JP2022140055A - Dress derivation system - Google Patents

Abstract

To provide a dress derivation system that proposes a dress in which comfort is taken into consideration.SOLUTION: A dress derivation system comprises: a processing unit that outputs performance required for a dress based on information on the weather of the day on which a user has clothing on, sensitivity information representing how the user feels about the weather in relation to a dress, and information on the user's schedule; a clothing database that stores information on dresses to be proposed to the user; a derivation unit that, based on the output from the processing unit, selects clothing to be proposed to the user from the clothing database; and a display that displays the selected clothing.SELECTED DRAWING: Figure 1

Description

The present invention relates to an outfit derivation system that proposes outfits.

Techniques for supporting coordination of clothes have been developed. For example, Patent Literature 1 discloses a virtual try-on system that generates a virtual try-on image by synthesizing a try-on person image and an image of clothes.

JP 2019-144890 A

Many people worry about choosing clothes that match the season and weather. Clothing suitable for the season is on sale, but it is not always the best for the user to wear clothes as recommended by the store, since each person has different feelings about the weather and temperature. The surrounding environment of the user varies depending on the season, the weather, and the place where the user is active, such as whether the user is outdoors or indoors during the day. Therefore, it may not be possible to feel comfortable with clothes suggested only based on weather and temperature information. SUMMARY OF THE INVENTION It is an object of the present invention to provide a system for suggesting clothing that takes comfort into account.

The clothing derivation system of the present invention includes a weather information acquisition unit that acquires information about the weather on the day the clothing is worn, a sensitivity information acquisition unit that acquires sensitivity information that indicates how the user feels about the weather related to the clothing, and a user's a schedule information acquisition unit that acquires information about a schedule; a processing unit that outputs performance required for clothing based on the information acquired by the weather information acquisition unit, the sensitivity information acquisition unit, and the schedule information acquisition unit; a clothing database that stores information on clothing to be proposed to a user; a deriving unit that selects clothing to be proposed to the user from the clothing database based on the output of the processing unit; and a display unit that displays the selected clothing. Prepare.

To provide a system for proposing clothes considering comfort.

1 is a configuration example of an outfit derivation system related to Embodiment 1; 4 is a flowchart showing an example of the flow of processing related to the first embodiment; FIG. 10 is a configuration example of an outfit derivation system related to Embodiment 2; FIG. 9 is a flowchart showing an example of the flow of processing related to the second embodiment; 9 is a flowchart showing an example of the flow of processing related to the second embodiment; A configuration example of an outfit derivation system related to the third embodiment.

Hereinafter, embodiments will be described in detail with reference to the accompanying drawings. In addition, the following embodiments do not limit the invention according to the scope of claims. Although multiple features are described in the embodiments, not all of these multiple features are essential to the invention, and multiple features may be combined arbitrarily. Furthermore, in the accompanying drawings, the same or similar configurations are denoted by the same reference numerals, and redundant description is omitted.

(Example 1)
The clothing derivation system of the present disclosure has an information processing device 100 and an external server 101 . Information processing apparatus 100 and external server 101 are connected via a communication line. The information processing device 100 is, for example, a personal computer, a tablet terminal, or a smart phone. Information processing apparatus 100 includes input unit 110 , communication unit 111 , storage unit 112 , photographing unit 113 , control unit 120 , and display unit 130 . Input unit 110, communication unit 111, storage unit 112, photographing unit 113, display unit 130, and control unit 120 are connected so as to be able to exchange data and signals.

The input unit 110 can receive various operation instructions from the user. The input unit 110 is, for example, a touch sensitive display or keyboard. The communication unit 111 communicates with the external server 101 via a communication line. The storage unit 112 stores various data. The storage unit 112 has a clothing image storage unit 112A that stores images of clothing. The storage unit 112 also has a clothes database (clothes DB) 112B that stores clothes. Here, when the clothes are suggested from among the clothes owned by the user, the clothes owned by the user are stored in the storage unit 112 . Clothing image storage unit 112A may store an image of clothing photographed by imaging unit 113 of information processing apparatus 100 by the user.

A clothes database (clothes DB) 112B can store captured clothes images in association with weather parameters, which are numerical values representing performance such as resistance to cold and heat required for the clothes. Further, the clothing DB 112B may calculate the weather parameters of the clothing based on the attributes of the clothing obtained by analyzing the photographed image by the control unit 120, and store the weather parameters in association with the image of the clothing. . Clothing analysis may be performed using a learning model whose input is a photographed image of the clothing. In that case, the learning model can classify and output clothes according to the attributes of the input image. Clothing attributes can include the type of clothing such as outerwear and innerwear, and the material, thickness, color, pattern, and feel of the clothing analyzed from the image. Attributes may include clothing usage. A comment by the user may be attached to the attribute information of the clothing. Comments by the user can be input from the input unit. Comments may include opinions about the garment, such as likes or dislikes. The impression may be expressed in words or sentences, or may be expressed by predetermined symbols.

The control unit 120 can have a first learning model unit 121 , a derivation unit 122 , a clothing information acquisition unit 123 and an image acquisition unit 124 . The first learning model unit 121 functions as a processing unit that receives weather information, schedule information, and sensitivity information representing the user's perception of the weather as input data and outputs weather parameters, which are the performance required for clothing. . More specifically, it is a learned model generated so as to input weather information, schedule information, and sensitivity information, and output weather parameters based on these information. The weather parameters output in this manner reflect not only weather and schedule information but also how the weather is perceived.

The first learning model unit 121 may be provided in the external server 101 . In that case, the information processing apparatus 100 may include an application functioning as an interface with the external first learning model unit 121 . The derivation unit 122 may also be provided in the external server 101 in the same manner as the first learning model 121 , and the information processing apparatus 100 may include an application functioning as an interface for the derivation unit 122 .

The derivation unit 122 derives clothes to be proposed using the clothing information and the weather parameters output by the first learning model unit 121 . The derivation unit 122 may be configured as a learning model that selects and outputs clothing based on weather parameters and clothing attribute information. The display unit 130 acquires the image of the outfit proposed by the derivation unit 122 from the storage unit 112 via the image acquisition unit 124 and displays it. The display unit 130 is, for example, a smartphone display.

Clothes information acquisition section 123 acquires clothes information in which clothes images are associated with weather parameters from clothes DB 112</b>B of storage section 112 , and outputs the clothes information to derivation section 122 . The clothing information acquisition unit 123 can acquire clothing information in response to a request from the derivation unit 122 . The image acquisition unit 124 acquires the image of the clothing derived by the derivation unit 122 from the clothing image storage unit 112A.

Weather information may include time of day, location, weather, temperature, humidity and wind speed. The weather information may be information based on the weather forecast for the date and time of the appointment, which is obtained based on the schedule information. Weather forecast information can be obtained via an external server or communication line. In that case, the control unit 120 functions as a weather information acquisition unit via the communication unit 111 .

The schedule information is information that associates a location and a date and time when the user is scheduled to be active. The schedule information may include information such as whether the clothes are to be worn outdoors or indoors, and whether the clothes are for formal or casual use such as watching sports. The input unit may display an input screen for inputting schedule information. At this time, the control unit 120 functions as a schedule information acquisition unit.

The information on how the user feels about the weather is, for example, information indicating whether the user is sensitive to heat or cold. Clothing includes outerwear, tops, bottoms, innerwear, and the like. Clothing is a combination of clothes.

A weather parameter is a numerical value for deriving clothes necessary to protect against heat and cold, and weather parameters are set for each piece of clothes. As described above, the weather parameters may be calculated by the control unit 120 when the clothes are photographed, may be set by the user, or may be changed according to the user's evaluation. The sensitivity information may be set in advance by how the user feels about the weather, or may be set by the user by evaluating the suitability of the clothes worn for the weather. For example, if the user feels colder or hotter than expected, the user's evaluation may be analyzed and fed back to the clothing DB and used to modify the weather parameters for the clothing. The control unit 120 functions as a sensitivity information acquisition unit when sensitivity information is set.

An example of the processing flow of the clothing derivation system according to this embodiment will be described with reference to FIG. In step S<b>201 , weather information is acquired from the external server 101 using the communication unit 111 . In step S202, schedule information is acquired using the input unit 110. FIG. In step S203, the user's sensitivity information is acquired using the input unit 110 . In step S204, weather parameters are output using the first learning model unit 121 based on weather information, schedule information, and sensitivity information.

In step S205, it is determined whether or not to photograph clothes. When photographing clothes, the photographing unit 113 is used to photograph the clothes in step S206. In step S<b>207 , the photographed image of the clothes is saved in the storage unit 112 . The image of the clothes and the storage in the storage unit 112 may be performed before starting this flow. You can take pictures of all the clothes you own.

In step S208, the clothing information acquisition unit 123 acquires clothing information from the clothing DB 112B. In step S209, using the derivation unit 122, clothes are derived from the clothes information and weather parameters. Clothes are selected so that the total value of the weather parameters of the clothes satisfies the target weather parameters of the clothes output from the first learning model unit 121 . The derivation can be based on a combinatorial optimization algorithm. Also, the derivation unit 122 may be configured as a learning model that outputs clothing based on weather parameters. The derivation unit 122 can request clothes having predetermined weather parameters from the clothes information acquisition unit 123 . The derivation unit 122 can request clothing information from the clothing information acquisition unit 123 until appropriate clothing can be derived. In step S210, display unit 130 displays the derived clothes.

The deriving unit 122 acquires the information of the clothes of the previous day on the day when the clothes are proposed from the clothes DB 112B, compares the derived clothes with the clothes of the previous day, and derives the clothes again when the clothes are worn. good. Alternatively, a plurality of clothes may be derived as candidates, and the previous day's clothes and non-wearing clothes may be derived from them. The deriving unit 122 may derive a plurality of clothes as candidates. When a plurality of clothes are displayed on the display unit 130, a predetermined number of clothing candidates with numerical values close to the target weather parameter may be proposed from among clothes candidates that satisfy the weather parameter.

The user can edit the clothes displayed by the display unit 130 by operating the input unit 110 . When the display unit 130 is a touch-sensitive display, an input button for instructing editing may be displayed together with the derived clothing display on the display unit 130 . The input button may have the function of changing clothes or specifying clothes. For example, when a down coat is displayed as an outer, use the input button to change the outer coat, specify the down coat, search the clothing DB to display other coats, and select the desired coat from among them. Select and replace with down coat. In this case, the weather parameters of the clothes after the change may be displayed to indicate whether or not the target values are met.

Further, the derivation unit 122 may suggest clothes in consideration of the weather and clothes of the previous day. In that case, the derivation unit 122 may obtain weather parameters based on the difference in weather from the previous day. In one example, derivation unit 122 modifies weather parameters to suggest clothing based on temperature differences. Further, by inputting schedule information for a plurality of consecutive days, the derivation unit 122 may continuously derive outfits for a plurality of consecutive days. The derivation unit 122 may be configured with a learning model.

As described above, by reflecting the user's perception of the weather in the suggested clothes, it is possible to derive the clothes for the user to spend comfortably.

(Example 2)
In the present embodiment, a numerical value (comfort information) indicating the comfort level evaluated by the user for a combination of weather information, schedule information, and clothing information for a given day is input to the second learning model unit 125 to determine the comfort level. Find parameters. The comfort information is a numerical value of how the user feels about the clothes worn, and is a value that can be set by the user. The comfort parameter is a parameter that indicates the user's comfort with respect to weather and schedule-related clothes. As the input of the first learning model 121, the comfort parameter output by the second learning model unit 125 is input instead of the sensitivity information. The above two points are different from the first embodiment.

FIG. 3 is a diagram showing a configuration example of an outfit derivation system according to this embodiment. The comfort DB 112C stores comfort parameters output by the second learning model unit 125 . The comfort parameter acquisition unit 126 retrieves and acquires comfort parameters that match the conditions (schedule, weather, etc.) that the user wants to propose from the comfort DB 112C. The acquired comfort parameter is input to the first learning model unit 121 together with the weather, schedule, and the like. The first learning model unit 121 outputs weather parameters based on the input information.

The flow of processing for outputting comfort parameters according to this embodiment will be described with reference to FIG. In step S401, using the communication unit 111, weather information for a day specified by the user is acquired from the external server 101. FIG. The date specified by the user is the date on which the user sets the comfort information. Next, in step S402, the input unit 110 is used to input schedule information for the day specified by the user. The schedule information may also be acquired from the external server 101 or the storage unit 112 that stores the information of the designated day.

In step S403, the input unit 110 is used to input information about clothes worn on a day specified by the user. Clothing information may also be obtained from the external server 101 or the storage unit 112 that stores the information of the designated day. In step S404, the input unit 110 is used to input comfort information that the user felt on the day specified by the user. For the comfort information, the user's feeling is input, for example, on a scale of 1 to 10. As described above, the weather information, schedule information, clothing information, and comfort information about them for the day specified by the user are input to the second learning model unit 125 . In step S405, the second learning model unit 125 outputs comfort parameters classified based on weather information, schedule information, clothing information, and comfort information based on these input information. In step S406, comfort parameters are registered in the comfort DB 112C.

Next, the flow of clothes derivation processing will be described with reference to FIG. First, the user inputs the date and time for which he/she would like clothing suggestions. In step S501, weather information is acquired from the external server 101 via the communication unit 111 based on the input date and time information. In step S502, schedule information is input from the input unit 110. FIG. In step S503, the comfort parameter acquisition unit 126 acquires a comfort parameter that matches the weather information and the schedule from the comfort DB 112C.

Also, in step S504, the input weather, schedule, and obtained comfort parameters are input to the first learning model 121 . Here, the first learning model unit 121 is a learning model trained to output weather parameters based on input information.

In step S505, it is determined whether to photograph the image of the clothes. When photographing a clothing image, the photographing unit 113 is used to photograph the clothing in step S206. In step S<b>507 , the clothing image is saved in storage unit 112 . The image of the clothes may be photographed and stored in the storage unit 112 before starting this flow.

In step S508, the clothes information acquisition unit 123 is used to acquire clothes information from the clothes DB 112B. In step S509, using the deriving unit 122, clothes are derived from the clothes information and weather parameters. The clothes are derived so that the total value of the weather parameters of the clothes satisfies the target weather parameters output from the first learning model unit 121 . The derivation may be based on a combinatorial optimization algorithm. Also, the derivation unit 122 can request clothes having predetermined weather parameters from the clothes information acquisition unit 123 . The derivation unit 122 can request clothing information from the clothing information acquisition unit 123 until appropriate clothing can be derived. In step S510, the derived clothes are displayed using the display unit 130. FIG.

Also, the comfort parameter can be said to be information associated with the weather and clothes. Therefore, it is also possible to input the weather and schedule of the day for which the user wants to propose clothes, and to propose clothes associated with good values among the comfort parameters that match them from the comfort DB 112C. The storage space of the comfort DB 112C is grouped according to the temperature of the weather information, which is advantageous for retrieval of the comfort parameter by the comfort parameter acquisition unit 126. FIG. Since items with poor comfort parameters are basically excluded from selection, storing only items with good comfort parameters in the comfort DB 112C is advantageous in reducing the amount of memory used.

As described above, by reflecting the user's perception of the weather, it is possible to derive clothes that the user can wear comfortably.

(Example 3)
This embodiment includes a sensitivity information estimation unit 127, and differs from the above-described embodiments in that sensitivity information is estimated from the BMI value calculated from the user's height and weight. An embodiment of the present invention will be described with reference to FIG.

The sensitivity information estimation unit 127 calculates a BMI value from the user's height and weight input using the input unit 110, and estimates sensitivity information from the calculated BMI value. The BMI value is called a body mass index, and is a body mass index that indicates the degree of obesity calculated from weight and height. As an example, the sensitivity information estimation unit 127 estimates that the person is sensitive to heat when the BMI value is high and sensitive to cold when the BMI value is low. As described above, by reflecting the user's perception of the weather, it is possible to derive clothes that the user can wear comfortably. The sensitivity information estimated by the sensitivity information estimation unit 127 is input to the first learning model unit 121 .

This embodiment does not require the user to input how they feel about the weather. Therefore, for example, a store clerk can estimate and input a customer's BMI value at a store. The obtained clothes information can be used when recommending clothes in the store to customers.

In the flow of processing described above, the order of each step is not limited to the above order and may be changed. For example, the photographing of clothes performed in step S206 of FIG. 2 or step S506 of FIG.

The invention is not limited to the embodiments described above, and various modifications and variations are possible without departing from the spirit and scope of the invention. Accordingly, the claims are appended to make public the scope of the invention.

100: information processing device, 101: external server, 110: input unit, 111: communication unit, 112: storage unit, 112A: clothing image storage unit, 112B: clothing DB, 113: photographing unit, 120: control unit, 121: First Learning Unit 122: Derivation Unit 123: Clothing Information Acquisition Unit 124: Image Acquisition Unit 130: Display Unit

Claims (10)

a weather information acquisition unit that acquires information about the weather on the day the clothes are worn;
a sensitivity information acquisition unit that acquires sensitivity information representing how the user feels about the weather in relation to clothes;
a schedule information acquisition unit that acquires information about a user's schedule;
a processing unit that outputs performance required for clothing based on the information acquired by the weather information acquisition unit, the sensitivity information acquisition unit, and the schedule information acquisition unit;
a clothing database that stores information on clothing to be proposed to the user;
a derivation unit that selects clothes to be proposed to a user from the clothes database based on the output of the processing unit;
and a display for displaying the selected clothing. 2. The method of claim 1, wherein the sensitivity information includes parameters obtained based on the clothing worn, the weather when the user wears the clothing, and a value representing the user's perceived comfort with the clothing worn. Outfit derivation system. 3. The outfit derivation system according to claim 1, wherein the derivation unit selects the clothing to be proposed to the user from the clothing database based on the weather and the clothing of the day before the wearing date. 4. The outfit derivation system according to claim 1, wherein the derivation unit selects a plurality of outfits. 5. The clothing derivation system according to claim 1, wherein the derivation unit evaluates compatibility between the selected clothing and the weather, and the display unit displays the evaluation. 6. The apparatus according to any one of claims 1 to 5, further comprising an input unit for inputting another outfit based on the displayed outfit, and displaying an evaluation of the input different outfit with respect to weather. The clothing derivation system described in . 7. The derivation unit according to any one of claims 1 to 6, wherein, when the wearing schedule includes a plurality of consecutive days, the derivation unit proposes clothes for each of the consecutive days. Apparel derivation system as described. 8. The clothing database according to any one of claims 1 to 7, wherein the clothing database receives clothing images, analyzes the images to obtain clothing attributes, and stores clothing information in association with the clothing images. The clothing derivation system according to item 1. 9. The outfit derivation system according to any one of claims 1 to 8, wherein the clothing database stores a date on which the clothing was worn in association with the clothing. 10. The clothing derivation system according to any one of claims 1 to 9, wherein the sensitivity information acquisition unit further calculates how the user perceives the weather based on a BMI value calculated from the user's height and weight. .

Images