an evaluator whose distributed representation is similar to that of user U1 (for example, a distributed representation based on attribute information, etc.). The information processing device 10 then determines that the clothing shown in the region of image P1 whose region vector is highly similar to the distributed representation of the identified evaluator is more likely to suit user U1, and the clothing shown in the region with a low degree of similarity to the distributed representation of the evaluator is less likely to suit user U1.

[0050] Furthermore, the information processing device 10 generates for use in the above-mentioned processing, and is not limited to CNNs; any distributed representation (embedded representation) can be generated as long as it is possible to obtain one.

[0051] Next, the information processing device 10 outputs image P1 to the user terminal 100 in a manner corresponding to the degree determined in step Sb3 (step Sb4). For example, the information processing device 10 outputs an image that shows the degree to which the clothing represented by each region of image P1 suits user U1 using a heat map. To give a specific example, if the distributed representation of the evaluator identified in step Sb3 is similar to the region vector of the region representing the bottoms in image P1, but the region vector of the region representing the tops in image P1 is not similar, the information processing device 10 outputs image P1-1, in which the region representing the bottoms in image P1 is displayed in dark red and the region representing the tops in dark blue. Also, if the distributed representation of the evaluator identified in step Sb3 is similar to the region vector of the region representing the tops in image P1, but the region vector of the region representing the bottoms in image P1 is not similar, the information processing device 10 outputs image P1-2, in which the region representing the tops in image P1 is displayed in dark red and the region representing the bottoms in dark blue.

[0052] Furthermore, when outputting an image, the information processing device 10 may output the image along with information corresponding to the degree to which the clothing in each area suits the user U1. For example, when outputting image P1-1, the information processing device 10 outputs image P1-1 along with information suggesting the purchase of clothing included in the area shown in red, information suggesting that the user should avoid wearing clothing included in the area shown in blue, and information on other outfits that make use of the clothing included in the area shown in red.

[0053] As described above, the information processing device 10 according to the embodiment can visualize the areas of the outfit that the evaluator has evaluated as suitable or unsuitable, so that it is possible to understand which of the multiple outfit combinations suits the wearer. Furthermore, the information processing device 10 according to the embodiment allows the user to understand which outfits an evaluator with similar attributes to themselves has evaluated as suitable, so that the user can obtain guidance on what clothes to wear.

[0054] [3. Other processing examples] The above-described process is merely an example, and the information processing device 10 may perform various processes using various types of information. Examples of this are listed below.

[0055] [3-1. Estimation of coordination based on user information] In the example shown in Figure 2, the information processing device 10 may estimate an outfit that suits user U1 based on the distributed representation space and the category selected by user U1. For example, if the information processing device 10 receives a request for an outfit from user terminal 100 along with information indicating that user U1 has selected the category "casual", the information processing device 10 estimates that the outfit shown by the image projected within a predetermined range from attribute information B2 in the distributed representation space F4, and the outfit shown by the image projected within a predetermined range from category T1 in the distributed representation space F3, is an outfit that suits user U1 in the category "casual", and provides user terminal 100 with information about the outfit that corresponds to information about user U1 and information related to the outfit.

[0056] Furthermore, the information processing device 10 may estimate a suitable outfit for user U1 based on the distributed representation space and user U1's responses to the questionnaire. For example, if user U1 answers "I am not confident in my fashion sense" to a questionnaire such as "Are you confident in your fashion sense?", the information processing device 10 estimates that the outfit shown by the image projected from the response C2 within a predetermined range in the distributed representation space F5 is a suitable outfit for user U1, and provides the user terminal 100 with information about the outfit that corresponds to information about user U1 and information related to the outfit.

[0057] Furthermore, the information processing device 10 may make coordination suggestions based on information about user U1. For example, if user U1 answers "I'm not confident in my fashion sense" to a questionnaire question such as "Are you confident in your fashion sense?", the information processing device 10 will suggest a coordination shown by an image projected within a predetermined range from a predetermined position (for example, an intermediate position) between responses C2 and C5 in the distributed representation space F5, along with a comment such as "Why not try this to gain confidence in your fashion sense?". In other words, the information processing device 10 may make coordination suggestions based on information about user U1's future.

[0058] [3-2. Estimation of coordination based on information about the target individuals] In the example shown in Figure 2, the information processing device 10 may estimate an outfit that a target person would consider suitable for user U1, based on the distributed representation space and information about a target person having a predetermined relationship with user U1 (for example, someone user U1 meets). For example, the information processing device 10 may estimate an outfit that a target person would consider suitable for user U1, based on the distributed representation space and the target person's attributes. To give a specific example, if the target person's attribute information is "early 20s," the information processing device 10 estimates that the outfit shown by the image projected within a predetermined range from the attribute information B6 in the distributed representation space F4 is an outfit that the target person would consider suitable, and provides the user terminal 100 with information about the outfit that corresponds to information about user U1 and information related to the outfit.

[0059] Furthermore, for example, the information processing device 10 may estimate an outfit that the subject considers to suit them based on the distributed representation space and the subject's responses to the questionnaire. To give a specific example, if the subject answers "I am confident in my fashion sense" to a questionnaire such as "Are you confident in your fashion sense?", the information processing device 10 estimates that the outfit shown by the image projected from the response C5 within a predetermined range in the distributed representation space F5 is an outfit that the subject considers to suit them, and provides the user terminal 100 with information about the outfit that corresponds to information about the user U1 and information related to the outfit.

[0060] Furthermore, the information processing device 10 may estimate the position of the information about the subject when it is projected onto the distributed representation space, based on the information projected onto the distributed representation space and the information about the subject, and based on the estimated position, estimate an outfit that the subject would consider suitable for the user. For example, if the subject's attribute information is "early 20s, lives in Tokyo," the information processing device 10 estimates that the position midway between attribute information B3 and B6 in the distributed representation space F4 is the position where the information about the subject is projected onto the distributed representation space F4. The information processing device 10 then estimates that the outfit shown by the image projected within a predetermined range from the estimated position is an outfit that the subject would consider suitable, and provides the user terminal 100 with information about the outfit that corresponds to the user's information and information related to the outfit.

[0061] This allows, for example, the system to estimate what outfits the target person would rate as suitable for the user, enabling the user to understand what to wear to make a good impression on the person they are about to meet.

[0062] Furthermore, if the target person provides information about an outfit that they evaluate as suitable for user U1, the information processing device 10 may obtain feedback from the target person regarding that outfit. For example, the information processing device 10 obtains information (feedback) from the user terminal 100 or a terminal device used by the target person indicating whether or not the target person had a positive opinion of user U1 wearing the outfit indicated by the provided information. If the target person had a positive opinion of user U1 wearing the outfit indicated by the provided information, the information processing device 10 updates the distributed representation space so that the information corresponding to the target person (e.g., attribute information or answers to a questionnaire) and the image showing the outfit are brought closer together by a predetermined distance. On the other hand, if the target person had a negative opinion of user U1 wearing the outfit indicated by the provided information, the information processing device 10 updates the distributed representation space so that the information corresponding to the target person and the image showing the outfit are moved further apart by a predetermined distance. Furthermore, since the feedback from the target person can also be interpreted as an evaluation of whether or not the image showing the outfit suits them, evaluator information indicating the evaluator (target person) who provided the evaluation (feedback) on the image may be projected into the distributed representation space.

[0063] Furthermore, the information processing device 10 may estimate, based on the distributed representation space, information about the user, and information about a target person having a predetermined relationship with the user, an outfit that the user evaluates as suitable for them and that the target person evaluates as suitable for the user. For example, the information processing device 10 estimates that an outfit shown by an image projected in the distributed representation space from information about the user to a predetermined range, and projected in the same range from information about the target person, is an outfit that both the user and the target person evaluate as suitable for them, and provides the user terminal 100 with information about the outfit that corresponds to information about the user and information related to the outfit. Alternatively, the information processing device 10 estimates that an outfit shown by an image projected in the distributed representation space from information showing the user's response to a questionnaire to a predetermined range, and projected in the same range from information showing the target person's response to a questionnaire, is an outfit that both the user and the target person evaluate as suitable for them, and provides the user terminal 100 with information about the outfit that corresponds to information about the user and information related to the outfit.

[0064] Furthermore, the information processing device 10 may estimate the position of information based on both the user and the target person when projected onto the distributed representation space, based on the information projected onto the distributed representation space, information about the user, and information about the target person. Based on the estimated position, it may estimate an outfit that the user evaluates as suitable for themselves and that the target person evaluates as suitable for the user. For example, if the user's attribute information is "late 20s" and the target person's attribute information is "early 20s," the information processing device 10 estimates that the position midway between attribute information B2 and B5 in the distributed representation space F4 is the position when information corresponding to both the user and the target person is projected onto the distributed representation space F4. The information processing device 10 then estimates that the outfit shown by the image projected within a predetermined range from that position is an outfit that both the user and the target person evaluate as suitable. The information processing device 10 then provides the user terminal 100 with information about the outfits that correspond to the estimated outfits, such as information about the user and information related to the outfits.

[0065] [3-3. Estimating suitable outfits for the subject] In the example shown in Figure 2, the information processing device 10 may estimate an outfit that the user U1 considers suitable for themselves based on the distributed representation space and information about a person who has a predetermined relationship with the user U1 (for example, the person to whom user U1 is giving a gift). For example, the information processing device 10 may estimate an outfit that the user considers suitable for themselves based on the distributed representation space and the user's attributes. To give a specific example, if the user's attribute information is "early 20s", the information processing device 10 estimates that the outfit shown by the image projected within a predetermined range from the attribute information B6 in the distributed representation space F4 is an outfit that the user considers suitable for themselves, and provides the user terminal 100 with information about the outfit that corresponds to the information about the user (attribute information, physical information (height, weight, body type, bone structure, hairstyle, hair color, skin color, eye color, etc.)) and information related to the outfit (for example, if the user is female, an outfit suitable for women).

[0066] Furthermore, for example, the information processing device 10 may estimate the outfit that the subject considers to suit them based on the distributed representation space and the subject's responses to the questionnaire. To give a specific example, if the subject answers "I am confident in my fashion sense" to a questionnaire such as "Are you confident in your fashion sense?", the information processing device 10 estimates that the outfit shown by the image projected within a predetermined range from the response C5 in the distributed representation space F5 is the outfit that the subject considers to suit them, and provides the user terminal 100 with information about the outfit that corresponds to information about the subject and information related to the outfit.

[0067] Furthermore, the information processing device 10 may estimate the position of the information about the subject when it is projected onto the distributed representation space, based on the information projected onto the distributed representation space and the information about the subject, and based on the estimated position, estimate an outfit that the subject would consider to suit them. For example, if the subject's attribute information is "early 20s, lives in Tokyo," the information processing device 10 estimates that the position midway between attribute information B3 and B6 in the distributed representation space F4 is the position where the information about the subject is projected onto the distributed representation space F4. The information processing device 10 then estimates that the outfit shown by the image projected within a predetermined range from the estimated position is an outfit that the subject would consider to suit them, and provides the user terminal 100 with information about the outfit that corresponds to the information about the subject and information related to the outfit.

[0068] Furthermore, if the subject provides information about an outfit that they evaluate as suiting them, the information processing device 10 may obtain feedback from the subject regarding that outfit. For example, when user U1 presents the subject with clothing included in the outfit indicated by the provided information, the information processing device 10 obtains information (feedback) indicating whether the subject was positive or negative from the user terminal 100 or the terminal device used by the subject. When user U1 presents the subject with clothing included in the outfit indicated by the information provided to user U1, and the subject is positive, the information processing device 10 updates the distributed representation space so that the information corresponding to the subject and the image showing the outfit are brought closer together by a predetermined distance. On the other hand, when user U1 presents the subject with clothing included in the outfit indicated by the information provided to user U1, and the subject is negative, the information processing device 10 updates the distributed representation space so that the information corresponding to the subject and the image showing the outfit are moved further apart by a predetermined distance. Furthermore, since feedback from the subjects can be interpreted as an evaluation of whether or not the outfit looks good in the image, evaluator information indicating the evaluator (subject) who provided the evaluation (feedback) on the image may be further projected into the distributed representation space.

[0069] Furthermore, the information processing device 10 may estimate, based on the distributed representation space, information about the user, and information about a target person having a predetermined relationship with the user, an outfit that the user evaluates as suitable for the target person, and that the target person evaluates as suitable for themselves. For example, the information processing device 10 estimates that an outfit shown by an image projected in the distributed representation space from information about the user to a predetermined range, and projected in the same range from information about the target person, is an outfit that both the user and the target person evaluate as suitable, and provides the user terminal 100 with information about the outfit that corresponds to information about the target person and information related to the outfit. Alternatively, the information processing device 10 estimates that an outfit shown by an image projected in the distributed representation space from information showing the user's response to a questionnaire to a predetermined range, and projected in the same range from information showing the target person's response to a questionnaire, is an outfit that both the user and the target person evaluate as suitable, and provides the user terminal 100 with information about the outfit that corresponds to information about the target person and information related to the outfit.

[0070] Furthermore, the information processing device 10 may estimate the position of information based on both the user and the target person when projected onto the distributed representation space, based on the information projected onto the distributed representation space, information about the user, and information about the target person. Based on the estimated position, it may estimate an outfit that the user evaluates as suitable for the target person, and that the target person evaluates as suitable for themselves. For example, if the user's attribute information is "late 20s" and the target person's attribute information is "early 20s," the information processing device 10 estimates that the position midway between attribute information B2 and B5 in the distributed representation space F4 is the position when information corresponding to both the user and the target person is projected onto the distributed representation space F4. The information processing device 10 then estimates that the outfit shown by the image projected within a predetermined range from that position is an outfit that both the user and the target person evaluate as suitable. The information processing device 10 then provides the user terminal 100 with information about the outfits that correspond to the estimated outfits, such as information about the target person and information related to the outfits.

[0071] [3-4. Estimation of outfit coordination based on images of wearers and users] In the example shown in Figure 2, the information processing device 10 may estimate a suitable outfit for a given target (user, subject, etc.) based on a distributed representation space onto which an image of the wearer's face is projected, and an image of the face of the target. For example, the information processing device 10 estimates that the outfit worn by a wearer whose facial features are similar to those of user U1 is a suitable outfit for the user, based on the distributed representation space F4 and the attribute information of user U1, among outfits #1, #3, and #4 that have been estimated.

[0072] Furthermore, the information processing device 10 may estimate a suitable outfit for a given target based on a distributed representation space onto which an image representing the wearer's body shape is projected, and an image representing the body shape of a predetermined target. For example, the information processing device 10 estimates that among outfits #1, #3, and #4 estimated based on the distributed representation space F4 and the attribute information of user U1, the outfit worn by a wearer whose body shape (e.g., slender, overweight) is similar to that of user U1 is a suitable outfit for the user.

[0073] [3-5. Regarding the output of images in a manner based on information about the subject] In the example shown in Figure 4, the information processing device 10 may output an image showing the coordination based on the CNN and information about a subject having a predetermined relationship with user U1 (for example, the person user U1 meets). For example, when providing image P1 to user U1, the information processing device 10 identifies an evaluator similar to the subject's distributed representation (for example, a distributed representation based on attribute information, etc.). The information processing device 10 then determines that the clothing shown in the region of image P1 where the region vector is highly similar to the distributed representation of the identified evaluator is more likely to be rated as suitable for user U1 by the subject, and the clothing shown in the region where the degree of similarity to the distributed representation of the evaluator is lower is less likely to be rated as suitable for user U1 by the subject, and outputs image P1 in a manner corresponding to the determined degree.

[0074] Furthermore, when outputting image P1, the information processing device 10 may output image P1 along with information corresponding to the degree to which the subject evaluates the clothing in each region as suiting user U1. For example, image P1-1 may be output along with information suggesting the purchase of clothing shown in regions where the subject evaluates the clothing as suiting user U1 to a high degree, information suggesting avoiding wearing clothing included in regions where the degree is low, and information on other outfits that make use of the clothing included in regions where the degree is high.

[0075] [4. Configuration of the Information Processing Device] Next, the configuration of the information processing device 10 will be described using Figure 5. Figure 5 is a diagram showing an example of the configuration of the information processing device 10 according to the embodiment. As shown in Figure 5, the information processing device 10 has a communication unit 20, a storage unit 30, and a control unit 40.

[0076] (Regarding Communications Section 20) The communication unit 20 is implemented, for example, by a NIC (Network Interface Card). The communication unit 20 is connected to the network N by wire or wireless connection and transmits and receives information between the user terminal 100, the evaluator terminal 200, etc.

[0077] (Regarding memory unit 30) The storage unit 30 is implemented by, for example, semiconductor memory elements such as RAM (Random Access Memory) and flash memory, or by storage devices such as hard disks and optical discs. As shown in Figure 3, the storage unit 30 has an evaluator information database 31 and a user information database 32.

[0078] (Regarding the evaluator information database 31) The evaluator information database 31 stores various types of information about evaluators. Here, an example of the information stored in the evaluator information database 31 will be explained using Figure 6. Figure 6 is a diagram showing an example of the evaluator information database 31 according to the embodiment. In the example in Figure 6, the evaluator information database 31 has items such as "evaluator ID," "attribute information," "questionnaire information," and "evaluation information."

[0079] "Evaluator ID" indicates identification information for identifying the evaluator. "Attribute Information" indicates the evaluator's attribute information. "Questionnaire Information" indicates the evaluator's responses to a prescribed questionnaire. "Evaluation Information" indicates information related to the evaluator's evaluation and includes items such as "Target ID," "Image Information," "Clothing Information," "Wearer Information," and "Evaluation."

[0080] "Target ID" indicates identification information used to identify the subject (image) being evaluated. "Image Information" indicates the image that is being evaluated. "Clothing Information" indicates information about the clothing included in the outfit shown in the image, such as identification information for identifying the clothing. "Wearer Information" indicates information about the wearer shown in the image, such as identification information for identifying the wearer, as well as information about the wearer's attributes and physical characteristics. "Evaluation" indicates the evaluator's evaluation of the image.

[0081] In other words, Figure 6 shows an example where the attribute information of the evaluator identified by the evaluator ID "AID#1" is "Attribute Information #1", the questionnaire information is "Questionnaire Information #1", the image information of the subject of evaluation identified by the target ID "DID#1" is "Image Information #1", the clothing information is "Clothing Information #1", the wearer information is "Wearer Information #1", and the evaluation is "Evaluation #1".

[0082] (Regarding User Information Database 32) The user information database 32 stores various types of information about the user. Here, an example of the information stored in the user information database 32 will be explained using Figure 7. Figure 7 is a diagram showing an example of the user information database 32 according to the embodiment. In the example in Figure 7, the user information database 32 has items such as "user ID," "attribute information," "questionnaire information," "user image," and "feedback information."

[0083] "User ID" indicates identification information used to identify the user. "Attribute Information" indicates the user's attribute information. "Survey Information" indicates the user's responses to a designated survey. "User Image" indicates an image showing the user's face and body shape. "Feedback Information" indicates information about the user's feedback on the information provided to the user.

[0084] In other words, Figure 7 shows an example where the user attribute information identified by the user ID "UID#1" is "Attribute Information#11", the survey information is "Survey Information#11", the user image is "User Image#1", and the feedback information is "Feedback Information#1".

[0085] (Regarding the control unit 40) The control unit 40 is a controller, and is realized, for example, by a CPU (Central Processing Unit) or MPU (Micro Processing Unit) executing various programs stored in the memory device inside the information processing device 10 using RAM as a working area. Alternatively, the control unit 40 is a controller, and is realized, for example, by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or FPGA (Field Programmable Gate Array). As shown in Figure 5, the control unit 40 according to this embodiment has a receiving unit 41, a generation unit 42, an estimation unit 43, a providing unit 44, a determination unit 45, an output unit 46, an acquisition unit 47, and an update unit 48, and realizes or executes the information processing functions and operations described below.

[0086] (Regarding Reception Desk 41) The reception unit 41 receives evaluations from multiple evaluators regarding images of a wearer wearing multiple combinations of clothing, indicating whether or not the combination suits the wearer. For example, in the example in Figure 2, the reception unit 41 presents the pair of images P1 and P2 to the evaluators, compares the two images, receives evaluations indicating which outfits in the images suit the wearer and which do not, and stores them in the evaluator information database 31.

[0087] Furthermore, the reception unit 41 may receive evaluations from multiple evaluators regarding images of a wearer wearing multiple combinations of clothing, indicating whether or not the combination suits the wearer. For example, in the example in Figure 4, the reception unit 41 receives evaluations indicating whether or not the outfit #1 shown in image P1 suits the wearer #1 shown in image P1.

[0088] (Regarding the generation unit 42) The generation unit 42 generates a distributed representation space that projects the image and evaluator information indicating the evaluator, based on the evaluation received by the reception unit 41. For example, in the example in Figure 2, the generation unit 42 refers to the evaluator information database 31 and uses VSE technology to generate a distributed representation space that projects the image and evaluator information indicating the evaluator.

[0089] Furthermore, the generation unit 42 may project the image and evaluator information closer together in the distributed representation space the more highly the evaluator rates the combination as suitable for the wearer, and project the image and evaluator information further apart in the distributed representation space the less highly the evaluator rates the combination as unsuitable for the wearer. For example, in the example in Figure 2, the generation unit 42 generates a distributed representation space F1 in which image P1 is projected closer to evaluator information D1 indicating evaluator A1 who rates image P1 as "suitable" than to evaluator information D2 indicating evaluator A2 who rates image P1 as "unsuitable". Also, the generation unit 42 generates a distributed representation space F1 in which image P2 is projected closer to evaluator information D2 indicating evaluator A2 who rates image P2 as "suitable" than to evaluator information D1 indicating evaluator A1 who rates image P2 as "unsuitable".

[0090] Furthermore, the generation unit 42 may generate a distributed representation space on which information indicating the clothing included in the combination is projected based on the evaluation. For example, in the example in Figure 2, the generation unit 42 generates a distributed representation space F2 on which the image of clothing I1 included in the coordination shown in image P1 is projected, closer to the evaluator information D1 indicating evaluator A1 who evaluated image P1 as "suitable" than to the evaluator information D2 indicating evaluator A2 who evaluated image P1 as "unsuitable". Also, the generation unit 42 generates a distributed representation space F2 on which the image of clothing I2 included in the coordination shown in image P2 is projected, closer to the evaluator information D2 indicating evaluator A2 who evaluated image P2 as "suitable" than to the evaluator information D1 indicating evaluator A1 who evaluated image P2 as "unsuitable".

[0091] Furthermore, the generation unit 42 may generate a distributed representation space on which information indicating the category to which the combination belongs is projected based on the evaluation. For example, in the example in Figure 2, the generation unit 42 generates a distributed representation space F3 on which the category T1 (casual) of the outfit shown by image P1 is projected, closer to the evaluator information D1 indicating evaluator A1 who evaluated image P1 as "suitable" than to the evaluator information D2 indicating evaluator A2 who evaluated image P1 as "unsuitable". Also, the generation unit 42 generates a distributed representation space F3 on which the category T2 (date) of image P2 is projected, closer to the evaluator information D2 indicating evaluator A2 who evaluated image P2 as "suitable" than to the evaluator information D1 indicating evaluator A1 who evaluated image P2 as "unsuitable".

[0092] Furthermore, the generation unit 42 may generate a distributed representation space on which information indicating the provider of the clothing included in the combination is projected based on the evaluation. For example, in the example in Figure 2, the generation unit 42 generates a distributed representation space on which information indicating the provider of the clothing included in the coordination shown in image P1 is projected closer to the evaluator information D1 indicating evaluator A1 who evaluated image P1 as "suitable" than to the evaluator information D2 indicating evaluator A2 who evaluated image P1 as "unsuitable". Also, the generation unit 42 generates a distributed representation space on which information indicating the provider of the clothing included in the coordination shown in image P2 is projected closer to the evaluator information D2 indicating evaluator A2 who evaluated image P2 as "suitable" than to the evaluator information D1 indicating evaluator A1 who evaluated image P2 as "unsuitable".

[0093] Furthermore, the generation unit 42 may generate a distributed representation space onto which evaluator information indicating the attributes of the evaluators is projected. For example, in the example in Figure 2, the generation unit 42 generates a distributed representation space F4 onto which image P1 is projected, closer to the attribute information of evaluator A1 who evaluated image P1 as "suitable" than closer to the attribute information of evaluator A2 who evaluated image P1 as "unsuitable". Similarly, the generation unit 42 generates a distributed representation space F4 onto which image P2 is projected, closer to the attribute information of evaluator A2 who evaluated image P2 as "suitable" than closer to the attribute information of evaluator A1 who evaluated image P2 as "unsuitable".

[0094] Furthermore, the generation unit 42 may generate a distributed representation space onto which evaluator information showing the evaluators' responses to a predetermined questionnaire is projected. For example, in the example in Figure 2, the generation unit 42 generates a distributed representation space F5 onto which image P1 is projected, closer to the response of evaluator A1 who evaluated image P1 as "suitable" than to the response of evaluator A2 who evaluated image P1 as "unsuitable". Similarly, the generation unit 42 generates a distributed representation space F5 onto which image P2 is projected, closer to the response of evaluator A2 who evaluated image P2 as "suitable" than to the response of evaluator A1 who evaluated image P2 as "unsuitable".

[0095] Furthermore, the generation unit 42 may generate a convolutional neural network based on the evaluations received by the reception unit 41. For example, in the example in Figure 4, the generation unit 42 generates a CNN using training data that includes an image, the evaluator's evaluation of the image, and evaluator information about the evaluator.

[0096] (Regarding Estimation Section 43) The estimation unit 43 estimates multiple clothing combinations that suit the user based on the distributed representation space and information about the user. For example, in the example in Figure 2, the estimation unit 43, in response to a request for coordination from the user terminal 100 via an e-commerce service or a coordination service, refers to the user information database 32 and estimates a coordination that suits user U1 based on the generated distributed representation space and information about user U1.

[0097] Furthermore, the estimation unit 43 may estimate multiple clothing combinations that suit a given target based on a distributed representation space that projects the image and evaluator information indicating the evaluators, which is generated based on evaluations from multiple evaluators of an image of a wearer wearing multiple clothing combinations, indicating whether or not the combination suits the wearer. For example, in the example in Figure 2, the estimation unit 43 estimates a suitable outfit for user U1 based on the generated distributed representation space in response to a request for outfit provision from user terminal 100 via an e-commerce service or a coordination service.

[0098] Furthermore, the estimation unit 43 may estimate a suitable combination for the user based on the distributed representation space and information about the user. For example, in the example in Figure 2, the estimation unit 43 estimates a suitable outfit for user U1 based on the generated distributed representation space and information about user U1.

[0099] Furthermore, the estimation unit 43 may estimate a suitable combination for the user based on at least one of the user's attributes, the clothing category selected by the user, or the user's responses to a predetermined questionnaire. For example, in the example in Figure 2, if the attribute information of user U1 is "late 20s", the estimation unit 43 estimates that the outfit shown by the image projected from attribute information B2 within a predetermined range in the distributed representation space F4 is a suitable outfit for user U1, and provides the user terminal 100 with information about the outfit that corresponds to information about user U1 and information related to the outfit. Also, if the estimation unit 43 receives information indicating that user U1 has selected the category "casual", it estimates that the outfit shown by the image projected from attribute information B2 within a predetermined range in the distributed representation space F4, and the outfit shown by the image projected from category T1 within a predetermined range in the distributed representation space F3, is a suitable outfit for user U1 in the "casual" category, and provides the user terminal 100 with information about the outfit that corresponds to information about user U1 and information related to the outfit. Furthermore, if user U1 answers "I'm not confident in my fashion sense" to a question such as "Are you confident in your fashion sense?", the estimation unit 43 estimates that the outfit shown by the image projected from the answer C2 within a predetermined range in the distributed representation space F5 is an outfit that suits user U1, and provides information about the outfit that corresponds to information about user U1 and information related to the outfit to the user terminal 100.

[0100] Furthermore, the estimation unit 43 may estimate combinations that the subject evaluates as suitable for the user, based on the distributed representation space and information about the subject. For example, in the example in Figure 2, the estimation unit 43 estimates coordinations that the subject evaluates as suitable for user U1, based on the distributed representation space and information about the subject having a predetermined relationship with user U1.

[0101] Furthermore, the estimation unit 43 may estimate combinations that the subject evaluates as suitable for the user based on at least one of the subject's attributes or the subject's responses to a predetermined questionnaire. For example, in the example in Figure 2, if the subject's attribute information is "early 20s," the estimation unit 43 estimates that the outfit shown by the image projected from the attribute information B6 within a predetermined range in the distributed representation space F4 is an outfit that the subject evaluates as suitable for the user, and provides the user terminal 100 with information about the outfit that corresponds to information about the user U1 and information related to the outfit. Also, if the subject answers "I am confident in my fashion sense" to a questionnaire such as "Are you confident in your fashion sense?", the estimation unit 43 estimates that the outfit shown by the image projected from the response C5 within a predetermined range in the distributed representation space F5 is an outfit that the subject evaluates as suitable for the user, and provides the user terminal 100 with information about the outfit that corresponds to information about the user U1 and information related to the outfit.

[0102] Furthermore, the estimation unit 43 may estimate a combination that suits the subject based on the distributed representation space and information about the subject. For example, in the example in Figure 2, the estimation unit 43 estimates an outfit that the subject evaluates as suiting them based on the distributed representation space and information about the subject who has a predetermined relationship with user U1.

[0103] Furthermore, the estimation unit 43 may estimate a combination that suits the subject based on at least one of the subject's attributes or the subject's responses to a predetermined questionnaire. For example, in the example in Figure 2, if the subject's attribute information is "early 20s," the estimation unit 43 estimates that the outfit shown by the image projected from the attribute information B6 within a predetermined range in the distributed representation space F4 is an outfit that the subject would rate as suitable, and provides the user terminal 100 with information about the outfit that corresponds to the subject's information and information related to the outfit. Also, if the subject answers "I am confident in my fashion sense" to a questionnaire such as "Are you confident in your fashion sense?", the estimation unit 43 estimates that the outfit shown by the image projected from the response C5 within a predetermined range in the distributed representation space F5 is an outfit that the subject would rate as suitable, and provides the user terminal 100 with information about the outfit that corresponds to the subject's information and information related to the outfit.

[0104] Furthermore, the estimation unit 43 may estimate combinations that suit a predetermined target based on a distributed representation space onto which an image showing the wearer's face is projected, and an image showing the face of a predetermined target. For example, in the example in Figure 2, the estimation unit 43 estimates that among the coordinates #1, #3, and #4 estimated based on the distributed representation space F4 and the attribute information of user U1, the coordinate worn by the wearer whose facial features are similar to those of user U1 is a coordinate that suits the user.

[0105] Furthermore, the estimation unit 43 may estimate combinations that suit a predetermined target based on a distributed representation space onto which an image showing the wearer's body shape is projected, and an image showing the body shape of a predetermined target. For example, in the example in Figure 2, the estimation unit 43 estimates that among the coordinates #1, #3, and #4 estimated based on the distributed representation space F4 and the attribute information of user U1, the coordinate worn by a wearer whose body shape is similar to that of user U1 is a coordinate that suits the user.

[0106] Furthermore, the estimation unit 43 may further estimate the degree to which the combination suits a predetermined target. For example, in the example in Figure 4, the estimation unit 43 estimates that the more the coordinate information indicated by the image closer to the attribute information B2 in the distributed representation space F4, the higher the degree to which it suits the user U1.

[0107] (Regarding Section 44) The provisioning unit 44 provides users with combination information relating to the combinations estimated by the estimation unit 43. For example, in the example in Figure 2, the provisioning unit 44 provides information on coordination #1, #3, and #4, as well as information on the clothing included in coordination #1, #3, and #4, in the e-commerce service. The provisioning unit 44 also provides information on coordination #1, #3, and #4 in the coordination service.

[0108] Furthermore, the provisioning unit 44 may provide the user with information regarding the combinations estimated by the estimation unit 43. For example, in the example in Figure 2, the provisioning unit 44 provides information on coordination #1, #3, and #4, as well as information on the clothing included in coordination #1, #3, and #4, in the e-commerce service. The provisioning unit 44 also provides information on coordination #1, #3, and #4 in the coordination service.

[0109] Furthermore, the providing unit 44 may provide combinations in a manner that depends on the degree of suitability. For example, in the example shown in Figure 2, the providing unit 44 estimates that the information regarding coordination indicated by images closer to the attribute information B2 in the distributed representation space F4 is more suitable for user U1, and provides it with a higher priority.

[0110] Here, using Figures 8 and 9, we will explain the manner in which information regarding clothing combinations is provided to the user terminal 100. First, using Figure 8, we will explain the manner in which information regarding clothing combinations is provided in an e-commerce service. Figure 8 is Figure (1), which shows an example of the screen of the user terminal 100.

[0111] As shown in Figure 8, the providing unit 44 provides a screen SC1 that displays search results corresponding to the search query "T-shirt" or "cut and sew" entered by user U1 in the e-commerce service. For example, the providing unit 44 provides a screen SC1 that displays images of outfits that include a T-shirt or cut and sew, in order of how well the outfits shown in the images suit user U1 (in other words, a screen SC1 that is specified to be displayed in order of how well it suits user U1 in the dropdown area AR1). The providing unit 44 also displays information that can be specified as the information displayed in the dropdown area AR1 (i.e., the conditions for narrowing down the search results), such as "in order of how well the target person rates you (user U1)" or "in order of how well the target person rates themselves."

[0112] For example, if "the order in which the target person evaluates what suits you" is specified in area AR1, the providing unit 44 provides screen SC1 in the distributed representation space, displaying images that represent outfits corresponding to information about user U1 and information related to the outfit, from images projected within a predetermined range from the information about the target person, in the order in which they are projected closer to the information about the target person (in other words, in order of how closely the target person evaluates what suits user U1). In such a case, the information processing device 10 may receive information from user U1 for identifying the target person in the e-commerce service (in other words, a target person ID linked to the information about the target person) and information about the target person. The providing unit 44 then provides screen SC1 based on the information received from user U1.

[0113] Furthermore, if "the order in which the subject evaluates what suits them" is specified in area AR1, the providing unit 44 provides screen SC1 in the distributed representation space, displaying images that represent outfits corresponding to information about the subject and information related to the outfit, from among the images projected within a predetermined range from the information about the subject, in the order in which they are projected closer to the information about the subject (in other words, in the order in which the subject evaluates what suits user U1 to the highest degree). In such a case, the information processing device 10 may receive information from user U1 for identifying the subject in the e-commerce service, as well as information about the subject. The providing unit 44 then provides screen SC1 based on the information received from user U1.

[0114] Next, using Figure 9, we will explain how information regarding clothing combinations is provided in the coordination service. Figure 9 is Figure (2), which shows an example of the screen of the user terminal 100.

[0115] As shown in Figure 9, the service provider 44 provides a screen SC2 that displays search results corresponding to the search query entered by user U1 in the coordination service. For example, the service provider 44 provides a screen SC2 that displays images of outfits corresponding to the search query entered by user U1, in order of how well the outfits shown in the images suit user U1 (in other words, a screen SC2 that is specified to display in order of how well it suits user U1 in the dropdown area AR2). The service provider 44 also allows users to specify information to be displayed in the dropdown area AR2, such as "in order of how well the target person rates you (user U1)" or "in order of how well the target person rates themselves."

[0116] For example, if "the order in which the target person evaluates what suits you" is specified in area AR2, the providing unit 44 provides screen SC2 in the distributed representation space, displaying images that represent outfits corresponding to information about user U1 and information related to the outfit, from among the images projected within a predetermined range from the information about the target person, in order of being projected closer to the information about the target person. In this case, the information processing device 10 may receive information from user U1 for identifying the target person in the outfit service, as well as information about the target person. The providing unit 44 then provides screen SC2 based on the information received from user U1.

[0117] Furthermore, if the "order in which the subject evaluates what suits them best" is specified in area AR2, the providing unit 44 provides screen SC2 in the distributed representation space, displaying images that represent the coordination corresponding to the subject's information and coordination-related information, among the images projected within a predetermined range from the subject's information, in order of being projected closer to the subject's information. In such a case, the information processing device 10 may receive information for identifying the subject in the coordination service and information about the subject from the user U1. The providing unit 44 then provides screen SC2 based on the information received from user U1.

[0118] In addition, in Figures 8 and 9, the providing unit 44 may use the region vectors generated by the CNN to estimate the regions that each evaluator judged to be suitable from the outfit shown in each image (in other words, the key clothing items in the outfit), and provide an image showing the estimated regions as a heat map.

[0119] In the example above, we showed an example of displaying product information in an order of suitability based on a specific search query, but the scope of the present invention is not limited to that case. For example, even if there is no search query related to a product category, the providing unit 44 may evaluate the degree of suitability for a group of products including any product category for the user in question and provide the recommendation result to the user.

[0120] Furthermore, even outside of search queries, it is possible to display the suitability evaluation results to the user in combination with various filters and other evaluation results. For example, in a recommendation system, candidate items worthy of recommendation to the user and their recommendation level evaluation results can be identified in advance. Then, the suitability level of these candidate items can be evaluated, and the items can be displayed to the user in order according to their recommendation level and suitability level. In this way, it is possible to provide users with information on items (products) and outfits using results that combine suitability level with other factors in a complex manner.

[0121] (Regarding the determination unit 45) The determination unit 45 determines how well each piece of clothing in a given region suits the user, based on multiple regions in the distributed representation and information about the user. For example, in the example in Figure 4, the determination unit 45 identifies an evaluator whose distributed representation is similar to that of user U1. The determination unit 45 then determines that the clothing in a region of image P1 whose region vector is highly similar to the distributed representation of the identified evaluator suits user U1 better, and the clothing in a region whose region is less similar to the distributed representation of the evaluator suits user U1 less.

[0122] Furthermore, the determination unit 45 may determine the degree to which the subject evaluates each of the garments contained in a region as suitable for the user, based on multiple regions in the distributed representation and information about the subject. For example, in the example in Figure 4, the determination unit 45 identifies an evaluator whose distributed representation is similar to that of the subject. The determination unit 45 then determines that the garments shown in regions of the image P1 whose region vectors are highly similar to the distributed representation of the identified evaluator are more likely to be evaluated as suitable for the user U1 by the subject, and the garments shown in regions whose region vectors are less similar to the distributed representation of the evaluator are less likely to be evaluated as suitable for the user U1 by the subject.

[0123] (Regarding output section 46) The output unit 46 outputs the image in a manner based on the distributed representation generated for each region of the image by the convolutional neural network generated by the generation unit 42. For example, in the example in Figure 4, the output unit 46 uses the region vectors generated by the CNN to estimate the regions that each evaluator judges to be suitable, and outputs a heat map showing the estimated regions.

[0124] Furthermore, the output unit 46 may output an image in a manner based on multiple regions in the distributed representation and evaluator information indicating the evaluator. For example, in the example in Figure 4, the output unit 46 uses region vectors generated by a CNN trained to output a distributed representation similar to the distributed representation of evaluators who evaluated an image as suitable, and a distributed representation that is not similar to the distributed representation of the evaluator information of evaluators who evaluated an image as unsuitable, to estimate the regions that each evaluator judged to be suitable, and outputs a heatmap showing the estimated regions.

[0125] Furthermore, the output unit 46 may output an image in which the degree of emphasis is higher for areas where the distributed representation is similar to the distributed representation of the evaluator information, and lower for areas where the distributed representation is not similar to the distributed representation of the evaluator information. For example, in the example in Figure 4, the output unit 46 outputs a heatmap in which the degree of red is higher for areas where the distributed representation is similar to the distributed representation of the evaluator information, and the degree of blue is higher for areas where the distributed representation is not similar to the distributed representation of the evaluator information.

[0126] Furthermore, the output unit 46 may output to the user an image that displays the regions in a manner corresponding to the degree to which the user is suitable, as determined by the determination unit 45. For example, in the example in Figure 4, if the distributed representation of the evaluator identified in step Sb3 is similar to the region vector of the region showing the bottoms in image P1, but the region vector of the region showing the tops in image P1 is not similar, the information processing device 10 outputs image P1-1, in which the region showing the bottoms in image P1 is displayed in dark red and the region showing the tops in dark blue. Also, if the distributed representation of the evaluator identified in step Sb3 is similar to the region vector of the region showing the tops in image P1, but the region vector of the region showing the bottoms in image P1 is not similar, the information processing device 10 outputs image P1-2, in which the region showing the tops in image P1 is displayed in dark red and the region showing the bottoms in dark blue.

[0127] Furthermore, the output unit 46 may output an image to the user that displays regions in a manner corresponding to the degree to which the subject, as determined by the determination unit 45, evaluates the clothing to suit the user. For example, in the example of Figure 4, the output unit 46 determines that among the regions of image P1, the clothing shown in regions where the region vector is similar to the distributed representation of the evaluator, which is similar to the distributed representation of the subject, is evaluated as being more suitable to the user U1 by the subject. Conversely, the clothing shown in regions where the degree of similarity to the distributed representation of the evaluator is lower is evaluated as being less suitable to the user U1 by the subject. The output unit 46 then outputs image P1 in a manner corresponding to the determined degree.

[0128] (Regarding acquisition section 47) The acquisition unit 47 acquires feedback from the user or other users who have a predetermined relationship with the user regarding the combination information provided by the provision unit 44. For example, in the example in Figure 2, the acquisition unit 47 acquires information (feedback) indicating whether user U1 purchased or applied for rental of clothing included in coordination #1, #3, and #4 in the e-commerce service, and stores it in the user information database 32. The acquisition unit 47 also acquires information (feedback) indicating whether user U1 registered coordination #1, #3, and #4 as favorites in the coordination service. Furthermore, the acquisition unit 47 acquires information (feedback) from the user terminal 100 or a terminal device used by the target person indicating whether the target person had a positive reaction to user U1 wearing the coordination indicated by the provided information. In addition, when user U1 gifted clothing included in the coordination indicated by the provided information to the target person, the acquisition unit 47 acquires information (feedback) from the user terminal 100 or a terminal device used by the target person indicating whether the target person had a positive reaction.

[0129] (Regarding update section 48) The update unit 48 updates the distributed representation space based on the feedback obtained by the acquisition unit 47. If user U1 is positive towards coordination #1, the update unit 48 updates the distributed representation space F4 so that attribute information B2 and image P1 are brought closer together by a predetermined distance. On the other hand, if user U1 is negative towards coordination #3, the update unit 48 updates the distributed representation space F4 so that attribute information B2 and image P3 are moved further apart by a predetermined distance.

[0130] Furthermore, if the target person has a positive view of user U1 wearing the outfit indicated by the provided information, the update unit 48 updates the distributed representation space so that the information corresponding to the target person (e.g., attribute information or responses to a questionnaire) and the image representing the outfit are brought closer together by a predetermined distance. On the other hand, if the target person has a negative view of user U1 wearing the outfit indicated by the provided information, the update unit 48 updates the distributed representation space so that the information corresponding to the target person and the image representing the outfit are moved further apart by a predetermined distance.

[0131] Furthermore, if user U1 presents a garment included in the outfit indicated by the information provided to user U1 to the target person, and the target person has a positive reaction, the update unit 48 updates the distributed representation space so that the information corresponding to the target person and the image representing the outfit are brought closer together by a predetermined distance. On the other hand, if user U1 presents a garment included in the outfit indicated by the information provided to user U1 to the target person, and the target person has a negative reaction, the update unit 48 updates the distributed representation space so that the information corresponding to the target person and the image representing the outfit are moved further apart by a predetermined distance.

[0132] [5. Information Processing Flow] Using Figure 10, the information processing procedure (1) of the information processing device 10 according to the embodiment will be explained. Figure 10 is a flowchart (1) showing an example of the information processing procedure according to the embodiment.

[0133] As shown in Figure 10, the information processing device 10 determines whether or not it has received an evaluation for the image of a wearer wearing multiple combinations of clothing (step S101). If no evaluation has been received (step S101; No), the information processing device 10 waits until an evaluation is received.

[0134] On the other hand, if an evaluation is accepted (step S101; Yes), the information processing device 10 generates a distributed representation space on which the image and evaluator information indicating the evaluator are projected based on the evaluation (step S102), and then terminates the process.

[0135] Next, the information processing procedure (2) of the information processing device 10 according to the embodiment will be described using Figure 11. Figure 11 is a flowchart (2) showing an example of the information processing procedure according to the embodiment.

[0136] As shown in Figure 11, the information processing device 10 determines whether or not it has received an evaluation for the image of the wearer wearing multiple combinations of clothing (step S201). If no evaluation has been received (step S201; No), the information processing device 10 waits until an evaluation is received.

[0137] On the other hand, if an evaluation is accepted (step S201; Yes), the information processing device 10 generates a convolutional neural network based on the evaluation (step S202). Subsequently, the information processing device 10 outputs the image in a manner based on the distributed representation generated by the convolutional neural network for each region of the image (step S203), and terminates the process.

[0138] Next, the information processing procedure (3) of the information processing device 10 according to the embodiment will be described using Figure 12. Figure 12 is a flowchart (3) showing an example of the information processing procedure according to the embodiment.

[0139] As shown in Figure 12, the information processing device 10 determines whether or not it has received a request to provide a combination of clothing (step S301). If the request has not been received (step S301; No), the information processing device 10 waits until it receives a request.

[0140] On the other hand, if a request for information is received (step S301; Yes), the information processing device 10 estimates multiple clothing combinations that would suit a given object based on a distributed representation space onto which images of a wearer wearing multiple clothing combinations, generated based on the evaluation from the evaluator, and evaluator information indicating the evaluator are projected (step 3202). Subsequently, the information processing device 10 provides the user with information regarding the estimated combinations (step S303) and terminates the process.

[0141] [6. Variant Example] The above-described embodiment is merely an example, and various modifications and applications are possible.

[0142] [6-1. Regarding the provision of information to evaluators] In the above-described embodiment, the information processing device 10 may provide information regarding coordination based on the evaluator's evaluation of the image. For example, the information processing device 10 may provide information regarding coordination corresponding to an image that the evaluator has deemed suitable in an e-commerce service or a coordination service.

[0143] [6-2. Regarding the target group] In the embodiments described above, the target person was shown to be someone the user meets or someone to whom a gift is given, but the target person is not limited to such examples. For example, the target person may be someone the user serves, such as a shop assistant or coordinator. In such cases, the information processing device 10 estimates the outfit that the target person evaluates as suiting them (in other words, the outfit that the target person likes) based on the distributed representation space and information about the target person the user serves, and outputs information about the estimated outfit to the user. The information processing device 10 also outputs an image showing the outfit (for example, an image showing which clothing item the target person likes) in a manner based on the CNN and information about the target person the user serves.

[0144] The information processing device 10 may also perform matching between users, such as shop staff or coordinators, and target individuals, who are customers. For example, based on information about each user and information about each target individual, the information processing device 10 estimates the position of each user and target individual when projected onto a distributed representation space, and provides information about users and target individuals located within a predetermined range from their projected positions to each user and target individual.

[0145] [6-3. Regarding the provision of information on clothing design] In the above-described embodiment, the information processing device 10 may receive evaluations from evaluators of clothing data designed by the manufacturer and provide the manufacturer with information based on those evaluations. For example, the information processing device 10 generates a CNN using training data that includes data showing a wearer wearing clothing designed by the manufacturer (e.g., CAD data), evaluations from evaluators of that data, and evaluator information about the evaluators. To give a specific example, the information processing device 10 generates a CNN that has been trained to generate region vectors (vectors that show the characteristics of each region after dividing the clothing shown in the data into multiple regions) that are similar to the distributed representations of each evaluator who is likely to have evaluated a certain region of clothing shown in the data as looking good, and different from the distributed representations of each evaluator who is likely to have evaluated it as looking bad.The information processing device 10 then uses the region vectors generated by such a trained CNN to estimate the regions that each evaluator judged to look good (in other words, parts of clothing such as collars and hems), and outputs a heat map showing the estimated regions.

[0146] To give a more specific example, the information processing device 10, based on the CNN and information about a predetermined user (for example, the clothing purchasing demographic), determines the degree to which the user would rate each part of the clothing represented by each region of the data as suitable for the user, and outputs data showing the determined degree as a heat map to the manufacturer.

[0147] This allows the information processing device 10 to understand which parts of the garment being designed are considered suitable for the user and which parts are considered unsuitable, thereby improving the convenience of garment design.

[0148] [6-4. Regarding the processing method] Of the processes described in the above embodiments, all or part of the processes described as being performed automatically can be performed manually, and conversely, all or part of the processes described as being performed manually can be performed automatically by known methods. In addition, the processing procedures, specific names, and information including various data and parameters shown in the above text and drawings can be arbitrarily changed unless otherwise specified. For example, the various information shown in each figure is not limited to the information shown.

[0149] Furthermore, the components of each illustrated device are functionally conceptual and do not necessarily need to be physically configured as shown. In other words, the specific forms of distribution and integration of each device are not limited to those shown, and all or part of them can be functionally or physically distributed and integrated in any unit according to various loads and usage conditions.

[0150] Furthermore, the embodiments described above can be combined as appropriate, provided that the processing content is not contradictory.

[0151] [6-5. Regarding the subjects of evaluation] In the above-described embodiment, the information processing device 10 receives evaluations from multiple evaluators regarding an image of a wearer wearing a combination of clothing, indicating whether the combination suits the wearer, and generates a distributed representation space or CNN based on the received evaluations. However, the subject of evaluation is not limited to combinations of clothing and may be any. For example, the information processing device 10 may receive evaluations from multiple evaluators regarding an image showing a combination of furniture, indicating whether the combination suits the wearer, and generate a distributed representation space or CNN based on the received evaluations. To give a specific example, the information processing device 10 generates a distributed representation space that projects images showing the furniture combination, information about the furniture shown in the image (e.g., furniture type, brand, shop name, place of manufacture, usage scene, color, category (casual, etc.), sales pitch, material, size, price range, etc.), and information about the evaluators, based on the evaluations received from the evaluators, and provides the user with information about the furniture combination that the user evaluates as suitable. Furthermore, the information processing device 10 generates a CNN based on the evaluation received from the evaluator, and outputs an image showing the furniture combination in a manner based on a distributed representation generated by the CNN for each region of the image showing the furniture combination.

[0152] Furthermore, for example, the information processing device 10 may receive evaluations from multiple evaluators regarding an image showing a combination of multiple dishes (e.g., a menu), indicating whether the combination is appropriate or not (e.g., whether the appearance or taste combination is appropriate), and generate a distributed representation space or a CNN based on the received evaluations. To give a specific example, the information processing device 10 generates a distributed representation space that projects an image showing the combination of dishes, information about the dishes shown in the image (e.g., dish name, ingredients, genre of cuisine, season, taste characteristics, price range, etc.), and information about the evaluators, based on the evaluations received from the evaluators, and provides the user with information about the combination of dishes that the user evaluates as appropriate. The information processing device 10 also generates a CNN based on the evaluations received from the evaluators and outputs an image showing the combination of dishes in a manner based on the distributed representation generated by the CNN for each region of the image showing the combination of dishes.

[0153] Furthermore, for example, the information processing device 10 may receive evaluations from multiple evaluators regarding an image showing a combination of multiple flowers (e.g., a bouquet), indicating whether the combination is appropriate or not, and generate a distributed representation space or a CNN based on the received evaluations. Specifically, the information processing device 10 generates a distributed representation space that projects an image showing the combination of flowers, information about the flowers shown in the image (e.g., usage scene, names of individual flowers, colors, season, purpose, category (e.g., casual), sales pitch, origin, price range, etc.), and information about the evaluators, based on the evaluations received from the evaluators, and provides the user with information about the combination of flowers that the user evaluates as appropriate. The information processing device 10 also generates a CNN based on the evaluations received from the evaluators and outputs an image showing the combination of flowers in a manner based on the distributed representation generated by the CNN for each region of the image showing the combination of flowers.

[0154] [6-6. About the images] In the embodiments described above, the information processing device 10 processes an image of a wearer wearing multiple combinations of clothing. However, the embodiments are not limited to this, and similar processing may be performed on an image of a wearer wearing only one piece of clothing.

[0155] [6-7. Estimation of clothing based on the user's self-reported physical characteristics] In the embodiments described above, the information processing device 10 may estimate clothing that suits the user based on physical information (user-reported physical information) that indicates the physical characteristics the user claims to have. Alternatively, the information processing device 10 may estimate clothing that suits the user based on physical information (user-reported physical information) that indicates the physical characteristics the user claims to have. For example, the information processing device 10 has an estimation unit 43 that estimates clothing that suits a predetermined target person based on a distributed representation space that is generated based on evaluations from multiple evaluators of an image of a wearer wearing clothing, indicating whether or not the clothing suits the wearer, and on a distributed representation space onto which the image and wearer physical information indicating the physical characteristics of the wearer are projected, and a provision unit 44 that provides the user with information about the clothing estimated by the estimation unit 43, and the estimation unit 43 may estimate clothing that suits a predetermined target person based on the distributed representation space, target physical information indicating the physical characteristics of a predetermined target person, and target self-reported physical information indicating the physical characteristics that a predetermined target person claims to have. Furthermore, for example, the information processing device 10 has an estimation unit 43 that estimates clothing that suits a predetermined subject based on a distributed representation space that is generated based on evaluations from multiple evaluators indicating whether or not the clothing suits the subject, and on a distributed representation space onto which the image, subject body information indicating the physical characteristics of the subject, and evaluator information indicating the evaluators are projected, and a provision unit 44 that provides information about the clothing estimated by the estimation unit 43 to the user, and the estimation unit 43 may estimate the clothing that suits a predetermined subject based on the distributed representation space, subject body information indicating the physical characteristics of the predetermined subject, and subject self-reported body information indicating the physical characteristics that the predetermined subject claims to be.Furthermore, for example, the information processing device 10 has an estimation unit 43 that estimates clothing that suits a predetermined subject based on a distributed representation space that is generated based on evaluations from multiple evaluators indicating whether or not the clothing suits them, and on a distributed representation space onto which the image and evaluator body information indicating the physical characteristics of the evaluators are projected, and a provision unit 44 that provides the user with information about the clothing estimated by the estimation unit 43, and the estimation unit 43 may estimate the clothing that suits a predetermined subject based on the distributed representation space, subject body information indicating the physical characteristics of the predetermined subject, and subject self-reported body information indicating the physical characteristics that the predetermined subject claims to be. Furthermore, for example, the information processing device 10 has an estimation unit 43 that estimates clothing that suits a predetermined subject based on a distributed representation space generated based on evaluations from multiple evaluators indicating whether or not the clothing suits them, where the distributed representation space is projected with the image, evaluator body information indicating the physical characteristics of the evaluators, and evaluator information (excluding the evaluator body information) indicating the evaluators, and a provision unit 44 that provides the user with information about the clothing estimated by the estimation unit 43. The estimation unit 43 may also estimate clothing that suits a predetermined subject based on the distributed representation space, subject body information indicating the physical characteristics of the predetermined subject, and subject self-reported body information indicating the physical characteristics that the predetermined subject claims to have. Furthermore, the estimation unit 43 may estimate clothing that suits a predetermined subject based on the positional relationship between the image, the subject body information, and the subject self-reported body information. Furthermore, the estimation unit 43 may estimate clothing that suits a predetermined subject based on the area of ​​the space connecting the image, the subject information, and the subject self-reported body information. Furthermore, the estimation unit 43 may estimate clothing suitable for a given subject based on the positional relationship between the image, the subject's physical information, the subject's self-reported physical information, and the evaluator's information. Alternatively, the estimation unit 43 may estimate clothing suitable for a given subject based on the area of ​​the space connecting the image, the subject's information, the subject's self-reported physical information, and the evaluator's information. This allows the information processing device 10 to take into account the user's self-reported body shape and provide information about clothing suitable for the user in e-commerce services and coordination services.For example, even if a user's actual body type is "thin," their self-perceived body type may be "slightly overweight" rather than "thin." In such cases, it is not always appropriate to suggest clothing that suits the user's body type, regardless of their self-perceived body type. In other words, even if a user's actual body type is "thin," if their self-perceived body type is "slightly overweight," it is not always appropriate to suggest clothing that suits a "thin" body type. By taking into account the user's self-perceived body type, it is possible to suggest clothing that is more suitable for the user. In this case, the information processing device 10 may estimate clothing that suits the user by changing the weights of the user's self-reported body information and the user's physical information (for example, by increasing the weight of one of them). The weights (or weight ratios, etc.) of the user's self-reported body information and the user's physical information can be set in any way and may be changed as appropriate in response to feedback from the user, etc.

[0156] (Second embodiment) Next, a second embodiment will be described. Descriptions similar to those of the first embodiment will be omitted as appropriate. Furthermore, in the following embodiments, the body information is not limited to body shape information; it may include any information that indicates physical characteristics. For example, the body information may include information about the shape of the face, head, eyes, ears, nose, and mouth; skin color (e.g., warm undertones, cool undertones); hair color (e.g., black hair, brown hair); hairstyle; face shape; skeletal structure; hand length; neck thickness; etc. Also, for example, the body information may include information about attributes such as age, gender, height, weight, race, and nationality. Furthermore, for example, the body information may include additional information such as whether or not glasses are worn, the shape of the glasses, whether or not makeup is worn, and makeup tendencies. The evaluator's evaluation may be an evaluation of a single item (e.g., a single piece of clothing) or an evaluation of a combination of items (e.g., a coordinated outfit). The evaluator's evaluation may also be an evaluation of how the single item or combination is worn. For example, variations in how the outfit is worn may be included in the dataset used for learning the coordinated outfit. For example, if there is data on variations in how to wear a single outfit, that data can be used for training. This allows the system to suggest which outfit and which style is closest to the query.

[0157] [7. Configuration of the Information Processing System] First, an information processing system 1A according to the embodiment will be described. Figure 13 is a diagram showing an example configuration of the information processing system 1A according to the embodiment. As shown in Figure 13, the information processing system 1A includes an information processing device 10A, a user terminal 100, and an evaluator terminal 200. The information processing device 10A, the user terminal 100, and the evaluator terminal 200 are connected to each other via a predetermined communication network (network N) by wired or wireless means. Note that the information processing system 1A shown in Figure 13 may include multiple information processing devices 10A, multiple user terminals 100, and multiple evaluator terminals 200.

[0158] The information processing device 10A is an information processing device that evaluates multiple images (images in which clothing is included in at least part of the image), receiving evaluations from multiple evaluators indicating whether or not the clothing suits the person, and based on the received evaluations, generates a distributed representation space that projects the image and physical information indicating the physical characteristics of the subject (the person wearing the clothing, or the evaluator themselves), and realizes information processing using the generated distributed representation space. For example, it can be implemented by an evaluator adding annotations (for example, annotations indicating evaluations) to the image. For example, if an evaluator adds an annotation indicating "suits," the clothing in the image is evaluated as suiting the person, and if an annotation indicating "does not suit," the clothing in the image is evaluated as not suiting the person. For this reason, it can be understood that the evaluator has added annotations. In addition, the image contains, for example, the poster's tags (for example, information entered by the poster when posting), and the wearer's physical information can be obtained from the poster's tags. Furthermore, more detailed body measurements may be obtained using a body measurement device as physical information of the wearer or evaluator. For example, body measurements may be obtained based on images of the wearer wearing body measurement clothing or body measurement glasses.

[0159] Furthermore, for example, the information processing device 10A provides an e-commerce service for providing clothing. The information processing device 10A also provides a service (which may be a coordination service) that accepts submissions of content showing clothing (which may also be content showing outfits) from users and provides them to other users.

[0160] Furthermore, the information processing device 10A may also function as a web server providing a website related to the service. Alternatively, the information processing device 10A may be a device that distributes information to be displayed on various service-related applications installed on the user terminal 100. Furthermore, the information processing device 10A may be a device that distributes the application data itself.

[0161] Furthermore, the information processing device 10A may function as a distribution device that distributes control information to the user terminal 100. Here, the control information is written, for example, in a scripting language such as JavaScript or a stylesheet language such as CSS. Note that the application itself distributed from the information processing device 10A may also be considered as control information.

[0162] The user terminal 100 is an information processing device used by the user. The user terminal 100 displays information distributed by the information processing device 10A and server devices providing predetermined services, using a web browser or application. In the example shown in Figure 14, the user terminal 100 is a smartphone.

[0163] The evaluator terminal 200 is an information processing device used by evaluators who evaluate images of a wearer wearing clothing (which may be a combination of multiple clothing items) to determine whether the clothing suits the wearer. The evaluator terminal 200 is also used by evaluators who evaluate whether the clothing suits the evaluator themselves (for example, whether the clothing suits the evaluator when they imagine themselves wearing it). Furthermore, the evaluator terminal 200 displays information distributed by the information processing device 10A and server devices providing predetermined services via a web browser or application. In the example shown in Figure 14, the evaluator terminal 200 is a smartphone. The evaluator performs the evaluation by adding annotations to the image. Once the evaluator has performed the evaluation, the evaluator terminal 200 transmits the evaluation information, which is then stored, for example, in the evaluator information database 31A, described later.

[0164] [8. An example of information processing] Next, an example of information processing realized by the information processing device, etc., according to this embodiment will be described using Figure 14.

[0165] The third information processing method implemented by the information processing device, etc. according to this embodiment will be described below with reference to Figure 14. Figure 14 is a diagram showing an example of information processing according to this embodiment.

[0166] First, the information processing device 10A receives the evaluator's evaluation of the image of the clothing (or outfit) from the evaluator terminal 200 (step Sc1). For example, the information processing device 10A receives an evaluation indicating whether the clothing in the image suits the person wearing it. Specifically, the information processing device 10A presents the evaluator with pairs of "suits" and "does not suit" and receives an evaluation indicating whether the clothing suits the person wearing it. The evaluation may also be an evaluation of how the clothing is worn, such as "it would be better to roll up the hems," "if the person is tall, it would be better to roll up the hems," or "if the person is short, it would be better not to roll up the hems." The evaluation may also be an evaluation of how the clothing is coordinated with other items.

[0167] Furthermore, the information processing device 10A is not limited to evaluations indicating whether the clothing suits the wearer, but also accepts evaluations indicating whether the clothing suits the evaluator. For example, the information processing device 10A determines whether the image includes a wearer of the clothing, and if the wearer is not included, it accepts an evaluation indicating whether the clothing (or outfit) suits the evaluator. In this case, the information processing device 10A presents the evaluator with pairs of "suits" and "does not suit," and accepts an evaluation indicating whether the clothing suits the evaluator. For example, the evaluation can be accepted by adding annotations indicating the evaluation to the image. The evaluation may also be an evaluation of how the clothing is worn, such as "I would look better with the hems rolled up" or "I would look better without the hems rolled up." The evaluation may also be an evaluation of how the clothing is worn in coordination with other items.

[0168] Thus, the information processing device 10A may accept evaluations indicating whether or not the garment suits the evaluator. For the sake of explanation, the following explanation will use the example of accepting evaluations indicating whether or not the garment suits the wearer.

[0169] The information processing device 10A receives evaluations from evaluator A1 indicating whether clothing (or outfit) #1 shown in image P1 suits wearer #1 shown in image P1, and whether clothing #2 shown in image P2 does not suit wearer #2 shown in image P2 (i.e., an evaluation of "suits" for image P1 and an evaluation of "does not suit" for image P2). If both examples are suitable, the device receives evaluations that clothing #1 shown in image P1 suits wearer #1 shown in image P1, and clothing #2 shown in image P2 suits wearer #2 shown in image P2. Conversely, if neither example is suitable, the device receives evaluations that clothing #1 shown in image P1 does not suit wearer #1 shown in image P1, and clothing #2 shown in image P2 does not suit wearer #2 shown in image P2. Furthermore, the information processing device 10A receives evaluations from evaluator A2, such as whether clothing #1 shown in image P1 does not suit wearer #1 shown in image P1, and whether clothing #2 shown in image P2 suits wearer #2 shown in image P2 (i.e., an evaluation of "does not suit" for image P1 and an evaluation of "suits" for image P2). Similarly, the information processing device 10A also receives evaluations of images P1 and P2 from other evaluators. In addition, the information processing device 10A is not limited to images P1 and P2, but may also receive evaluations of other images (such as images P3 and P4).

[0170] Next, the information processing device 10A generates a distributed representation space (step Sc2) on which the image and physical information indicating the physical characteristics of the subject (the wearer is the subject if the evaluation was performed on the wearer, and the evaluator is the subject if the evaluation was performed on the evaluator) are projected based on the evaluator's evaluation of the image. The distributed representation space includes, for example, pairs of images and the wearer's physical information for evaluations of the wearer, and pairs of images and the evaluator's physical information for evaluations of the evaluator. For example, the information processing device 10A generates the distributed representation space using VSE technology. To give a specific example, the information processing device 10A generates a distributed representation space F6 on which image P1 is projected closer to the physical information E1 of wearer #1, which evaluator A1 evaluated as "suitable," than to the physical information E2 of wearer #2, which evaluator A1 evaluated as "unsuitable." Furthermore, the information processing device 10A generates a distributed representation space F6 by projecting image P2 closer to the body information E2 of wearer #2, which evaluator A2 judged to "suit" the wearer, than to the body information E1 of wearer #1, which evaluator A2 judged to "not suit" the wearer. The information processing device 10A may also generate a distributed representation space on which body information, which represents various information about the image and various information about the wearer's body, is projected.

[0171] Next, the information processing device 10A receives a request for the provision of predetermined information from the user terminal 100 via an e-commerce service or a coordination service (step Sc3). Specifically, the information processing device 10A receives a request for the provision of information regarding physical characteristics that suit a predetermined garment (or coordination). For example, when a user U1 specifies a garment on a service such as an e-commerce service or a coordination service, the information processing device 10A obtains the garment specification information from the user terminal 100 and receives a request for the provision of information regarding physical characteristics that suit that garment.

[0172] Then, in response to a request, the information processing device 10A estimates physical characteristics that would suit the specified clothing based on the generated distributed representation space and the information about the clothing specified by user U1 (step Sc4). Alternatively, the information processing device 10A may estimate the position of the information about the clothing specified by user U1 when it is projected onto the distributed representation space, based on the information projected onto the distributed representation space and the information about the clothing specified by user U1, and then estimate physical characteristics that would suit the specified clothing based on the estimated position.

[0173] Here, in the example in Figure 14, the information processing device 10A determines that the estimated position is close to the positions of clothing #1 shown in image P1, clothing #3 shown in image P3, and clothing #4 shown in image P4. In such a case, the information processing device 10A provides the user terminal 100 with information about physical characteristics that suit the specified clothing, based on the information about clothing #1, #3, and #4 (step Sc5).

[0174] (Regarding the information provided to users) In step Sc5, the information processing device 10A may provide user U1 with information that is not limited to body shape, but also includes information such as skin color, hair color, hairstyle, face shape, and makeup. The information processing device 10A may also identify the closest item in the distributed representation space (for example, one item from each category) and provide the information, or it may identify all items included within a predetermined range in the distributed representation space and provide the information (the information may be provided in order of priority from closest to closest in the distributed representation space, or all items included within a predetermined range in the distributed space may be treated as equivalent and provided as such). Furthermore, the information processing device 10A may provide additional services by providing information corresponding to the estimated physical characteristics.

[0175] Here, indicators showing physical characteristics such as personal color, body type, and face type will be referred to as "physical characteristic axes" as appropriate. Specific values ​​corresponding to the physical characteristic axes will be referred to as "physical characteristic values" as appropriate. For example, the physical characteristic value corresponding to the "body type" of user U1 is "straight." Furthermore, a snapshot image or individual item images (or a single item image) of multiple items constituting a coordinate will be referred to as a "coordinate image" as appropriate. The information processing device 10A may provide a service that outputs physical characteristics suitable for a coordinate image when user U1 specifies a coordinate image on a service such as an e-commerce service or a coordination service. In this case, physical characteristics may be represented as a pair of physical characteristic axes and physical characteristic values. The physical characteristic axes may also be represented as a set of assumed physical characteristic values.

[0176] Based on type diagnosis in image consulting theory, personal colors are classified as "Spring, Summer, Autumn, Winter," body types as "Straight, Wave, Natural," and face types as "Cool, Cool Casual, Fresh, Soft Elegant, Elegant, Cute, Active Cute, Feminine," thereby setting physical characteristic values. Note that the design of the "physical characteristic axes" and "physical characteristic values" based on the above image consulting theory is merely an example, and may be added or modified as needed. Furthermore, physical characteristic values ​​are set by classifying hair color and makeup styles based on predetermined information. Note that physical characteristic values ​​are not limited to classifications and may be continuous values. Also, physical characteristic values ​​corresponding to physical characteristic axes may be output as a distribution. For example, for the physical characteristic axis of "body type," information such as "Straight: 70%, Wave: 20%, Natural: 10%" may be output. For example, the information processing device 10A may provide such information according to the closeness between the image in the distributed representation space and the physical characteristics (each physical characteristic value). This allows users to not only find products that match their physical characteristics through using the service, but also to learn which combinations of fashion items suit users with which physical characteristics. Therefore, it is thought that users can deepen their understanding of fashion that better suits their physical characteristics, thus eliminating their aversion to fashion and contributing to increased purchasing intent. Furthermore, when users are browsing outfits on a designated service or when product or outfit recommendations are sent via email, information such as the physical characteristics that suit the items or outfits being viewed or recommended may also be output.

[0177] (Regarding the handling of cases where a user specifies a combination of outfits that have never been used before) In step Sc4, if user U1 specifies an outfit that has not been seen before (is not in the training data), the information processing device 10A may estimate the physical characteristics that suit that outfit. For example, for outfits not in the training data, the information processing device 10A may learn a mapping (model) for projecting the outfit image onto the VSE space on which physical characteristic values ​​are projected, use this mapping to project the unknown outfit onto the VSE space, and output the physical characteristic values ​​that exist in the vicinity of the representation corresponding to the outfit image. In this case, the nearest neighbor search may be performed independently for each physical characteristic axis. The VSE space may also be configured separately for each physical characteristic axis. In that case, the information processing device 10A may learn mappings for projecting outfit images individually. The following describes a method for learning such mappings.

[0178] Assume that a dataset exists consisting of pairs of coordinated outfit images that an annotator has evaluated as "suitable" and physical characteristic values ​​(tags). Here, the physical characteristic values ​​may belong to the subject captured in the coordinated outfit image. Alternatively, more detailed physical measurements obtained using a body measurement device may also be used. The information processing device 10A may use both the physical characteristic values ​​and the physical measurements to construct an embedding vector relating to the body of the person who posted the coordinated outfit image. The annotator may be the person who posted the coordinated outfit image or a third party other than the person who posted it.

[0179] Furthermore, if the coordinated image consists of snapshot images, the information processing device 10A may learn using a method similar to the method for embedding regions corresponding to fashion items in the snapshot images into the VSE space. Specifically, the information processing device 10A may extract image regions corresponding to each item in the snapshot image and project a vector obtained by combining the image features of each image region into the VSE space.

[0180] Furthermore, if the coordinated image consists of multiple individual item images, the information processing device 10A may perform embedding into the VSE space by constructing a single embedding vector from the embedding vectors of each individual item image using a Set Transformer or the like. Alternatively, the information processing device 10A may project a vector formed by combining the embeddings of each individual item image into the VSE space. Alternatively, the information processing device 10A may construct a single embedding vector by taking the average of the feature vectors of multiple individual item images.

[0181] (Regarding the handling of cases where the presented physical characteristics do not match the user's physical characteristics) In step Sc5, if the physical characteristics presented to user U1 do not match user U1's physical characteristics, the information processing device 10A may suggest changing the item specified by user U1. In this case, the information processing device 10A may also suggest which item to change to. The information processing device 10A may perform different processing depending on whether user U1 judges the presented information to be true (correct) or false (incorrect). The information processing device 10A may also automatically determine whether it is true or false. If user U1 judges the presented information to be true, the information processing device 10A may, for example, encourage user U1 to change their physical appearance within a realistic range. For example, if the information processing device 10A determines that "losing weight will improve the system's evaluation score (you will have physical characteristics that match the specified item)," it may provide feedback to user U1 saying "Let's exercise." In such cases, the information processing device 10 may present the physical characteristics to be presented (physical characteristics that suit the specified item) in a manner that allows the user to set them as target physical information indicating the physical characteristics they aim for in a predetermined body management service provided by the information processing device 10 (a service that periodically acquires information indicating the user's physical characteristics from captured images of the user and supports the user's body shape management, etc.) (provided together with a setting button for setting as a target). In addition, when suggesting changes to items, the information processing device 10A may randomly change only, for example, "tops" from the coordinate images currently entered by user U1, and suggest a combination of items that will have a higher evaluation score than the current one (become more suitable for the user's physical characteristics). In this case, the number of trials for random changes may be set by the system designer, or it may be repeated until an item with a higher evaluation score than the current one is found. The information processing device 10A may also perform similar processing on items other than "tops". Furthermore, the information processing device 10A may estimate items that the user is unlikely to want to change (for example, the item initially specified, or items specified multiple times, etc.) and perform similar processing only on those other items.Furthermore, the information processing device 10A may not change items randomly, but instead use the user U1's physical information to perform a nearest neighbor search in the VSE space to find a replacement item. For example, the information processing device 1 may determine a replacement item based on an item corresponding to an image located near the user U1's physical information in the VSE space. Also, if the user U1 determines that the presented information is false, the information processing device 10A may, for example, collect feedback from the user U1 and update the system if it is reasonable. For example, if the confidence level of the system's output is lower than a predetermined threshold, the information processing device 10A may receive feedback from the user U1 and update the VSE space. In this case, the information processing device 10A may interactively obtain feedback from the user U1 using methods such as active learning. The method for calculating the confidence level may be anything. Also, the information processing device 10A may calculate a lower confidence level if the number of training data similar to the user U1's physical information is less than a certain amount.

[0182] [9. Configuration of Information Processing Equipment] Next, the configuration of the information processing device 10A will be described using Figure 15. Figure 15 is a diagram showing an example of the configuration of the information processing device 10A according to the embodiment. As shown in Figure 15, the information processing device 10A has a communication unit 20, a storage unit 30A, and a control unit 40A. Note that the communication unit 20 is the same as in the first embodiment, so its description will be omitted.

[0183] (Regarding memory unit 30A) The storage unit 30A is implemented by, for example, semiconductor memory elements such as RAM and flash memory, or storage devices such as hard disks and optical discs. As shown in Figure 15, the storage unit 30A has an evaluator information database 31A and a user information database 32A.

[0184] (Regarding the evaluator information database 31A) The evaluator information database 31A stores various types of information about evaluators. Here, an example of the information stored in the evaluator information database 31A will be explained using Figure 16. Figure 16 is a diagram showing an example of the evaluator information database 31A according to the embodiment. In the example in Figure 16, the evaluator information database 31A has items such as "evaluator ID", "evaluator information", "questionnaire information", and "evaluation information".

[0185] The "Evaluator ID," "Questionnaire Information," and "Evaluation Information" are the same as in the first embodiment, so their explanation is omitted. "Evaluator Information" shows information about the evaluator, and stores information such as the evaluator's attributes and physical information.

[0186] In other words, Figure 16 shows an example where the evaluator information of the evaluator identified by the evaluator ID "AID#1" is "Evaluator Information #1", the questionnaire information is "Questionnaire Information #1", the image information of the subject of evaluation identified by the target ID "DID#1" is "Image Information #1", the clothing information is "Clothing Information #1", the wearer information is "Wearer Information #1", and the evaluation is "Evaluation #1".

[0187] (Regarding User Information Database 32A) The user information database 32A stores various types of information about the user. Here, an example of the information stored in the user information database 32A will be explained using Figure 17. Figure 17 is a diagram showing an example of the user information database 32A according to this embodiment. In the example in Figure 17, the user information database 32A has items such as "User ID", "User Information", "Questionnaire Information", "User Image", and "Feedback Information".

[0188] The "User ID," "Questionnaire Information," "User Image," and "Feedback Information" are the same as in the first embodiment, so their explanation is omitted. "User Information" refers to information about the user, and stores information such as the user's attributes and physical information.

[0189] In other words, Figure 17 shows an example where the user information of a user identified by the user ID "UID#1" is "User Information#11", the survey information is "Survey Information#11", the user image is "User Image#1", and the feedback information is "Feedback Information#1".

[0190] (Regarding control unit 40A) The control unit 40A is a controller, and is realized, for example, by a CPU or MPU executing various programs stored in the memory device inside the information processing device 10A using RAM as a working area. Alternatively, the control unit 40A can be realized by an integrated circuit such as an ASIC or FPGA. As shown in Figure 15, the control unit 40A according to this embodiment has a receiving unit 41A, a generation unit 42A, an estimation unit 43A, a providing unit 44A, a determination unit 45, an output unit 46, an acquisition unit 47A, and an update unit 48A, and realizes or executes the information processing functions and operations described below. Note that the determination unit 45 and the output unit 46 are the same as in the first embodiment, so their description is omitted.

[0191] (Regarding reception desk 41A) The reception unit 41A receives the same information as the reception unit 41. The reception unit 41A also receives evaluations from multiple evaluators regarding images of a person wearing clothing, indicating whether or not the clothing suits the person. For example, in the example in Figure 14, the reception unit 41A presents image P1, etc., to the evaluators, receives evaluations for each image indicating whether or not the clothing in the image suits the person, and stores them in the evaluator information database 31A. The reception unit 41A also receives evaluations from multiple evaluators regarding images of clothing, indicating whether or not the clothing suits the evaluator themselves. For example, in the example in Figure 14, the reception unit 41A presents image P1, etc., to the evaluators, receives evaluations for each image indicating whether or not the clothing in the image suits the evaluator themselves, and stores them in the evaluator information database 31A. Furthermore, the reception unit 41A may determine whether or not there is a wearer in each image, and if there is a wearer, it may receive an evaluation from the evaluator indicating whether or not the clothing suits the wearer, and if there is no wearer, it may receive an evaluation from the evaluator indicating whether or not the clothing suits the evaluator themselves.

[0192] (Regarding the generation unit 42A) The generation unit 42A generates the same information as the generation unit 42. The generation unit 42A also generates a distributed representation space that projects the image and physical information indicating the physical characteristics of the subject (if the evaluator evaluates the wearer, the wearer is the subject; if the evaluator evaluates themselves, the evaluator is the subject) based on the evaluation received by the reception unit 41A. For example, in the example in Figure 14, the generation unit 42A refers to the evaluator information database 31A and uses VSE technology to generate a distributed representation space that projects the image and physical information indicating the physical characteristics of the subject.

[0193] Furthermore, the generation unit 42A may project the image and the wearer's (or evaluator's) physical information closer together in the distributed representation space the more highly the evaluator rates the clothing as suitable for the wearer (or suitable for the evaluator themselves), and project the image and the wearer's (or evaluator's) physical information further apart in the distributed representation space the more highly the evaluator rates the clothing as unsuitable for the wearer (or unsuitable for the evaluator themselves). For example, in the example in Figure 14, the generation unit 42A generates a distributed representation space F6 in which image P1 is projected closer to the body information E1 of wearer #1 of image P1, which evaluator A1 rates as "suitable", than to the body information E2 of wearer #2 of image P2, which evaluator A1 rates as "unsuitable". Furthermore, the generation unit 42A generates a distributed representation space F6 in which image P2 is projected closer to the body information E2 of wearer #2 of image P2, which evaluator A2 judged to "suit" the wearer, than to the body information E1 of wearer #1 of image P1, which evaluator A2 judged to "not suit" the wearer.

[0194] (Regarding the estimation unit 43A) The estimation unit 43A estimates physical characteristics that suit the clothing specified by the user, based on the distributed representation space and information about the clothing specified by the user. For example, in the example in Figure 14, the estimation unit 43A estimates physical characteristics that suit the specified clothing, based on the distributed representation space generated in response to a request for predetermined information from the user terminal 100 via an e-commerce service or a coordination service, and information about the clothing specified by user U1. The estimation unit 43A also estimates physical characteristics that suit a predetermined clothing based on a distributed representation space that projects an image of a wearer wearing the clothing onto the wearer's physical information, which is generated based on evaluations from multiple evaluators indicating whether the clothing suits the wearer. The estimation unit 43A also estimates physical characteristics that suit a predetermined clothing based on a distributed representation space that projects an image of a clothing onto the evaluator's physical information, which is generated based on evaluations from multiple evaluators indicating whether the clothing suits the evaluator themselves. Furthermore, the estimation unit 43A estimates physical characteristics that suit a given garment based on a distributed representation space that projects the image and the subject's physical information, which is generated based on evaluations from multiple evaluators of an image of a wearer wearing the garment, indicating whether or not the garment suits the wearer, and evaluations from multiple evaluators of an image of the garment, indicating whether or not the garment suits the evaluator themselves.

[0195] (Regarding Section 44A) The provisioning unit 44A provides the user with information about physical characteristics estimated by the estimation unit 43A. For example, in the example in Figure 14, the provisioning unit 44A provides information about physical characteristics that suit a specified garment in an e-commerce service. The provisioning unit 44A also provides information about physical characteristics that suit a specified garment in a coordination service.

[0196] (Regarding acquisition section 47A) The acquisition unit 47A acquires feedback from the user or other users who have a predetermined relationship with the user regarding the information on physical characteristics provided by the provision unit 44A. For example, in the example in Figure 14, the acquisition unit 47A acquires information (feedback) indicating whether user U1 has purchased clothing #1, #3, and #4 or applied for rental in the e-commerce service, and stores it in the user information database 32A. The acquisition unit 47A also acquires information (feedback) indicating whether user U1 has registered clothing #1, #3, and #4 as favorites in the coordination service. Furthermore, for example, if the physical characteristics that suit the clothing specified by user U1 match the user's physical characteristics, the acquisition unit 47A may acquire information (feedback) indicating whether user U1 has purchased the clothing specified in the e-commerce service or applied for rental, and store it in the user information database 32A. The acquisition unit 47A may also acquire information (feedback) indicating whether user U1 has registered the clothing specified in the coordination service as a favorite. Furthermore, the acquisition unit 47A acquires information (feedback) from the user terminal 100 or a terminal device used by the target person indicating whether or not the target person had a positive reaction to user U1 wearing the clothing indicated by the provided information. The acquisition unit 47A may also acquire information (feedback) from the user terminal 100 or a terminal device used by the target person indicating whether or not the target person had a positive reaction (whether they evaluated it as suiting user U1's physical characteristics) to user U1 wearing the clothing specified by user U1. Furthermore, when user U1 presents the clothing indicated by the provided information to the target person, the acquisition unit 47A acquires information (feedback) from the user terminal 100 or a terminal device used by the target person indicating whether or not the target person had a positive reaction. Furthermore, if the physical characteristics that suit the clothing specified by user U1 match the physical characteristics of the target person to whom the gift is to be given, the acquisition unit 47A may acquire information (feedback) indicating whether or not the target person had a positive reaction (whether they evaluated it as suiting the target person's physical characteristics).

[0197] (Regarding update section 48A) The update unit 48A updates the distributed representation space based on the feedback obtained by the acquisition unit 47A. If user U1 has a positive view of clothing #1, the update unit 48A updates the distributed representation space F6 to bring body information E1 and image P1 closer together by a predetermined distance. On the other hand, if user U1 has a negative view of clothing #3, the update unit 48A updates the distributed representation space F6 to move body information E3 and image P3 further apart by a predetermined distance. For example, the update unit 48A obtains information indicating whether user U1 has purchased the clothing specified by user U1 or applied for rental, and based on this feedback, updates user U1's body information and information indicating the clothing specified by user U1 closer together by a predetermined distance. Alternatively, for example, the update unit 48A obtains information indicating whether the subject had a positive view of user U1 wearing the clothing specified by user U1, and based on this feedback, updates user U1's body information and information indicating the clothing specified by user U1 closer together by a predetermined distance. Furthermore, for example, the update unit 48A acquires information indicating whether the recipient of the clothing gift specified by user U1 was positive or negative, and based on this feedback, updates the recipient's physical information and the information indicating the clothing specified by user U1 to bring them closer together by a predetermined distance.

[0198] Furthermore, if the subject has a positive view of user U1 wearing the clothing indicated by the provided information, the update unit 48A updates the distributed representation space so that the information corresponding to the subject (e.g., physical information or answers to a questionnaire) and the image showing the clothing are brought closer together by a predetermined distance. On the other hand, if the subject has a negative view of user U1 wearing the clothing indicated by the provided information, the update unit 48A updates the distributed representation space so that the information corresponding to the subject and the image showing the clothing are moved further apart by a predetermined distance.

[0199] Furthermore, when user U1 presents the clothing indicated by the information provided to user U1 to the target person, if the target person has a positive reaction, the update unit 48A updates the distributed representation space so that the information corresponding to the target person and the image representing the clothing are brought closer together by a predetermined distance. On the other hand, when user U1 presents the clothing indicated by the information provided to user U1 to the target person, if the target person has a negative reaction, the update unit 48A updates the distributed representation space so that the information corresponding to the target person and the image representing the clothing are moved further apart by a predetermined distance.

[0200] [10. Information Processing Flow] Using Figure 18, the information processing procedure (4) of the information processing device 10A according to the embodiment will be explained. Figure 18 is a flowchart (4) showing an example of the information processing procedure according to the embodiment.

[0201] As shown in Figure 18, the information processing device 10A determines whether or not it has received an evaluation of the image of the clothing (step S401). If no evaluation has been received (step S401; No), the information processing device 10A waits until an evaluation is received.

[0202] On the other hand, if an evaluation is accepted (step S401; Yes), the information processing device 10A generates a distributed representation space on which the image and physical information indicating the physical characteristics of the subject are projected based on the evaluation (step S402), and then terminates the process.

[0203] Next, the information processing procedure (5) of the information processing device 10A according to the embodiment will be described using Figure 19. Figure 19 is a flowchart (5) showing an example of the information processing procedure according to the embodiment.

[0204] As shown in Figure 19, the information processing device 10A determines whether or not it has received a request for the provision of predetermined information (step S501). If the request has not been received (step S501; No), the information processing device 10A waits until it receives a request.

[0205] On the other hand, if a request for information is received (Step S501; Yes), the information processing device 10A estimates the physical characteristics that would suit a given garment based on a distributed representation space that projects an image of the garment and physical information indicating the physical characteristics of the subject, which is generated based on the evaluation from the evaluator (Step 502). Subsequently, the information processing device 10A provides the user with information regarding the estimated physical characteristics (Step S503) and terminates the process.

[0206] [11. Effects] As described above, the information processing device 10A according to the embodiment includes an estimation unit 43A and a provision unit 44A. The estimation unit 43A estimates physical characteristics that suit a given garment based on a distributed representation space that is generated based on evaluations from multiple evaluators indicating whether or not the garment suits the person, and which projects the image and physical information indicating the physical characteristics of the person. The provision unit 44A provides the user with information regarding the physical characteristics estimated by the estimation unit 43A.

[0207] As a result, the information processing device 10A according to the embodiment can estimate physical characteristics that suit a given garment by configuring a space that can measure the closeness between an image of the garment and the physical information of the subject. Furthermore, since the information processing device 10A according to the embodiment can provide information about physical characteristics that are estimated to suit the given garment using the generated distributed representation space, it has the effect of enabling the user to understand the physical characteristics that suit the given garment.

[0208] Furthermore, the estimation unit 43A makes estimations based on a distributed representation space generated based on evaluations from evaluators indicating whether or not the clothing suits the wearer.

[0209] As a result, the information processing device 10A according to the embodiment allows the user to understand the physical characteristics that suit a given garment, based on a distributed representation space generated based on the subjective opinions of each evaluator indicating whether or not the garment suits the wearer.

[0210] Furthermore, the estimation unit 43A performs estimations based on a distributed representation space onto which the physical information of the wearer, who is the target, is projected.

[0211] As a result, the information processing device 10A according to the embodiment allows the user to understand the physical characteristics that suit a given garment based on a distributed representation space onto which the wearer's physical information is projected.

[0212] Furthermore, the estimation unit 43A performs estimation based on a distributed representation space generated based on evaluations from evaluators indicating whether or not the product is suitable for the evaluator.

[0213] As a result, the information processing device 10A according to the embodiment allows users to understand the physical characteristics that suit a given garment, based on a distributed representation space generated based on the subjective opinion of each evaluator indicating whether or not the garment suits them.

[0214] Furthermore, the estimation unit 43A performs estimations based on a distributed representation space onto which the physical information of the evaluator, who is the subject of the estimation, is projected.

[0215] As a result, the information processing device 10A according to the embodiment allows the user to understand the physical characteristics that suit a given garment based on a distributed representation space onto which the evaluator's physical information is projected.

[0216] Furthermore, if the image includes a person wearing the clothing, the estimation unit 43A estimates based on a distributed representation space generated based on an evaluator's evaluation indicating whether or not the clothing suits the person wearing it. If the image does not include a person wearing the clothing, the estimation unit 43A estimates based on a distributed representation space generated based on an evaluator's evaluation indicating whether or not the clothing suits the person wearing it.

[0217] As a result, the information processing device 10A according to the embodiment allows the user to understand the physical characteristics that suit a given piece of clothing, based on a distributed representation space onto which the subject's physical information is projected depending on whether or not the image includes a person wearing the clothing.

[0218] Furthermore, the estimation unit 43A estimates physical characteristics that are suitable for the clothing specified by the user, based on the distributed representation space.

[0219] As a result, the information processing device 10A according to the embodiment allows the user to understand the physical characteristics that suit the clothing specified by the user.

[0220] Furthermore, the estimation unit 43A evaluates images of clothing combinations, and estimates physical characteristics that suit a given clothing combination based on a distributed representation space generated based on evaluations from multiple evaluators indicating whether or not the clothing combination suits the person, and which projects images and physical information indicating the physical characteristics of the person. The provision unit 44A provides the user with information regarding the physical characteristics estimated by the estimation unit 43A.

[0221] As a result, the information processing device 10A according to the embodiment can estimate physical characteristics that suit a given combination of clothing by configuring a space that can measure the closeness between a coordinated image and the subject's physical information. Furthermore, since the information processing device 10A according to the embodiment can provide information about physical characteristics that are estimated to suit a given combination of clothing using the generated distributed representation space, it has the effect of enabling users to understand the physical characteristics that suit a given combination of clothing.

[0222] Furthermore, the estimation unit 43A estimates physical characteristics that suit a given way of wearing clothing, based on a distributed representation space generated from evaluations by evaluators indicating whether or not the way of wearing the clothing suits the person, in relation to images showing different ways of wearing clothing.

[0223] As a result, the information processing device 10A according to the embodiment allows users to understand the physical characteristics that suit a given garment, based on a distributed representation space generated by adding the style of dressing to the evaluation of each evaluator.

[0224] [12. Hardware Configuration] Furthermore, the information processing device 10A according to each embodiment described above can be implemented by a computer 1000 having a configuration such as that shown in Figure 20. The following explanation will use the information processing device 10A as an example. Figure 20 is a hardware configuration diagram showing an example of a computer that implements the functions of the information processing device 10A. The computer 1000 has a CPU 1100, ROM 1200, RAM 1300, HDD 1400, communication interface (I / F) 1500, input / output interface (I / F) 1600, and media interface (I / F) 1700.

[0225] The CPU 1100 operates based on programs stored in the ROM 1200 or HDD 1400, and controls various parts. The ROM 1200 stores boot programs executed by the CPU 1100 when the computer 1000 starts up, as well as programs that depend on the computer 1000's hardware.

[0226] The HDD 1400 stores programs executed by the CPU 1100, as well as data used by such programs. The communication interface 1500 receives data from other devices via the communication network 500 (corresponding to network N in this embodiment) and sends it to the CPU 1100, and also transmits data generated by the CPU 1100 to other devices via the communication network 500.

[0227] The CPU 1100 controls output devices such as displays and printers, and input devices such as keyboards and mice, via the input / output interface 1600. The CPU 1100 acquires data from input devices via the input / output interface 1600. The CPU 1100 also outputs the data it generates to output devices via the input / output interface 1600.

[0228] The media interface 1700 reads a program or data stored in the recording medium 1800 and provides it to the CPU 1100 via the RAM 1300. The CPU 1100 loads the program from the recording medium 1800 onto the RAM 1300 via the media interface 1700 and executes the loaded program. The recording medium 1800 is, for example, an optical recording medium such as a DVD (Digital Versatile Disc) or PD (Phase Change Rewritable Disk), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory.

[0229] For example, when computer 1000 functions as information processing device 10A, the CPU 1100 of computer 1000 implements the functions of control unit 40A by executing a program loaded onto RAM 1300. The HDD 1400 stores the data from the storage device of information processing device 10A. The CPU 1100 of computer 1000 reads and executes these programs from the recording medium 1800, but as another example, these programs may be obtained from other devices via a predetermined communication network.

[0230] [13. Other] Although some embodiments of the present invention have been described in detail above with reference to the drawings, these are illustrative examples, and the present invention can be implemented in various other forms with modifications and improvements based on the knowledge of those skilled in the art, starting with the embodiments described in the disclosure section of the invention.

[0231] Furthermore, the configuration of the aforementioned information processing device 10A can be flexibly changed, for example, by calling external platforms, etc., via APIs (Application Programming Interfaces) or network computing, depending on the function.

[0232] Furthermore, the term "part" in the claims can be replaced with "means," "circuit," etc. For example, "reception part" can be replaced with "reception means" or "reception circuit." [Explanation of symbols]

[0233] 10, 10A Information Processing Device 20 Communications Department 30, 30A storage section 31, 31A Evaluator Information Database 32, 32A User Information Database 40, 40A Control Unit 41, 41A Reception Desk 42, 42A generation section 43, 43A Estimation part 44, 44A supply department 45 Judgment section 46 Output section 47, 47A Acquisition Department 48, 48A update section 100 User Terminals 200 evaluator terminals

Claims

1. An evaluation of an image containing elements that constitute fashion, comprising a distributed representation space generated based on evaluations from multiple evaluators indicating whether the elements that constitute the fashion suit the evaluator, and evaluator information indicating the attribute information of the evaluator, wherein the distributed representation space onto which the image and the evaluator information are projected, and an estimation unit that estimates the elements that constitute fashion that suit a predetermined subject based on subject information indicating the attribute information of a predetermined subject, A providing unit that provides to the user information regarding the elements constituting fashion estimated by the estimation unit. It has, The aforementioned designated persons are the users or persons who have a predetermined relationship with the users. An information processing device characterized by the following:

2. The estimation unit, An evaluation of an image containing multiple elements that constitute fashion, wherein a distributed representation space is generated based on evaluations from multiple evaluators indicating whether or not the multiple elements that constitute the fashion suit the evaluator, and evaluator information indicating the attribute information of the evaluator, wherein the distributed representation space onto which the image and the evaluator information are projected, and subject information indicating the attribute information of a predetermined subject, the combination of multiple elements that constitute fashion that suits the predetermined subject is estimated. The aforementioned supply unit is, The estimation unit provides the user with information about combinations of multiple elements that constitute fashion. The information processing apparatus according to feature 1.

3. The estimation unit, An evaluation of an image showing how to wear the elements that constitute fashion, wherein a distributed representation space is generated based on evaluations from multiple evaluators indicating whether the way the elements that constitute fashion are worn suits the evaluator, and evaluator information indicating the attribute information of the evaluator, wherein the distributed representation space onto which the image and the evaluator information are projected, and subject information indicating the attribute information of a predetermined subject, estimates how to wear the elements that constitute fashion that suit the predetermined subject, The aforementioned supply unit is, The estimation unit provides the user with information on how to wear the elements that make up the fashion estimated by the estimation unit. The information processing apparatus according to feature 1.

4. The evaluator information, which shows the attribute information of the evaluator, is evaluator information that shows the physical information of the evaluator. The subject information indicating the attribute information of the predetermined subject is subject information indicating the physical information of the predetermined subject. The information processing apparatus according to feature 1.

5. An evaluation of an image containing elements that constitute fashion, with a reception unit that receives evaluations from multiple evaluators indicating whether or not the elements that constitute the fashion suit the evaluator, A generation unit generates a distributed representation space onto which the image and the evaluator information are projected, based on the evaluation received by the reception unit and evaluator information indicating the attribute information of the evaluator. Based on the distributed representation space generated by the generation unit, an estimation unit estimates the elements that constitute fashion suitable for a predetermined target person. The aforementioned designated target persons are users who provide information regarding the elements constituting fashion estimated by the estimation unit, or persons who have a predetermined relationship with such users. An information processing device characterized by the following:

6. The person having a predetermined relationship with the user is a person different from the user, and is a person recognized by the user. The information processing apparatus according to claim 1 or 5.

7. A method of information processing performed by a computer, An evaluation of an image containing elements that constitute fashion, comprising a distributed representation space generated based on evaluations from multiple evaluators indicating whether the elements that constitute the fashion suit the evaluator, and evaluator information indicating the attribute information of the evaluator, wherein the distributed representation space onto which the image and the evaluator information are projected, and an estimation step of estimating the elements that constitute fashion that suit a predetermined subject based on subject information indicating the attribute information of a predetermined subject, A provisioning step which provides to the user information about the elements that constitute fashion estimated by the estimation step, and Includes, The aforementioned designated persons are the users or persons who have a predetermined relationship with the users. An information processing method characterized by the following:

8. An evaluation of an image containing elements that constitute fashion, comprising a distributed representation space generated based on evaluations from multiple evaluators indicating whether the elements of the fashion suit the evaluator, and evaluator information indicating the attribute information of the evaluator, wherein the distributed representation space onto which the image and the evaluator information are projected, and an estimation procedure for estimating the elements that constitute fashion that suit a predetermined subject, based on subject information indicating the attribute information of a predetermined subject, A provision procedure for providing users with information about the elements that constitute fashion estimated by the estimation procedure described above. Have the computer run it, The aforementioned designated persons are the users or persons who have a predetermined relationship with the users. An information processing program characterized by the following features.

9. A method of information processing performed by a computer, An evaluation of an image containing elements that constitute fashion, comprising a receiving process that receives evaluations from multiple evaluators indicating whether or not the elements that constitute the fashion suit the evaluator, A generation step generates a distributed representation space onto which the image and the evaluator information are projected, based on the evaluation received in the aforementioned reception step and evaluator information indicating the attribute information of the evaluator. An estimation step is performed to estimate the elements that constitute fashion suitable for a given target person, based on the distributed representation space generated by the generation step. Includes, The aforementioned designated persons are users who provide information regarding the elements constituting fashion estimated by the estimation process, or persons who have a predetermined relationship with such users. An information processing method characterized by the following:

10. An evaluation procedure for an image containing elements that constitute fashion, in which evaluations indicating whether or not the elements that constitute the fashion suit the evaluator are received from multiple evaluators, A generation procedure for generating a distributed representation space onto which the image and the evaluator information are projected, based on the evaluation received through the above-mentioned reception procedure and evaluator information indicating the attribute information of the evaluator, An estimation procedure for estimating the elements that constitute fashion suitable for a given subject, based on the distributed representation space generated by the above generation procedure, Have the computer run it, The aforementioned designated persons are users who provide information on the elements constituting fashion estimated by the estimation procedure, or persons who have a predetermined relationship with such users. An information processing program characterized by the following features.

Images