KR20220151488A - Network server and method to provide data associated with fashion trend - Google Patents

Abstract

A method for providing the data related to a fashion trend comprises: a step of storing the target images in a database; a step of determining the times corresponding to each of the target images using a learned artificial intelligence model to output a time corresponding to an image when the image is inputted; a step of generating the fashion attribute datasets by processing each of the target images; and a step of transmitting the information related to the fashion attribute datasets and the determined times to a client terminal through a network. Therefore, the present invention is capable of enabling a design trend to be identified effectively.

Description

Network server and method for providing data related to fashion trends {NETWORK SERVER AND METHOD TO PROVIDE DATA ASSOCIATED WITH FASHION TREND}

The present invention relates to a network system, and more particularly to a network server and method for providing data related to fashion trends.

An artificial intelligence (AI) system is a computer system that implements human-level intelligence, and unlike existing rule-based smart systems, a machine learns and judges itself and derives a desired result or performs a desired action. The more AI systems are used, the higher the recognition rate and the more accurate understanding of user preferences.

Artificial intelligence technology consists of machine learning (deep learning) and element technologies using machine learning. Machine learning is an algorithm technology that classifies/learns the characteristics of input data by itself, and consists of technical fields such as linguistic understanding, visual understanding, reasoning/prediction, knowledge expression, and motion control by utilizing machine learning algorithms such as deep learning. do.

Artificial intelligence technology is being actively researched in various fields. For example, element technologies for implementing artificial intelligence technology include linguistic understanding technology for recognizing human language/characters, visual understanding technology for recognizing objects as human eyes, and logical reasoning and prediction by judging information. It may include at least one of inference/prediction technology, knowledge expression technology that processes human experience information into knowledge data, and motion control technology that controls autonomous vehicle driving and robot movement.

The above description is only intended to help understand the background of the technical ideas of the present invention, and therefore, it cannot be understood as the prior art known to those skilled in the art.

Embodiments of the present invention are intended to provide a network server and method that enable a user to easily grasp when and where a product or design is popular by providing data related to a fashion trend.

A method for providing data related to fashion trends according to an embodiment of the present invention includes storing target images in the database; determining times corresponding to the target images by using an artificial intelligence model learned to output times corresponding to the images when an image is input; generating fashion attribute data sets by processing each of the target images; and transmitting information associated with the fashion attribute data sets and the determined times to a client terminal via a network.

The method may further include mapping each of the fashion attribute data sets to one of the determined times, and the associated information may include at least a portion of the mapped fashion attribute data sets.

The artificial intelligence model may be trained to output a position corresponding to the image in addition to the time when the image is input, and the step of determining the times may include using the artificial intelligence model in addition to the times to output the target image. It may include determining positions respectively corresponding to the .

The method may further include mapping each of the fashion attribute data sets to one of the determined times and to one of the determined locations, wherein the associated information is at least part of the mapped fashion attribute data sets. can include

Each of the fashion attribute data sets may include attribute values associated with an object in a corresponding target image.

The method includes extracting those having the same attribute value among the fashion attribute data sets; and analyzing a trend of times corresponding to the extracted fashion attribute data sets among the determined times. In this case, the related information may include information indicating the trend.

The method includes extracting those having the same attribute value among the fashion attribute data sets; and detecting at least one period between times corresponding to the extracted fashion attribute data sets among the determined times, and the associated information may include information indicating the period. .

Another aspect of the present invention relates to a network server for communicating with a client terminal over a network. The network server may include a communicator connected to the network; and a processor configured to communicate with the client terminal through the communicator, wherein the processor accesses a database storing target images, and when an image is input, an artificial intelligence model learned to output a time corresponding to the image. determining times respectively corresponding to the target images by using ; and processing the target images respectively to generate fashion attribute data sets, and transmitting information associated with the fashion attribute data sets and the determined times to the client terminal through the communicator.

The processor may be configured to map each of the fashion attribute data sets to one of the determined times, and the associated information may include at least some of the mapped fashion attribute data sets.

The artificial intelligence model may be trained to output a position corresponding to the image in addition to the time when the image is input, and the processor may correspond to the target images in addition to the times using the artificial intelligence model. It may be configured to further determine the positions of

The processor may be configured to map each of the fashion attribute data sets to one of the determined times and to one of the determined locations, and the associated information may include at least some of the mapped fashion attribute data sets. can

Each of the fashion attribute data sets may include attribute values associated with an object in a corresponding target image.

The processor extracts one of the fashion attribute data sets having the same attribute value, analyzes a trend for times corresponding to the extracted fashion attribute data sets among the determined times, and information representing the trend. may be configured to transmit to the client terminal as the associated information.

The processor extracts one of the fashion attribute data sets having the same attribute value, detects at least one cycle between times corresponding to the extracted fashion attribute data sets among the determined times, It may be configured to transmit information indicating one period to the client terminal as the associated information.

According to embodiments of the present invention, a network server and method are provided that enable users to easily identify when and where a product or design is popular by providing data related to fashion trends. For example, the network server and method determine times corresponding to images using an artificial intelligence model learned so that times corresponding to images are output when images are input, and based thereon, provide data related to fashion trends. can By referring to the data related to the fashion trend, the user can grasp the trend of the corresponding product or design more effectively than in the case of solely relying on memory.

1 is a block diagram showing a network system according to an embodiment of the present invention.
Figure 2 is a block diagram showing an embodiment of the information providing server of Figure 1;
FIG. 3 is a diagram conceptually showing data generated by the meta data generating unit of FIG. 2 .
FIG. 4 is a diagram conceptually showing data generated by the fashion attribute detection unit of FIG. 2 .
FIG. 5 is a diagram conceptually showing data generated by the fashion trend data generation unit of FIG. 2 .
6 is a block diagram showing an embodiment of the meta data generation unit of FIG. 2 .
7 is a flowchart illustrating a method of providing data related to fashion trends according to an embodiment of the present invention.
8 is a flowchart showing an embodiment of step S150 of FIG. 7 .
9 is a flowchart showing another embodiment of step S150 of FIG. 7 .
10 is a block diagram showing another embodiment of the information providing server of FIG. 1 .

Hereinafter, a preferred embodiment according to the present invention will be described in detail with reference to the accompanying drawings. It should be noted that in the following description, only parts necessary for understanding the operation according to the present invention are described, and descriptions of other parts will be omitted in order not to obscure the subject matter of the present invention. In addition, the present invention may be embodied in other forms without being limited to the embodiments described herein. However, the embodiments described herein are provided to explain in detail enough to easily implement the technical idea of the present invention to those skilled in the art to which the present invention belongs.

Throughout the specification, when a part is said to be "connected" to another part, this includes not only the case where it is "directly connected" but also the case where it is "indirectly connected" with another element interposed therebetween. . Terminology used herein is for describing specific embodiments and is not intended to limit the present invention. Throughout the specification, when a certain component is said to "include", it means that it may further include other components without excluding other components unless otherwise stated. "At least one of X, Y, and Z" and "at least one selected from the group consisting of X, Y, and Z" means one X, one Y, one Z, or two or more of X, Y, and Z It can be interpreted as any combination of more (e.g., XYZ, XYY, YZ, ZZ). Here, "and/or" includes any combination of one or more of the elements.

1 is a block diagram showing a network system according to an embodiment of the present invention.

Referring to FIG. 1 , a network system 100 may include a network 50 , a client terminal 110 , an information providing server 120 , and a database server 130 .

Network system 100 may include a plurality of devices, servers, and/or software components that operate to perform various methods according to embodiments of the invention described herein. The devices and/or servers shown in FIG. 1 may be configured in different ways, and the operations and services provided by the devices and/or servers may be combined or separated for the embodiments described herein. and may be performed by more or fewer devices and/or servers. One or more devices and/or servers may be operated and/or maintained by the same or different entities.

The network 50 connects components within the network system 100, such as the client terminal 110, the information providing server 120, and the database server 130. Network 50 may include at least one of a public network, at least one private network, a wired network, a wireless network, any other suitable type of network, and combinations thereof. Each of the components in the network system 100 may include at least one of a wired communication function and a wireless communication function, and thus may communicate with each other through the network 50 .

The client terminal 110 is configured to communicate with the information providing server 120 through the network 50 . The client terminal 110 may receive fashion trend data (FTD) and/or related data from the information providing server 120 through the network 50 and display the received data on a display device. For example, the client terminal 110 may receive at least a portion of the fashion trend data FTD illustrated in FIG. 5 and visualize the received data on a display device. As another example, the client terminal 110 may receive data analyzed based on the fashion trend data (FTD) and visualize the received data on a display device.

In embodiments, the client terminal 110 may include hardware, software, and/or firmware suitable for performing the above operations. In embodiments, the client terminal 110 may include a computer device, an ultra mobile PC (UMPC), a workstation, a net-book, a personal digital assistant (PDA), a portable computer, a web tablet, information such as wireless phones, mobile phones, smart phones, e-books, portable multimedia players (PMPs), and portable game consoles in a wired and/or wireless environment. It may include a device capable of transmitting and receiving.

The information providing server 120 may be a network server configured to communicate with the client terminal 110 and the database server 130 through the network 50 . The information providing server 120 may acquire a plurality of images by accessing at least one database server 130 . The database server 130 may include at least one of various types of network servers, for example, a portal server that provides a plurality of images through web pages.

The information providing server 120 may generate fashion attribute data sets respectively corresponding to the obtained images and metadata sets respectively corresponding to the acquired images. Each of the meta data sets may include information indicating the time and location at which the corresponding image was published (or created). The information providing server 120 may generate fashion trend data FTD from fashion attribute data sets and meta data sets, and store the generated fashion trend data FTD in its database DB. Thereafter, the information providing server 120 may provide at least a portion of the fashion trend data FTD from the database DB to the client terminal 110 . Also, the information providing server 120 may analyze the fashion trend data (FTD) according to a request from the client terminal 110 and provide the analyzed data to the client terminal 110 .

Figure 2 is a block diagram showing an embodiment of the information providing server of Figure 1; FIG. 3 is a diagram conceptually showing data generated by the meta data generating unit of FIG. 2 . FIG. 4 is a diagram conceptually showing data generated by the fashion attribute detection unit of FIG. 2 . FIG. 5 is a diagram conceptually showing data generated by the fashion trend data generation unit of FIG. 2 .

Referring to FIG. 2 , the information providing server 200 may include a trend information providing device 201 and first and second databases DB1 and DB2. In Figure 2, the information providing server 200 is shown as including two databases (DB1, DB2). However, embodiments of the present invention are not limited thereto. For example, the first and second databases DB1 and DB2 may be disposed outside of the information providing server 200, and the information providing server 200 provides the first and second databases through the network 50. Data can be read or written by accessing fields (DB1, DB2). The first and second databases DB1 and DB2 may be included in the database DB of FIG. 1 .

The trend information providing device 201 collects target images TIMD and stores them in a first database DB1, while generating fashion trend data FTD based on the target images TIMD and generating fashion trend data. It is configured to store the data FTD in the second database DB2. The trend information providing device 201 may include a database server interface 210, a client terminal interface 220, and a trend information generator 230.

Database server interface 210 may interact with database server 130 (see FIG. 1) over network 50 (see FIG. 1). The database server interface 210 may provide elements of the trend information generator 230 with an interface to the database server 130 . Client terminal interface 220 can interact with client terminal 110 over network 50 . The client terminal interface 220 may provide an interface for the client terminal 110 to components of the trend information generator 230 .

The trend information generator 230 may include an image collection unit 231, a meta data generator 232, a fashion attribute detector 233, and a fashion trend data generator 234. The image collection unit 231 accesses the database server 130 through the database server interface 210, collects target images TIMG, and stores the collected target images TIMG in the first database DB1. is configured to

The meta data generation unit 232 includes an artificial intelligence model trained to output a meta data set corresponding to the image when an image is input, and meta data respectively corresponding to the target images TIMG using the artificial intelligence model. configured to create sets. Referring to FIG. 3 , the artificial intelligence model (AI) included in the meta data generator 232 generates first to m th metadata sets MDS1 to MDSm from the first to m th target images TIMG1 to TIMGm. ) can be created respectively. In this case, each of the first to m th meta data sets MDS1 to MDSm may include a time DT at which the corresponding image was issued. Each of the first to m th meta data sets MDS1 to MDSm may further include a location LC where the corresponding image was issued (or related to). For example, each of the first to m th meta data sets MDS1 to MDSm may include information indicating the year, year, month, or date when the corresponding image was issued, and a region (eg, country).

Referring back to FIG. 2 , the fashion attribute detector 233 is configured to generate fashion attribute data sets by processing each of the target images TIMG. Referring to FIG. 4 , the fashion attribute detector 233 may generate first to m th fashion attribute data sets FP1 to FPm from the first to m th target images TIMG1 to TIMGm, respectively. The fashion attribute detection unit 233 detects objects such as clothes (eg coats, dresses, etc.) and sundries (eg shoes, hair bands) included in the target image, and creates a fashion attribute data set from the detected objects. can do. Each of the first to mth fashion attribute data sets FP1 to FPm may include various types of attribute values TP1 , TP2 , ... , TPn associated with the detected object. For example, the first attribute value TP1 represents a category of an object such as “one piece” or “coat”, and the second attribute value TP2 represents a category of an object such as “light pink”, “red”, or “beige”. It represents a color, and the third attribute value TP3 may represent the material of an object such as “cotton” or “wool”. Each fashion attribute data set may further include additional design attributes related to object categories. For example, the fourth attribute value TP4 may indicate "long sleeves" when the first attribute value TP1 indicates "one piece", and may indicate "duffel" when the first attribute value TP1 indicates "coat". can refer to "court".

In embodiments, the fashion attribute detector 233 may include an artificial intelligence model learned to output a fashion attribute data set corresponding to the image when an image is input, and the fashion attribute detector 233 uses such an artificial intelligence model. Thus, a fashion attribute data set corresponding to each target image may be generated.

Referring back to FIG. 2 , the fashion trend data generation unit 234 is configured to generate fashion trend data FTD by mapping each fashion attribute data set to a time and/or location defined by a corresponding meta data set. The fashion trend data FTD may be stored in the database DB2.

Referring to FIG. 5 , each of the fashion attribute data sets FP is mapped to a specific coordinate in a vector space having time DT as the horizontal axis and position LC as the vertical axis. The mapped coordinates are determined by the time and location of the corresponding metadata set. The fashion trend data FTD of FIG. 2 is provided as fashion attribute data sets mapped in a vector space as described above, and accordingly, the fashion trend data FTD are fashion attribute data sets associated with each other (eg, having the same attribute values). fashion attribute data sets) may include information on trends. For example, fashion attribute data sets in which the category is “coat” and the design is “duffle coat” among fashion attribute data sets in which the location (LC) is Korean may appear relatively frequently in 2006 and 2016. For example, among fashion attribute data sets in which the location (LC) is American, fashion attribute data sets in which the category is “one piece” and the design is “long sleeve” appear relatively frequently in 1987, 2002, and 2017. have.

Referring back to FIG. 2 , the fashion trend data generating unit 234 may provide at least a portion of the fashion trend data FTD to the client terminal 110 through the client terminal interface 220 . As shown in FIG. 5 , the fashion trend data generation unit 234 provides at least some of the fashion attribute data sets mapped to the vector space to the client terminal 110, and the client terminal 110 provides the fashion in the vector space. Attribute data sets can be visualized through a display device. For example, fashion attribute data sets mapped to a vector space, with a category of “coat” and a design of “duffle coat,” may be provided to the client terminal 110 . In embodiments, the fashion trend data generation unit 234 may provide fashion attribute data sets having attribute values selected by the user of the client terminal 110 to the client terminal 110 . By generating the fashion trend data (FTD) and providing it to the client terminal 110 in this way, the user of the client terminal 110 can easily grasp when and where a specific product or design was popular, and can track the trend of the product or design. Therefore, it is possible to effectively predict when the product or design will become popular in the future.

In embodiments, data associated with the fashion trend data (FTD) may be provided to the client terminal 110 . The fashion trend data generation unit 234 extracts fashion attribute data sets having attribute values selected by the user, analyzes a trend of the extracted fashion attribute data sets in a vector space, and transmits data representing the analyzed trend to a client terminal. (110) can be provided. For example, the fashion trend data generation unit 234 may include an artificial intelligence model trained for trend analysis, and may analyze trends using the artificial intelligence model. In embodiments, the fashion trend data generation unit 234 detects at least one cycle between times DT corresponding to the extracted fashion attribute data sets, and transmits data representing the detected cycle to the client terminal 110. can be provided to For example, as shown in FIG. 5, if fashion attribute data sets having the category “one piece” and design “long sleeves” in the vector space appear relatively frequently in 1987, 2002, and 2017 in the United States, the corresponding fashion attribute The data sets may have a trend that appears in a cycle of about 15 years in the United States, and the fashion trend data generator 234 transmits data indicating that a corresponding product or design appears in a cycle of about 15 years in the United States to the client terminal 110. can In addition, the fashion trend data generation unit 234 may further transmit data indicating that a corresponding product or design may be popular in 2032 or 2047 corresponding to a 15-year cycle from 2017 to the client terminal 110. . In this way, the fashion trend data generation unit 234 provides the client terminal 110 with information indicating when and where the product or design has become popular, and can further provide information indicating when the product or design will become popular in the future. .

In embodiments, each of the image collection unit 231, the meta data generation unit 232, the fashion attribute detection unit 233, and the fashion trend data generation unit 234 may include at least one of hardware, software, firmware, and a combination thereof. can be implemented through In embodiments, at least some of the image collection unit 231, the meta data generation unit 232, the fashion attribute detection unit 233, and the fashion trend data generation unit 234 may be integrated, and separated into more components It can be.

6 is a block diagram showing an embodiment of the meta data generation unit of FIG. 2 .

Referring to FIG. 6 , the meta data generator 300 may include an artificial intelligence model 310 and an artificial intelligence processor 320.

The artificial intelligence model 310 may be trained in advance to output the time DT and location LC when an image IMG such as the target image TIMG of FIG. 2 is input. In embodiments, the artificial intelligence model 310 may include one or more neural networks (L1, L2, ... , L_k-1, L_k), and when the image IMG is input, the time ( DT) and position (LC). For example, the neural networks (L1, L2, ... , L_k-1, L_k) are neural networks corresponding to an encoder for outputting a vector representing feature information from an input image (IMG), and feature information It may include neural networks corresponding to a decoder for converting a vector representing Δ into time (DT) and position (LC).

The artificial intelligence processor 320 is configured to control the artificial intelligence model 310 . The artificial intelligence processor 320 may include a data learning unit 321 and a data processing unit 322 . The data learning unit 321 uses learning data including an image and its corresponding time and location so that when the image IMG is input to the artificial intelligence model 310, the time DT and location LC are output. The artificial intelligence model 310 may be trained. Data for such learning may also be acquired from at least one database server 130 on the network 50 (see FIG. 1).

The data processing unit 322 may acquire the time DT and location LC as result values by inputting the target image TIMG of FIG. 2 as an image IMG to the learned artificial intelligence model 310 . The obtained time DT and location LC may form a metadata set corresponding to the target image TIMG as described with reference to FIG. 3 .

In embodiments, the artificial intelligence model 310 and the artificial intelligence processor 320 may be implemented as a processor and a memory. A processor may include one or more cores, such as single core, dual core, quad core, etc. The processor may provide each of the artificial intelligence model 310 and the artificial intelligence processor 320 by loading programs and/or instructions into memory and executing the loaded programs and/or instructions.

7 is a flowchart illustrating a method of providing data related to fashion trends according to an embodiment of the present invention.

Referring to FIG. 7 , in step S110, the information providing server 120 (see FIG. 1) accesses the database server 130 and collects target images TIMD.

In step S120, the information providing server 120 determines the time and location corresponding to each of the target images by using an artificial intelligence model learned to output a meta data set corresponding to the image when an image is input. For example, the time may indicate at least one of the year, season, month, and date when the corresponding target image was issued, and the location may indicate a specific region such as a country or city.

In step S130, the information providing server 120 generates fashion attribute data sets by processing the target images. A fashion attribute data set corresponding to each target image may include attribute values of various types associated with an object (eg, coat, dress, etc.) in the corresponding target image. The information providing server 120 may detect objects such as clothes and sundries included in the target image, and generate a fashion attribute data set from the detected objects. Attribute values may be determined by processing each target image using at least one of a variety of algorithms known in the art. For example, the fashion attribute data set includes attribute values representing object categories such as “dress” and “coat”, attribute values representing object colors such as “light pink”, “red” and “beige”, and “face”. , "wool", etc. may include attribute values indicating the material of the object, and attribute values indicating additional design attributes related to the category of the object.

According to embodiments, step S130 may be performed before step S120, performed later than step S120, or performed in parallel with step S120.

In step S140, the information providing server 120 maps each of the fashion attribute data sets to the time and location of the corresponding target image. Each of the fashion attribute data sets is mapped in a vector space determined by time and location, and the fashion attribute data sets mapped in the vector space can be stored in a database (DB, see FIG. 1) as fashion trend data (FTD). .

In step S150, the information providing server 120 transmits information to the client terminal 110 (see FIG. 1) based on the fashion attribute data sets mapped to time and location. In embodiments, the information providing server 120 may transmit at least some of the fashion attribute data sets mapped to time and location to the client terminal 110 .

8 is a flowchart showing an embodiment of step S150 of FIG. 7 .

Referring to FIG. 8 , in step S210, the information providing server (130, see FIG. 1) extracts fashion attribute data sets having the same attribute value among fashion attribute data sets mapped to time and location. In embodiments, a user may select a specific attribute value such as “duffel coat”, and the information providing server 130 may extract fashion attribute data sets having attribute values selected by the user.

In step S220, the information providing server 130 analyzes the trend of the extracted fashion attribute data sets. The information providing server 130 may analyze trends of times and locations corresponding to the extracted fashion attribute data sets.

In step S230, the information providing server 130 may transmit information indicating the analyzed trend to the client terminal 110 (see FIG. 1). Furthermore, the information providing server 130 may estimate a time and location where a fashion attribute data set having the same attribute value appears based on the analyzed trend, and may also transmit the estimated information to the client terminal 110 .

9 is a flowchart showing another embodiment of step S150 of FIG. 7 .

Referring to FIG. 9 , in step S310, the information providing server (130, see FIG. 1) extracts items having selected attribute values from fashion attribute data sets mapped to time and location.

In step S320, the information providing server 130 determines whether at least one cycle is detected between times corresponding to the extracted fashion attribute data sets. If so, step S330 is performed. If not, step S340 is performed.

In step S330, the information providing server 130 may transmit information indicating the detected period to the client terminal 110 (see FIG. 1). In step S340, the information providing server 130 may transmit a message indicating that there is no trend associated with the selected attribute value to the client terminal 110.

10 is a block diagram showing another embodiment of the information providing server of FIG. 1 .

Referring to FIG. 10, the information providing server 1000 includes a bus 1100, at least one processor 1200, a system memory 1300, a storage interface (I/F) 1400, a communication interface 1500, a storage medium 1600, and communicator 1700.

The bus 1100 is connected to various components of the information providing server 1000 and transfers data, signals, and information. The processor 1200 may be either a general-purpose or dedicated processor, and may control overall operations of the information providing server 1000.

The processor 1200 is configured to load program codes and instructions providing various functions into the system memory 1300 when executed, and to process the loaded program codes and instructions. The system memory 1300 may be provided as a working memory of the processor 1200 . As an embodiment, the system memory 1300 may include at least one of a random access memory (RAM), a read only memory (ROM), and other types of computer-readable media.

The processor 1200 may load the trend information generation module 1310 , which provides functions of the trend information providing device 201 of FIG. 2 , into the system memory 1300 when executed by the processor 1200 . Such program codes and/or instructions may be executed by the processor 1200 to perform functions and/or operations of the trend information providing device 201 described with reference to FIG. 2 . In order to perform such functions and/or operations, the trend information generation module 1310 executed by the processor 1200 is a component of the information providing server 1000, such as the storage interface 1400 and the communication interface 1500. can use them.

Program codes and/or instructions may be loaded into the system memory 1300 from a storage medium 1600, which is a recording medium readable by a separate computer. Alternatively, program codes and/or instructions may be loaded into the system memory 1300 from the outside of the information providing server 1000 through the communicator 1700. In addition, the system memory 1300 may function as a buffer memory for the content design recommendation module 1310 .

In FIG. 10 , the system memory 1300 is shown as a separate configuration from the processor 1200 , but at least a portion of the system memory 1300 may be included in the processor 1200 . The system memory 1300 may be provided as a plurality of memories physically and/or logically separated from each other according to embodiments.

The storage interface 1400 is connected to the storage medium 1600 . The storage interface 1400 may interface between components such as the processor 1200 and the system memory 1300 connected to the bus 1100 and the storage medium 1600 . The communication interface 1500 is coupled to the communicator 1700 . The communication interface 1500 may interface between components connected to the bus 1100 and the communicator 1700 .

In embodiments, the bus 1100, processor 1200, and system memory 1300 may be integrated into one board 1050. For example, the bus 1100, the processor 1200, and the system memory 1300 may be mounted on a single semiconductor chip. In embodiments, the board 1050 may further include a storage interface 1400 and a communication interface 1500 .

The storage medium 1600 may include various types of nonvolatile storage media, such as a flash memory and a hard disk, which retain stored data even when power is cut off. The storage medium 1600 may be provided as at least a part of the first and second databases DB1 and DB2 of FIG. 2 .

The communicator 1700 (or transceiver) may transmit/receive signals between the information providing server 1000 and other devices and/or servers in the network system 100 (see FIG. 1) through the network 105.

Although specific embodiments and application examples have been described herein, they are only provided to help a more general understanding of the present invention, the present invention is not limited to the above embodiments, and those of ordinary skill in the field to which the present invention belongs Numerous modifications and variations are possible from this description when grown.

Therefore, the spirit of the present invention should not be limited to the described embodiments, and it will be said that not only the claims to be described later, but also all modifications equivalent or equivalent to these claims belong to the scope of the present invention. .

110: client terminal
120, 200: information providing server
130: database server
201: trend information providing device
210: image server interface
220: client server interface
230: trend information generator

Claims (14)

A method for providing data associated with fashion trends:
storing target images in the database;
determining times corresponding to the target images by using an artificial intelligence model learned to output times corresponding to the images when an image is input;
generating fashion attribute data sets by processing each of the target images; and
and transmitting information associated with the fashion attribute data sets and the determined times to a client terminal via a network. According to claim 1,
Further comprising mapping each of the fashion attribute data sets to one of the determined times,
wherein the associated information includes at least some of the mapped fashion attribute data sets. According to claim 1,
The artificial intelligence model learns to output a position corresponding to the image in addition to the time when the image is input,
The step of determining the times includes determining positions respectively corresponding to the target images in addition to the times using the artificial intelligence model. According to claim 3,
further comprising mapping each of the fashion attribute data sets to one of the determined times and one of the determined locations;
wherein the associated information includes at least some of the mapped fashion attribute data sets. According to claim 1,
Each of the fashion attribute data sets includes attribute values associated with an object in a corresponding target image. According to claim 5,
extracting those having the same attribute value from among the fashion attribute data sets; and
Further comprising analyzing trends for times corresponding to the extracted fashion attribute data sets among the determined times,
The method of claim 1, wherein the associated information includes information indicating the trend. According to claim 5,
extracting those having the same attribute value from among the fashion attribute data sets; and
Further comprising detecting at least one cycle between times corresponding to the extracted fashion attribute data sets among the determined times,
The method of claim 1, wherein the associated information includes information indicating the period. In a network server for communicating with a client terminal through a network:
a communicator connected to the network; and
A processor configured to communicate with the client terminal through the communicator;
the processor,
accessing a database storing target images;
determining times corresponding to the target images, respectively, by using an artificial intelligence model learned so that times corresponding to the images are output when an image is input;
processing each of the target images to generate fashion attribute data sets;
A network server configured to transmit information associated with the fashion attribute data sets and the determined times to the client terminal via the communicator. According to claim 8,
the processor is configured to map each of the fashion attribute data sets to one of the determined times;
The network server of claim 1 , wherein the associated information includes at least some of the mapped fashion attribute data sets. According to claim 8,
The artificial intelligence model learns to output a position corresponding to the image in addition to the time when the image is input,
wherein the processor is configured to further determine locations respectively corresponding to the target images in addition to the times using the artificial intelligence model. According to claim 10,
the processor is configured to map each of the fashion attribute data sets to one of the determined times and one of the determined locations;
The network server of claim 1 , wherein the associated information includes at least some of the mapped fashion attribute data sets. According to claim 8,
Each of the fashion attribute data sets includes attribute values associated with an object in a corresponding target image. According to claim 12,
the processor,
Extracting those having the same attribute value among the fashion attribute data sets;
Analyzing a trend for times corresponding to the extracted fashion attribute data sets among the determined times;
A network server configured to transmit the information indicating the trend to the client terminal as the associated information. According to claim 12,
the processor,
Extracting those having the same attribute value among the fashion attribute data sets;
detecting at least one cycle between times corresponding to the extracted fashion attribute data sets among the determined times;
The network server configured to transmit information indicating the at least one period to the client terminal as the associated information.

Images