Clothing recommendation method, device, equipment and system

Abstract

The present application belongs to the field of image technology, and specifically relates to a clothing recommendation method, device, equipment and system. The present application aims to solve the problem of poor recommendation accuracy in existing clothing recommendation methods. By collecting user image information, the corresponding clothing information is determined based on the user image information, wherein the clothing information includes multiple clothing elements, and the associated matching features between the multiple clothing elements are determined, wherein the associated matching features are used to characterize the image features of each clothing element after matching, and the clothing recommendation information is determined based on the associated matching features. Since the associated matching features between the multiple clothing elements are obtained based on the multiple clothing elements in the clothing information, the correlation and integrity between different clothing elements can be better reflected, and the user's clothing characteristics can be better reflected. On this basis, clothing recommendations are made, which improves the accuracy of the clothing recommendation results and the clothing recommendation effect.

Description

Technical Field

[0001] The present application belongs to the field of image technology, and specifically relates to a clothing recommendation method, device, equipment and system. Background Art

[0002] With the progress and development of society, people's clothing is no longer just a means and tool to resist the cold, but also a symbol of people's taste and culture. Good clothing matching can often highlight a person's temperament and physical beauty. However, not everyone is an expert in clothing matching. For many ordinary people, it is not easy to choose and match clothing properly.

[0003] Therefore, in related technologies, in order to solve the problem of people's clothing matching, some smart terminal devices, such as smart phones, laptops, etc., can be used to collect users' clothing information and recommend clothing based on users' appearance characteristics or current popular styles.

[0004] However, in the solutions of the related art, the smart terminal device only recommends a piece of clothing based on simple features in the user's clothing image data, which has the problem of low accuracy and poor effect of clothing recommendation results.

[0005] Accordingly, the art requires a new clothing recommendation method, device, equipment and system to solve the above problems. Summary of the Invention

[0006] In order to solve the above-mentioned problems in the prior art, that is, to solve the problems of low accuracy and poor effect of existing clothing recommendation results, the present application provides a clothing recommendation method, device, equipment and system.

[0007] According to a first aspect of the embodiments of the present application, the present application provides a clothing recommendation method, comprising:

[0008] Collect user image information; determine corresponding clothing information based on the user image information; wherein the clothing information includes multiple clothing elements; determine associated matching features between the multiple clothing elements, wherein the associated matching features are used to characterize the image features of the clothing elements after matching; determine clothing recommendation information based on the associated matching features.

[0009] In the preferred technical solution of the above-mentioned clothing recommendation method, the clothing recommendation information includes target clothing information; and determining the clothing recommendation information based on the associated matching features includes:

[0010] Obtaining a preset clothing matching strategy, where the clothing matching strategy is used to characterize a mapping relationship between the associated matching features and the clothing database;

[0011] Determining, according to the clothing matching strategy, a clothing database corresponding to the associated matching features;

[0012] A set of one or more clothing elements in the clothing database is determined as target clothing information.

[0013] In the preferred technical solution of the above clothing recommendation method, determining a set of one or more clothing elements in the clothing database as target clothing information includes:

[0014] Get the preset user appearance information;

[0015] Performing a matching degree test on the clothing elements in the clothing database according to the user appearance information to obtain a matching degree corresponding to each clothing element;

[0016] A set of clothing elements corresponding to the matching degree greater than a preset matching degree threshold is determined as target clothing information.

[0017] In the preferred technical solution of the above clothing recommendation method, determining corresponding clothing information based on the user image information includes:

[0018] Acquiring preset clothing characteristic parameters, wherein the clothing characteristic parameters are used to characterize the characteristics of the clothing information;

[0019] screening the plurality of image features according to the clothing feature parameters to determine a plurality of clothing elements, wherein the clothing elements match the clothing features;

[0020] The set of clothing elements is determined as the clothing information.

[0021] In the preferred technical solution of the above clothing recommendation method, each clothing element has a corresponding clothing feature, and determining the associated matching features between the multiple clothing elements includes:

[0022] Get the preset clothing element weight;

[0023] According to the weights of the clothing elements, weighted feature fusion is performed on the clothing features corresponding to the clothing elements to generate associated matching features.

[0024] In the preferred technical solution of the above-mentioned clothing recommendation method, the associated matching features include at least one of the following: color matching features between different clothing elements, style matching features between different clothing elements, and material matching features between different clothing elements.

[0025] According to a second aspect of the embodiments of the present application, the present application provides a clothing recommendation device, comprising:

[0026] An acquisition module is used to acquire user image information; a first determination module is used to determine corresponding clothing information based on the user image information; wherein the clothing information includes multiple clothing elements; a second determination module is used to determine the associated matching features between the multiple clothing elements, wherein the associated matching features are used to characterize the image features of the clothing elements after matching; and a third determination module is used to determine clothing recommendation information based on the associated matching features.

[0027] In the preferred technical solution of the above-mentioned clothing recommendation device, the clothing recommendation information includes target clothing information, and the third determination module is specifically used to: obtain a preset clothing matching strategy, where the clothing matching strategy is used to characterize the mapping relationship between the associated matching features and the clothing database; determine the clothing database corresponding to the associated matching features based on the clothing matching strategy; and determine a set of one or more clothing elements in the clothing database as target clothing information.

[0028] In the preferred technical solution of the above-mentioned clothing recommendation device, when the third determination module determines a set of one or more clothing elements in the clothing database as target clothing information, it is specifically used to: obtain preset user appearance information; perform a matching degree detection on the clothing elements in the clothing database based on the user appearance information to obtain the matching degree corresponding to each clothing element; and determine the set of clothing elements corresponding to the matching degree greater than a preset matching degree threshold as the target clothing information.

[0029] In the preferred technical solution of the above-mentioned clothing recommendation device, the first determination module is specifically used to: obtain preset clothing feature parameters, wherein the clothing feature parameters are used to characterize the characteristics of the clothing information; based on the clothing feature parameters, screen the multiple image features to determine multiple clothing elements, wherein the clothing elements match the clothing features; and determine the set of clothing elements as the clothing information.

[0030] In the preferred technical solution of the above-mentioned clothing recommendation device, each of the clothing elements has a corresponding clothing feature, and the second determination module is specifically used to: obtain a preset clothing element weight; and perform weighted feature fusion on the clothing features corresponding to the clothing elements according to the clothing element weight to generate associated matching features.

[0031] In the preferred technical solution of the above-mentioned clothing recommendation device, the associated matching features include at least one of the following: color matching features between different clothing elements, style matching features between different clothing elements, and material matching features between different clothing elements.

[0032] According to a third aspect of the embodiments of the present application, the present application provides an electronic device comprising: a memory, a processor, and a computer program; wherein the computer program is stored in the memory and is configured to be executed by the processor as the clothing recommendation method described in any one of the first aspects of the embodiments of the present application.

[0033] In the preferred technical solution of the above electronic device, the electronic device further includes: a camera component, which is connected to the processor and the memory, and is used to obtain user image information and send the user image information to the memory and / or processor.

[0034] According to the fourth aspect of the embodiment of the present application, the present application provides a smart home system, including a display screen device and the electronic device provided by the fourth aspect of the embodiment of the present application, the display screen device is communicatively connected to the electronic device, wherein the display screen device is used to receive and display clothing recommendation information sent by the electronic device.

[0035] According to the fifth aspect of the embodiments of the present application, the present application provides a computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are executed by a processor, they are used to implement the clothing recommendation method as described in any one of the first aspects of the embodiments of the present application.

[0036] The clothing recommendation method, device, equipment and system provided by the present application collect user image information, determine corresponding clothing information based on the user image information, wherein the clothing information includes multiple clothing elements, determine the associated matching features between the multiple clothing elements, wherein the associated matching features are used to characterize the image features of the various clothing elements after matching, and determine clothing recommendation information based on the associated matching features. Since the associated matching features between the multiple clothing elements are obtained based on the multiple clothing elements in the clothing information, the correlation and integrity between different clothing elements can be better reflected, and the user's clothing characteristics can be better reflected. Clothing recommendations are made on this basis, thereby improving the accuracy of the clothing recommendation results and the clothing recommendation effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The following describes preferred embodiments of the clothing recommendation method, device, apparatus, and system of the present application with reference to the accompanying drawings.

[0038] Figure 1 This is a diagram of an application scenario of the clothing recommendation method provided in an embodiment of the present application;

[0039] Figure 2 A flowchart of a clothing recommendation method provided in one embodiment of the present application;

[0040] Figure 3 A flowchart of a clothing recommendation method provided in another embodiment of the present application;

[0041] Figure 4 for Figure 3 Flowchart of step S209 in the illustrated embodiment;

[0042] Figure 5 A schematic diagram of the structure of a clothing recommendation device provided in one embodiment of the present application;

[0043] Figure 6 A schematic structural diagram of a clothing recommendation device provided in another embodiment of the present application;

[0044] Figure 7 A schematic diagram of an electronic device provided in accordance with an embodiment of the present application. DETAILED DESCRIPTION

[0045] First, those skilled in the art should understand that these embodiments are merely used to explain the technical principles of this application and are not intended to limit the scope of protection of this application. Those skilled in the art may adjust them as needed to suit specific applications.

[0046] The following exemplary embodiments are described in detail, with examples illustrated in the accompanying drawings. In the following description, when referring to the drawings, identical numbers in different figures represent identical or similar elements, unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with the present application. Rather, they are merely examples of apparatuses and methods consistent with certain aspects of the present application, as detailed in the appended claims.

[0047] First, the application scenarios of the embodiments of the present application are explained:

[0048] Figure 1 This is an application scenario diagram of the clothing recommendation method provided in the embodiment of the present application, such as Figure 1 As shown, the execution subject of the clothing recommendation method provided in the embodiment of the present application can be an electronic device, such as a smart camera. The smart camera can be a standalone smart terminal device or part of a smart home system. The smart camera collects user image information, such as a full-body photo or video of the user, and obtains clothing recommendation information suitable for the user by performing offline analysis of the user image information, such as a set of clothing photos. The clothing photos are then displayed on a display screen that is communicatively connected to the smart camera, allowing the user to obtain clothing information suitable for themselves, either individually or in combination.

[0049] Currently, in related technologies, the technical solutions for recommending clothing to users usually use pictures of single clothing items worn or specified by the user, and then make matching recommendations based on the clothing pictures based on big data. However, when users choose clothing combinations, one or two pieces of clothing cannot reflect the user's true dressing purpose or dressing style. For example, there are many scenarios where users need to wear sneakers, such as outdoor sports or shopping in a mall. In this case, if clothing recommendations are made based solely on the user's sports shoes, the recommendation results may be inaccurate. In the above technical solution, since the connection and integrity between the various clothing items when the user is already wearing the specified clothing are not taken into account, it is impossible to accurately recommend suitable and matching clothing information to the user, resulting in inaccurate recommendations such as the recommended clothing being inappropriate or failing to meet the user's dressing needs.

[0050] The following specific embodiments describe in detail the technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0051] Figure 2 This is a flowchart of a clothing recommendation method provided by an embodiment of the present application. For example, the execution subject of the method may be a smart camera. Figure 2 As shown, the clothing recommendation method provided in this embodiment includes the following steps:

[0052] Step S101: Collect user image information.

[0053] Exemplarily, user image information is information used to describe the user's external features, and more specifically, includes one or more of the user's facial features, body features, and clothing features. Exemplarily, user image information can be implemented in various forms, such as one or more photos or videos containing the user's appearance, or as a three-dimensional profile scan of the user. Any information that can represent the user's external features is acceptable, and the specific implementation of user image information is not limited herein.

[0054] There are also many ways to collect user image information. For example, the image acquisition unit of the execution subject of the method provided by this embodiment is used to collect real-time user image information of the user. More specifically, for example, the execution subject of the method provided by this embodiment is a smart camera, and the smart camera collects the user's image information through an image acquisition unit including a lens. For example, image acquisition information can also be collected by other electronic devices that are communicatively connected to the execution subject of the method provided by this embodiment. For example, the execution subject of the method provided by this embodiment is a smart home terminal, and the smart home terminal is communicatively connected to the camera. The smart home terminal collects user image information through the camera and performs subsequent processing. For another example, the smart home terminal is connected to a storage medium, and the smart home directly reads the user image information from the storage medium. The specific implementation method of collecting user image information is not limited here.

[0055] Step S102: determining corresponding clothing information according to the user image information; wherein the clothing information includes a plurality of clothing elements.

[0056] User image information includes external clothing features of the user, such as the number, type, style, and color of the clothing items worn. Therefore, by processing and extracting features from the user image information, it is possible to identify information about the clothing items worn by the user, namely, clothing information. Clothing information includes multiple clothing elements. For example, user A is wearing white sweatpants and a pair of black soccer shoes, while user B is wearing a floral skirt and white socks. These features constitute the clothing information corresponding to user A and user B. User A's clothing information includes two clothing elements: white sweatpants and black soccer shoes; while user B's clothing information includes two clothing elements: a floral skirt and white socks. Of course, it is understood that clothing elements can be further refined, for example, by describing specific clothing elements using labels such as "soccer shoe a" and "soccer shoe b." In this embodiment, the specific implementation of clothing elements can be configured based on specific needs and is not specifically limited here.

[0057] Exemplarily, there are many methods for determining clothing information through user image information, such as performing feature recognition and classification on user image information through a convolutional neural network to obtain multiple clothing elements, and a collection of multiple clothing elements can be used as clothing information; at the same time, the corresponding user appearance information can also be determined by receiving instruction information input by the user, or a combination of the two. The specific implementation methods are existing technologies in this field and will not be repeated here.

[0058] Step S103 : determining the associated matching features between the plurality of clothing elements, wherein the associated matching features are used to represent the image features of the clothing elements after matching.

[0059] Exemplarily, the associated matching feature is feature information that reflects the association and integrity between clothing elements. For example, clothing element a is white sports shorts; clothing element b is sports shoes, then there is an associated matching feature A between clothing element a and clothing element b; at the same time, clothing element c is slippers; then there is an associated matching feature B between clothing element a and clothing element c; that is, since different clothing elements have different associations with each other after being combined, correspondingly, the associated matching feature A and the associated matching feature B are also different.

[0060] It should be noted that, since the associated matching features are calculated based on all clothing elements in the user's image information, the more clothing elements there are, the more complex the associated matching features become. While the above example illustrates the associated matching features between two clothing elements for ease of understanding, this does not necessarily mean that the associated matching features are limited to just two clothing elements; they can also be formed across multiple clothing elements.

[0061] Compared with related technologies, since the correlation and matching characteristics between multiple clothing elements are taken into consideration, it can further express the style of existing clothing, and then recommend clothing that matches the style of existing clothing.

[0062] Step S104: determining clothing recommendation information based on the associated matching features.

[0063] Since different clothing elements have different characteristics, the corresponding associated matching features after multiple clothing elements are matched and combined are also different. However, the associated matching features that take into account the matching characteristics between multiple clothing elements can better reflect the characteristics of the user's current clothing matching. Specifically, the associated matching features can be in the form of pixel information implemented in a matrix, or can be identifiers corresponding to different pixel information combinations. No specific limitation is made here. Furthermore, by inputting the associated matching features into the clothing recommendation model, corresponding clothing recommendation information can be obtained. Among them, the clothing recommendation model can be a mapping relationship table or logical expression preset by the user; it can also be a neural network model that has been trained to convergence and obtained through self-learning. The clothing recommendation model has the ability to map the associated matching features to clothing recommendation information. The specific model training method is the existing technology in this field and will not be repeated here.

[0064] In this embodiment, by collecting user image information, corresponding clothing information is determined based on the user image information, wherein the clothing information includes multiple clothing elements, and the associated matching features between the multiple clothing elements are determined, wherein the associated matching features are used to characterize the image features of each clothing element after matching, and clothing recommendation information is determined based on the associated matching features. Since the associated matching features between the multiple clothing elements are obtained based on the multiple clothing elements in the clothing information, the correlation and integrity between different clothing elements can be better reflected, and the user's clothing characteristics can be better reflected. Clothing recommendations are made on this basis, thereby improving the accuracy of the clothing recommendation results and the clothing recommendation effect.

[0065] Figure 3 A flowchart of a clothing recommendation method provided in another embodiment of the present application is shown as follows: Figure 3 As shown, the clothing recommendation method provided in this embodiment is Figure 2 Based on the clothing recommendation method provided in the illustrated embodiment, S102-S104 are further refined. The clothing recommendation method provided in this embodiment includes the following steps:

[0066] Step S201: Collect user image information.

[0067] Step S202: obtaining preset clothing characteristic parameters, wherein the clothing characteristic parameters are used to characterize the characteristics of clothing information.

[0068] For example, clothing characteristic parameters can be configuration information pre-entered by the user as needed. Specifically, the electronic device has an interactive interface through which the user interacts with the electronic device. The user enters configuration instructions containing clothing characteristic parameters, which the electronic device receives and stores in a storage medium. Clothing characteristic parameters are used to characterize different types of clothing information. Specifically, pixel combinations that meet the description of clothing characteristic parameters are considered to have clothing characteristics and can be used as clothing information. Clothing characteristic parameters can distinguish clothing information from non-clothing information in user image information.

[0069] In the steps of this embodiment, since the clothing feature parameters are preset by the user, the clothing information can be accurately defined by adjusting the clothing feature parameters. For example, the tops, bottoms, and shoes worn by the user are regarded as the components of the clothing information, and the associated matching features between the tops, bottoms, and shoes are evaluated, while the socks are distinguished from them. This can achieve more flexible clothing recommendations and better meet the user's clothing recommendation needs.

[0070] Exemplarily, each feature of clothing information corresponds to a unique identifier, and the clothing feature parameters have a mapping relationship with the identifier. According to the clothing feature parameters, the identifier corresponding to the feature of the clothing information can be determined, and then the feature of the clothing information can be determined.

[0071] Step S203 , screening multiple image features according to clothing feature parameters to determine multiple clothing elements, wherein the clothing elements match the clothing features.

[0072] Image features are filtered based on preset clothing feature parameters to identify target information within the user's image information that needs to be considered for clothing matching, namely clothing elements. For example, an electronic device collects user image information of user A, where the corresponding clothing information includes the following clothing elements: top, bottoms, gloves, scarf, and shoes. Only tops, bottoms, scarves, and shoes are considered as clothing elements for clothing matching evaluation, while gloves are excluded. This allows for more flexible clothing recommendations within certain restrictions.

[0073] For example, the method for screening multiple image features based on clothing feature parameters can be to use a pre-trained image recognition model to retrieve user image information based on image features, and obtain basic image features in all dimensions of the user image information that match the image features corresponding to the clothing information as clothing elements. The training and use of the image recognition model is known in the art and will not be further described here.

[0074] Step S204: determining the set of clothing elements as clothing information.

[0075] Specifically, clothing elements represent the clothing corresponding to the user's image information, such as a green top and gray sweatpants. Once the clothing elements corresponding to the clothing information are determined, the collection of multiple clothing elements, as a whole, can be used as the clothing information. Exemplarily, the collection of clothing elements can be a simple superposition of multiple clothing elements, or a collection obtained by feature fusion according to a preset algorithm. This is not specifically limited here and can be implemented according to specific needs.

[0076] Step S205: Obtain preset clothing element weights.

[0077] When it comes to matching clothing elements, we don't simply consider the characteristics of each element equally. Instead, we make recommendations based on their relationships. Among multiple clothing elements, there are primary and secondary elements. For example, a user's top often corresponds to a primary element, while shoes, in contrast, correspond to secondary elements.

[0078] Clothing element weights represent the primary and secondary relationships between clothing elements. For example, the normalized clothing element weight for a top is 0.8, and the normalized clothing element weight for shoes is 0.1. Based on clothing element weights, the importance of multiple clothing elements in an outfit can be reflected.

[0079] Exemplarily, there are many methods for determining the weights of clothing elements. The user can set them based on experience or preferences and preset the weight information in the electronic device; or the electronic device or other electronic device running the method provided in this embodiment can obtain weight information that conforms to current popular elements or the user's clothing preferences in a self-learning manner. No specific limitations are made here and it can be implemented as needed.

[0080] Step S206 , performing weighted feature fusion on clothing features corresponding to clothing elements according to clothing element weights to generate associated matching features.

[0081] Based on the weights of clothing elements, weighted feature fusion is performed on the clothing features corresponding to the clothing elements. Apparel elements with high weights play a dominant role in the overall clothing pairing, while clothing elements with low weights play a secondary role. After weighted feature fusion of multiple clothing elements, a set of information representing the overall and associated characteristics of the user's current clothing can be generated, namely, associated matching features. In this embodiment, by weighting different clothing elements using clothing element weights, the dominant and secondary elements among multiple clothing elements can be reflected, thereby forming associated matching features dominated by one or more clothing elements, thereby further improving the accuracy of clothing recommendations and better meeting the user's clothing recommendation needs.

[0082] Step S207: obtaining a preset clothing matching strategy, where the clothing matching strategy is used to represent a mapping relationship between associated matching features and a clothing database.

[0083] Exemplarily, the clothing database is a pre-set data file or database. The clothing database can be stored in a storage medium within the electronic device or in an external storage medium communicatively connected to the electronic device. The electronic device can access the pre-set clothing database via real-time or non-real-time means. The clothing database includes a variety of clothing elements with common characteristics. Each clothing database corresponds to an identifier, namely, a clothing database identifier. The clothing database identifier uniquely identifies a clothing database.

[0084] Step S208: determining a clothing database corresponding to the associated matching feature according to the clothing matching strategy.

[0085] The clothing matching strategy is a model that represents the mapping relationship between the associated matching features and the clothing database identifiers. According to the clothing matching strategy, the clothing database corresponding to the associated matching features can be uniquely determined.

[0086] Step S209: determining a set of one or more clothing elements in the clothing database as target clothing information.

[0087] For example, the clothing database includes multiple clothing elements that have common features, such as belonging to the same matching style, the same color, etc. By matching these clothing elements, a better clothing matching effect can be obtained. The collection of these clothing elements is the target clothing information.

[0088] For example, Figure 4 As shown, step S209 includes three specific implementation steps: step S2091, S2092 and S2093:

[0089] Step S2091: Obtain preset user appearance information.

[0090] User image information includes external features of the user, such as the user's face and torso. Therefore, by performing data processing and feature extraction on the user image information, information such as the user's face, hairstyle, body shape, etc., that is, user appearance information, can be identified. Different users have different user appearance information. For example, user image information A corresponds to the user appearance information of user A. User A is tall, has long brown hair, and has a fair complexion. The above features are the specific content of the corresponding description of the user appearance information. Of course, it is understandable that the user appearance information can be further refined, for example, by describing specific user appearance features through labels such as hairstyle a and hairstyle b. In this embodiment, the specific implementation method of the user appearance information can be set according to specific needs and is not specifically limited here.

[0091] Step S2092: performing a matching degree test on the clothing elements in the clothing database based on the user's appearance information to obtain the matching degree corresponding to each clothing element.

[0092] Since different users have different corresponding user appearance information, for example, some users are tall, and some users have fair skin, the clothing elements that match different user appearance information are also different. Therefore, the matching degree between the user appearance information and the clothing elements can be detected to determine the clothing elements that match the user appearance information.

[0093] Specifically, the electronic device retrieves multiple clothing elements from a clothing database, evaluates the degree of match between the user's appearance information and the clothing elements, and determines the degree of match between the user's appearance information and the clothing elements. For example, a specific implementation method may first label external data including the user's appearance information and clothing elements, then perform model training to obtain a trained and converged matching evaluation model. This matching evaluation model can receive the user's appearance information, evaluate the match with multiple preset clothing elements, and output the corresponding matching degree. The specific sample labeling and model training processes are conventional techniques in the art and will not be further described here.

[0094] Step S2093 : determining a set of clothing elements corresponding to a matching degree greater than a preset matching degree threshold as target clothing information.

[0095] Specifically, clothing elements corresponding to a matching degree greater than a matching degree threshold can be regarded as clothing elements that match the user's appearance features, and a set of these clothing elements is the target clothing information.

[0096] Among them, since the matching threshold can be set according to user needs, the higher the matching threshold, the less target clothing information is obtained, and the more accurate the matching result. Therefore, it can be flexibly set according to user needs, improving the flexibility of the method of this embodiment.

[0097] Figure 5 This is a structural diagram of a clothing recommendation device provided in one embodiment of the present application, as shown in FIG. Figure 5 As shown, the clothing recommendation device 3 provided in this embodiment includes:

[0098] The acquisition module 31 is used to acquire user image information.

[0099] The first determining module 32 is configured to determine corresponding clothing information according to the user image information; wherein the clothing information includes a plurality of clothing elements.

[0100] The second determining module 33 is used to determine the associated matching features between multiple clothing elements, wherein the associated matching features are used to represent the image features of each clothing element after matching.

[0101] The third determining module 34 is used to determine clothing recommendation information based on the associated matching features.

[0102] Optionally, the clothing recommendation information includes target clothing information. The third determination module 34 is specifically used to: obtain a preset clothing matching strategy, where the clothing matching strategy is used to characterize the mapping relationship between associated matching features and the clothing database; determine the clothing database corresponding to the associated matching features based on the clothing matching strategy; and determine a set of one or more clothing elements in the clothing database as target clothing information.

[0103] Optionally, when determining a set of one or more clothing elements in the clothing database as target clothing information, the third determination module 34 is specifically used to: obtain preset user appearance information; perform a matching degree detection on the clothing elements in the clothing database based on the user appearance information to obtain the matching degree corresponding to each clothing element; and determine the set of clothing elements corresponding to a matching degree greater than a preset matching degree threshold as the target clothing information.

[0104] Optionally, the first determination module 32 is specifically used to: obtain preset clothing feature parameters, wherein the clothing feature parameters are used to characterize the characteristics of clothing information; based on the clothing feature parameters, screen multiple image features to determine multiple clothing elements, wherein the clothing elements match the clothing features; and determine the set of clothing elements as clothing information.

[0105] Optionally, each clothing element has a corresponding clothing feature. The second determining module 33 is specifically used to: obtain a preset clothing element weight; perform weighted feature fusion on the clothing features corresponding to the clothing element according to the clothing element weight, and generate associated matching features.

[0106] Optionally, the associated matching features include at least one of the following: color matching features between different clothing elements, style matching features between different clothing elements, and material matching features between different clothing elements.

[0107] Figure 6 This is a structural diagram of a clothing recommendation device provided in another embodiment of the present application, such as Figure 6 As shown, the clothing recommendation device provided in this embodiment is Figure 5 Based on the clothing recommendation device provided in the illustrated embodiment, a display module 41 is added. The clothing recommendation device 4 provided in this embodiment further includes:

[0108] The display module 41 is used to send the clothing recommendation information to a display screen device for display.

[0109] The collection module 31, the first determination module 32, the second determination module 33, the third determination module 34, and the display module 41 are connected in sequence. Figure 2-Figure 4 The technical solutions of any of the method embodiments shown have similar implementation principles and technical effects, which will not be repeated here.

[0110] Figure 7 A schematic diagram of an electronic device provided in one embodiment of the present application is shown as follows: Figure 7 As shown, the electronic device 5 provided in this embodiment includes: a memory 51, a processor 52 and a computer program.

[0111] The computer program is stored in the memory 51 and is configured to be executed by the processor 52 to implement the present application. Figure 2-Figure 4 The clothing recommendation method provided by any corresponding embodiment.

[0112] The memory 51 and the processor 52 are connected via a bus 53 .

[0113] Optionally, the electronic device further includes: a camera component 54 , which is connected to the processor 52 and the memory 51 via a bus 53 , and is used to obtain user image information and send the user image information to the memory 51 and / or the processor 52 .

[0114] For related instructions, please refer to Figure 2-Figure 4 The relevant descriptions and effects corresponding to the steps in any corresponding embodiment can be understood, and no further details are given here.

[0115] One embodiment of the present application provides a computer-readable storage medium having a computer program stored thereon, which is executed by a processor to implement the present application. Figure 2-Figure 4 The clothing recommendation method provided in any one of the corresponding embodiments.

[0116] Among them, the computer-readable storage medium may be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, and the like.

[0117] In the several embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of modules is only a logical function division. In actual implementation, there may be other division methods, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or modules, which can be electrical, mechanical or other forms.

[0118] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the application disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present application that follow the general principles of the present application and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0119] It should be understood that the present application is not limited to the exact structure described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

[0120] Thus far, the technical solutions of the present application have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it is readily understood by those skilled in the art that the scope of protection of the present application is obviously not limited to these specific embodiments. Without departing from the principles of the present application, those skilled in the art may make equivalent changes or substitutions to the relevant technical features, and the technical solutions after such changes or substitutions will fall within the scope of protection of the present application.

Claims

1. A clothing recommendation method, characterized in that: Collect user image information; Determining corresponding clothing information according to the user image information; wherein the clothing information includes multiple clothing elements; Each of the clothing elements has a corresponding clothing feature, and a preset clothing element weight is obtained, wherein the clothing element weight represents information about the primary and secondary relationships between the clothing elements; Based on the clothing element weights, weighted feature fusion is performed on clothing features corresponding to all clothing elements in the user image information to generate associated matching features; wherein the associated matching features are used to characterize the overall image features after the clothing elements are matched; the associated matching features include at least one of the following: color matching features between different clothing elements, style matching features between different clothing elements, and material matching features between different clothing elements; Determine clothing recommendation information based on the associated matching features.

2. The method according to claim 1, characterized in that The clothing recommendation information includes target clothing information; Determining clothing recommendation information based on the associated matching features includes: Obtaining a preset clothing matching strategy, where the clothing matching strategy is used to characterize a mapping relationship between the associated matching features and a preset clothing database; Determining, according to the clothing matching strategy, a clothing database corresponding to the associated matching features; A set of one or more clothing elements in the clothing database is determined as target clothing information.

3. The method according to claim 2, characterized in that Determining a set of one or more clothing elements in the clothing database as target clothing information includes: Get the preset user appearance information; Performing a matching degree test on the clothing elements in the clothing database according to the user appearance information to obtain a matching degree corresponding to each clothing element; A set of clothing elements corresponding to the matching degree greater than a preset matching degree threshold is determined as target clothing information.

4. The method according to claim 1, wherein Determining corresponding clothing information based on the user image information includes: Acquiring preset clothing characteristic parameters, wherein the clothing characteristic parameters are used to characterize the characteristics of the clothing information; screening a plurality of image features according to the clothing feature parameters to determine a plurality of clothing elements, wherein the clothing elements match the clothing features; The set of clothing elements is determined as the clothing information.

5. A clothing recommendation device, characterized in that: include: An acquisition module, used to acquire user image information; A first determining module determines corresponding clothing information based on the user image information; wherein the clothing information includes a plurality of clothing elements; The second determination module obtains preset clothing element weights based on the corresponding clothing features of each clothing element, wherein the clothing element weights represent information about the primary and secondary relationships between clothing elements; performs weighted feature fusion on the clothing features corresponding to all clothing elements in the user image information based on the clothing element weights to generate associated matching features, wherein the associated matching features are used to represent the overall image features of the clothing elements after matching; the associated matching features include at least one of the following: color matching features between different clothing elements, style matching features between different clothing elements, and material matching features between different clothing elements; The third determining module determines clothing recommendation information according to the associated matching features.

6. An electronic device, characterized in that: include: memory, processors and computer programs; The computer program is stored in the memory and is configured to be executed by the processor to implement the clothing recommendation method according to any one of claims 1 to 4.

7. The electronic device according to claim 6, wherein: The electronic device further comprises: A camera component is connected to the processor and the memory, and is used to obtain user image information and send the user image information to the memory and / or the processor.

8. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, which, when executed by a processor, are used to implement the clothing recommendation method according to any one of claims 1 to 4.

9. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the clothing recommendation method according to any one of claims 1 to 4 is implemented.

Images