Computer-implemented method for the automated generation of a clothing suggestion and / or an accessory suggestion.

Abstract

The invention relates to a computer-implemented method for the automated generation of a clothing suggestion (3) and / or an accessory suggestion, comprising the steps - Providing a vehicle configuration data set (2) of a vehicle with a plurality of vehicle configuration data, - Processing the vehicle configuration data set (2) using artificial intelligence (1), wherein design-specific features of the vehicle are extracted from the vehicle configuration data set (2) and a clothing suggestion (3) and / or an accessory suggestion for a user of the vehicle is automatically generated from the design-specific features, - Displaying the clothing suggestion (3) and / or accessory suggestion generated in the previous step by means of a display device (4) and / or storing the clothing suggestion (3) and / or accessory suggestion generated in the previous step in a non-volatile storage medium (5).

Description

[0001] The present invention relates to a computer-implemented method for the automated generation of a clothing suggestion and / or an accessory suggestion.

[0002] US Patent 2020 / 143454 A1 discloses a system for generating fashion tips. This system takes into account photos of the person being advised and fashion rules from fashion experts. Location-specific information can also be considered. Artificial intelligence systems, such as machine learning models, are used to generate the fashion tips.

[0003] The invention aims to provide a further improved computer-implemented method for the automated generation of a clothing suggestion and / or an accessory suggestion.

[0004] The solution to this problem is provided by a computer-implemented method for the automated generation of a clothing suggestion and / or an accessory suggestion with the features of claim 1. The dependent claims relate to advantageous further developments of the invention.

[0005] A computer-implemented method according to the invention for the automated generation of a clothing suggestion and / or an accessory suggestion comprises the steps - Providing a vehicle configuration dataset for a vehicle with a plurality of vehicle configuration data, - Processing the vehicle configuration dataset using artificial intelligence, whereby design-specific features of the vehicle are extracted from the vehicle configuration dataset and a clothing suggestion and / or an accessory suggestion for a user of the vehicle is automatically generated from the design-specific features, - Displaying the clothing suggestion and / or accessory suggestion generated in the previous step using a display device and / or storing the clothing suggestion and / or accessory suggestion generated in the previous step in a non-volatile storage medium.

[0006] The method according to the invention advantageously enables the automated generation of clothing and / or accessory suggestions for a vehicle user based on design-specific features contained in the associated vehicle configuration data set. By automatically generating clothing and / or accessory suggestions using artificial intelligence, the often time-consuming selection of clothing and / or accessories suitable for the vehicle is significantly simplified for the vehicle user. Examples of accessories, which are expressly not exhaustive, include headwear, in particular hats or caps, gloves, belts, bags, umbrellas, and jewelry. The non-volatile storage medium can be integrated into the computing unit that executes the artificial intelligence.The non-volatile storage medium can also be decentralized cloud storage.

[0007] In one embodiment, the vehicle configuration data set contains design-specific vehicle features including information about the vehicle model, exterior features, and interior features. Exterior features might include, for example, the vehicle's exterior color, wheel type, and other exterior design elements. Interior features might include, for example, the interior color, seat color and material, stitching color, and other interior design elements.

[0008] In one embodiment, it is proposed that the artificial intelligence classifies the design-specific features contained in the vehicle configuration dataset and assigns them to specific attributes, and uses these attributes in the automated generation of clothing and / or accessory suggestions. For example, the artificial intelligence can automatically classify information about the vehicle type, exterior features, and interior features contained in a given vehicle configuration dataset and assign them to specific attributes, such as "sporty," "luxurious," "classic," or "futuristic," and use these attributes for the automated generation of clothing and / or accessory suggestions for clothing or accessories that match the vehicle.

[0009] In one embodiment, the computer-implemented method comprises a step of training the artificial intelligence, in which the artificial intelligence is trained by machine learning using a large amount of training data, comprising at least a plurality of vehicle configuration datasets and information about the product portfolio of at least one clothing manufacturer and / or at least one accessory manufacturer. The training is designed to generate clothing suggestions and / or accessories suitable for the vehicle from the vehicle configuration datasets and the information about the product portfolio of the at least one clothing manufacturer and / or the at least one accessory manufacturer. Preferably, information about fashion trends and color combinations from various information sources, such as fashion magazines, trend agencies, or social media platforms, can also be used for training the artificial intelligence.In principle, there is the possibility of supervised machine learning, in which additional human expert knowledge, such as the knowledge of fashion consultants, is incorporated into the training of the artificial intelligence.

[0010] In an advantageous embodiment, it is proposed that the artificial intelligence be trained using contrastive learning. This allows color and style combinations that frequently occur together to be advantageously grouped, while rarely related color and style combinations can be discarded.

[0011] Further features and advantages of the present invention will become clear from the following description of a preferred embodiment with reference to the enclosed Fig. Figure 1 shows a system 100 for carrying out a computer-implemented procedure for the automated generation of a clothing suggestion 3 and / or an accessory suggestion in a highly simplified schematic form. For the sake of simplicity, only the automated generation of the clothing suggestion 3 will be discussed below. An accessory suggestion can be generated automatically in a completely analogous manner.

[0012] System 100 comprises a computing unit 101 in which an artificial intelligence 1 is implemented. During the execution of the procedure, the artificial intelligence 1 is provided with a vehicle configuration dataset 2 of a vehicle, containing a plurality of vehicle configuration data as input variables. The vehicle configuration data includes data that, in addition to information about the vehicle model of a vehicle manufacturer, which constitutes a design-specific feature, also includes additional design-specific features of the vehicle, in particular exterior equipment features, such as the exterior color, the rim type, and other exterior design features, as well as interior equipment features, such as the interior color, the color and material of the seats, the color of decorative stitching, and other interior design features.The artificial intelligence 1 is trained to extract from the vehicle configuration data set 2 not only the information about the vehicle model but also the other design-specific features of the configured vehicle contained therein, and to automatically generate a clothing suggestion 3 for clothing items suitable for the vehicle, which may also include shoes, on this data basis.

[0013] Before going into production, the artificial intelligence 1 is trained on a large number of training datasets to automatically generate clothing suggestions 3 for a user of the vehicle associated with the respective vehicle configuration dataset 2. These clothing suggestions should be as close as possible to the vehicle model and its exterior and interior features. The artificial intelligence 1 is trained using machine learning, utilizing not only the vehicle configuration datasets 2 but also the product portfolios of numerous clothing manufacturers and, preferably, information on fashion trends and color combinations from various sources, such as fashion magazines, trend agencies, and social media platforms.In principle, there is also the possibility of supervised machine learning, in which additional human expert knowledge, such as the knowledge of fashion consultants, is incorporated into the training of the artificial intelligence 1.

[0014] Artificial intelligence 1 can be trained preferably through contrastive learning. This allows color and style combinations that occur frequently to be grouped together advantageously, while rarely related color and style combinations are discarded.

[0015] The artificial intelligence 1 can also be trained in such a way as to automatically classify the exterior and interior equipment features contained in a given vehicle configuration data set 2 and assign them to specific attributes, such as "sporty", "luxurious", "classic" or "futuristic", and use these attributes for the automated generation of a clothing suggestion 3 for suitable clothing items for the user of the vehicle.

[0016] Once training is complete, the artificial intelligence 1 can be deployed in production. In production, the trained artificial intelligence 1 receives a vehicle configuration data set 2 and automatically generates a clothing suggestion for the user of a vehicle to which the vehicle configuration data set 2 is assigned.

[0017] The clothing suggestion 3 can be visualized for the user by means of a display device 4 of the system 100 and / or stored in a non-volatile memory medium 5, 5' of the system 100. Fig. Figure 1 shows that the non-volatile storage medium 5 can, for example, be integrated into the computing device 101. The non-volatile storage medium 5' can also, for example, be implemented as decentralized cloud storage. QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] US 2020 / 143454 A1

[0002]

Claims

[1] Computer-implemented method for the automated generation of a clothing suggestion (3) and / or an accessory suggestion, comprising the steps - Providing a vehicle configuration data set (2) of a vehicle with a plurality of vehicle configuration data, - Processing the vehicle configuration data set (2) using artificial intelligence (1), wherein design-specific features of the vehicle are extracted from the vehicle configuration data set (2) and a clothing suggestion (3) and / or an accessory suggestion for a user of the vehicle is automatically generated from the design-specific features, - Displaying the clothing suggestion (3) and / or accessory suggestion generated in the previous step by means of a display device (4) and / or storing the clothing suggestion (3) and / or accessory suggestion generated in the previous step in a non-volatile storage medium (5). [2] Computer-implemented method according to claim 1, characterized by , that the design-specific features of the vehicle contained in the vehicle configuration data set (2) include information about the vehicle model, information about exterior equipment features and information about interior equipment features. [3] Computer-implemented method according to one of claims 1 or 2, characterized by, that the artificial intelligence (1) classifies the design-specific features contained in the vehicle configuration data set (2) and assigns them to specific attributes and uses these attributes in the automated generation of the clothing suggestion (3) and / or accessory suggestion. [4] Computer-implemented method according to one of claims 1 to 3, comprising a step of training the artificial intelligence (1), wherein the artificial intelligence (1) is trained by machine learning with a plurality of training data comprising at least a plurality of vehicle configuration data sets (2) and information about a product portfolio of at least one clothing manufacturer and / or at least one accessory manufacturer to generate clothing suggestions (3) and / or accessory suggestions suitable for the vehicle from the vehicle configuration data sets (2) and the information about the product portfolio of the at least one clothing manufacturer and / or the at least one accessory manufacturer. [5] Computer-implemented method according to claim 4, characterized by , that the artificial intelligence (1) is trained through contrastive learning.

Images