KR102119253B1 - System, method and program for obtaining abstract data from image data - Google Patents

Abstract

The present invention relates to a method for obtaining abstract characteristics of image data, an apparatus thereof, and a program thereof. According to one embodiment of the present invention, the method for obtaining abstract characteristics of image data comprises the steps of: inputting a plurality of individual appearance characteristics calculated for the image data into an abstract characteristic recognition model and calculating abstract characteristics by a server (S4100); and combining the calculated individual appearance characteristics and the abstract characteristics to generate appearance descriptive data by the server (S4200).

Description

Method, device and program for obtaining abstract characteristics of image data {SYSTEM, METHOD AND PROGRAM FOR OBTAINING ABSTRACT DATA FROM IMAGE DATA}

The present invention relates to a method, apparatus and program for obtaining abstract characteristics of image data.

With the recent development of the Internet, social media network services are rapidly growing. As a result, as the amount of multimedia explosively increases, an effective image retrieval system is required, and image annotation is becoming increasingly important due to the necessity of efficient image retrieval according to explosively increasing web images.

In most image search studies, a content-based image retrieval (CBIR) method, which mainly analyzes the contents of an image, has been conducted. Content-based image retrieval analyzes the contents of an image using visual features such as color, texture, and shape. This method works well when the number of defined tags is small, but performance decreases as the dataset becomes larger and the types of tags vary.

Text-based image retrieval (TBIR) is a method of retrieving an image corresponding to text using text as a query. In this method, the visual content of an image is represented by a manually tagged text descriptor, and is used to perform image search in a dataset management system. That is, in the existing image or video search method, a search is performed based on information directly tagged by a user. In addition, the method of acquiring the user interest information through the image is obtained based on the information directly tagged by the user, and thus, if the user incorrectly tags the keyword in the image, the acquisition result is incorrect. In addition, since there may be a difference in keywords defined for each user, there is a problem in that results provided according to keywords selected by a user inputting an image are different.

Published Patent Publication No. 10-2018-0133200, 2018.12.13

The present invention is a method for acquiring abstract characteristics of image data, obtaining abstract characteristics, which are emotional characteristics felt for a specific object, based on appearance description data in which a plurality of individual appearance characteristics calculated from image data are combined in a code string form, It is intended to provide devices and programs.

The problems to be solved by the present invention are not limited to the problems mentioned above, and other problems not mentioned will be clearly understood by those skilled in the art from the following description.

A method for acquiring abstract characteristics of image data according to an embodiment of the present invention includes: a server inputting a plurality of individual appearance characteristics calculated for image data into an abstract characteristic recognition model, thereby calculating abstract characteristics; And generating, by the server, appearance descriptive data by combining the calculated plurality of individual appearance characteristics and abstract characteristics, wherein the individual appearance characteristics include a specific classification criterion, which is a specific classification criterion for describing the appearance of an object. It expresses various external appearance characteristics, and the abstract characteristic includes a first emotional characteristic recognized based on the external appearance of the object.

In addition, in another embodiment, the individual external characteristics are calculated by inputting the image data into each of a plurality of individual characteristic recognition modules for determining different external classification criteria in the external characteristic recognition model, and the individual characteristics The recognition module calculates individual appearance characteristics included in specific appearance classification criteria of image data.

In addition, in another embodiment, the appearance classification criteria include a specialized appearance classification criterion applied only to a specific type of object and a universal appearance classification criterion applied to all types of objects.

In another embodiment, the abstract characteristic recognition model includes a first emotional characteristic recognition model, and the first emotional characteristic recognition model scores for each of a plurality of first individual emotional characteristics for each individual external characteristic. As is set, the first emotional characteristic is calculated by summing the scores for each of the first individual emotional characteristics set in the inputted plurality of individual external characteristics.

In addition, in another embodiment, the external description data generation step extracts code values corresponding to each of a plurality of individual appearance characteristics and first emotion characteristics of the image data, and the plurality of code values are combined. It is to generate the column-shaped appearance description data, and the code value corresponding to the first emotional characteristic includes information on the scores summed for each of the first individual emotional characteristics.

In another embodiment, the abstract characteristic further includes a second emotional characteristic recognized based on information given to the product of the object, and the abstract characteristic recognition model receives product information of the object And a second emotional characteristic recognition model for calculating a second emotional characteristic, wherein the second emotional characteristic recognition model includes a plurality of second individual emotional characteristic recognition modules for determining a second emotional characteristic for different product information. will be.

In addition, in another embodiment, the server calculates the recommended outline description data matching the outline description data, and generates recommended image data information that is information about one or more image data including the recommended outline description data. ; Further includes.

In another embodiment, in the step of generating the recommended image data information, the server calculates the recommended outline description data based on the degree of association between the outline description data and the first individual emotional characteristics, thereby generating recommended image data information. Characterized in that, and the degree of association between the first individual emotional characteristics, it includes a similarity and dissimilarity between each of the first individual emotional characteristics.

Further, in another embodiment, in the generating of the recommended image data information, the server calculates the recommended appearance description data based on the correlation between the appearance description data, the first individual emotional characteristics, and user preference information, and recommends It is characterized in that it generates image data information, and the user preference information is preference information for each of the first individual emotion characteristics of the user.

Further, in another embodiment, the server matching one or more first individual sentiment characteristics to the keyword; And extracting, by the server, a first individual sentiment characteristic matching the search keyword received from the user client, and transmitting image data including the first individual sentiment characteristic to the user client.

Further, according to another embodiment, the server inputting the calculated first emotional characteristic into a style (style) recognition model, and calculating style information; further comprising, the style recognition model, the first emotion As each region of a plurality of styles is set in the characteristic spatial map, style information of an area in which the input first emotional characteristic is located is calculated, and the first emotional characteristic spatial map is configured to respectively determine a plurality of first individual emotional characteristics. It is a spatial map arranged on a plane based on the correlation between the first individual emotional characteristics of.

An apparatus for acquiring abstract characteristics of image data according to another embodiment of the present invention includes one or more computers, and performs the above-mentioned abstract characteristic acquisition method.

The program for acquiring abstract characteristics of image data according to another embodiment of the present invention is implemented in combination with hardware to execute the above-mentioned abstract characteristic acquisition method, and is stored in a medium.

Other specific matters of the present invention are included in the detailed description and drawings.

According to the present invention, it is possible to efficiently process data by calculating abstract characteristics by using appearance description data calculated from the image data, not the image data itself.

In addition, according to the present invention, abstract characteristics can be calculated by reflecting each characteristic by subdividing the specific calculation method of abstract characteristics of image data by country, region, or individual.

In addition, according to the present invention, by calculating not only the first emotional characteristics of the object's appearance, but also the second emotional characteristics of the object's product information such as price, and receiving preferences for each individual emotional characteristic for each user , Personalized recommendation image data or search results can be provided.

The effects of the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the following description.

1 is a flowchart of a method for obtaining abstract characteristics of image data according to an embodiment of the present invention.
2 is a block diagram of an appearance characteristic recognition model according to an embodiment of the present invention.
3 is a block diagram of an abstract feature recognition model according to an embodiment of the present invention.
4 is an exemplary diagram for explaining setting of a first individual emotion characteristic score for individual appearance characteristics according to an embodiment of the present invention.
5 is a flowchart of an abstract characteristic acquisition method further comprising the step of generating recommended image data information according to an embodiment of the present invention.
6 is a flowchart of a method for acquiring abstract characteristics, which further includes a first individual emotion characteristic matching step and a user search step of keywords according to an embodiment of the present invention.
7 is a flowchart of a method for acquiring abstract characteristics, further comprising a step of calculating style information according to an embodiment of the present invention.
8 is an exemplary view for explaining a first emotional characteristic spatial map according to an embodiment of the present invention.
9 is a block diagram of an abstract characteristic acquisition server according to an embodiment of the present invention.

Advantages and features of the present invention, and methods for achieving them will be clarified with reference to embodiments described below in detail together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but may be implemented in various different forms, and only the present embodiments allow the disclosure of the present invention to be complete, and are common in the technical field to which the present invention pertains. It is provided to fully inform the skilled person of the scope of the present invention, and the present invention is only defined by the scope of the claims.

The terminology used herein is for describing the embodiments and is not intended to limit the present invention. In the present specification, the singular form also includes the plural form unless otherwise specified in the phrase. As used herein, “comprises” and/or “comprising” does not exclude the presence or addition of one or more other components other than the components mentioned. Throughout the specification, the same reference numerals refer to the same components, and “and/or” includes each and every combination of one or more of the components mentioned. Although "first", "second", etc. are used to describe various components, it goes without saying that these components are not limited by these terms. These terms are only used to distinguish one component from another component. Therefore, it goes without saying that the first component mentioned below may be the second component within the technical spirit of the present invention.

Unless otherwise defined, all terms (including technical and scientific terms) used in the present specification may be used as meanings commonly understood by those skilled in the art to which the present invention pertains. In addition, terms defined in the commonly used dictionary are not ideally or excessively interpreted unless explicitly defined.

In the present specification,'computer' includes all of various devices capable of performing arithmetic processing and providing a result to a user. For example, the computer is not only a desktop PC and a notebook, but also a smart phone, a tablet PC, a cellular phone, a personal communication service phone (PCS phone), and synchronous/asynchronous. The mobile terminal of the International Mobile Telecommunication-2000 (IMT-2000), a Palm Personal Computer (PC), and a Personal Digital Assistant (PDA) may also be applicable. In addition, when a head mounted display (HMD) device includes a computing function, the HMD device may be a computer. Also, the computer may be a server 10 that receives a request from a client and performs information processing.

In this specification,'client' refers to all devices including communication functions that users can install and use programs (or applications). That is, the client device may include one or more of a telecommunication device, a remote controller, such as a smart phone, a tablet, a PDA, a laptop, a smart watch, and a smart camera, but is not limited thereto.

In the present specification, the term'object' means an article of a specific classification or category included in image data.

In this specification,'image data' means a two-dimensional or three-dimensional static or dynamic image including a specific object. That is, the'image data' may be static image data that is one frame, or may be dynamic image data (ie, video data) in which a plurality of frames are consecutive.

In the present specification, the'appearance classification criteria' refers to a classification criterion of an appearance expression required for description or annotation of a specific object. That is, the'appearance classification criteria' is a specific classification criterion for describing the appearance of a specific object, and includes a plurality of individual appearance characteristics expressing various appearance characteristics within the same classification criterion of the object. For example, when the object is clothing, the appearance classification criteria are classification criteria for the appearance of clothing, and may include patterns, colors, fits, and lengths. In other words, if the classification criteria for a specific object are increased, the appearance of a specific object belonging to the object can be described in detail.

In the present specification,'individual appearance characteristics' means various characteristics included in a specific appearance classification criterion. For example, when the external classification criteria are colors, individual appearance characteristics refer to various individual colors.

In the present specification,'abstract characteristic' is an abstract characteristic perceived for a specific object, and includes a first emotional characteristic or a second emotional characteristic.

In the present specification,'first emotional characteristic' refers to an emotional characteristic recognized based on the appearance of a specific object. For example, it may be an emotional or fashionable expression such as'cute' or'vintage' for the appearance of a specific object.

In the present specification,'second emotional characteristics' refers to emotional characteristics recognized based on information given to a product of a specific object. For example, it may be an emotional expression such as'cheap' or'expensive' perceived for a price among product information of a specific object.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a flowchart of a method for obtaining abstract characteristics of image data according to an embodiment of the present invention.

Referring to FIG. 1, in the method of acquiring abstract characteristics of image data according to an embodiment of the present invention, a server inputs a plurality of individual appearance characteristics calculated for image data into an abstract characteristic recognition model to calculate abstract characteristics Step S4100; And generating, by the server, appearance descriptive data by combining the calculated plurality of individual appearance characteristics and abstract characteristics (S4200). Hereinafter, detailed description of each step will be described.

The server 10 inputs a plurality of individual appearance characteristics calculated on the image data into the abstract characteristic recognition model 300 to calculate the abstract characteristics (S4100). The individual external characteristics may be to express various external characteristics within the external classification criteria, which are specific classification criteria for describing the external shape of the subject.

In one embodiment, the individual appearance characteristics may be calculated by inputting image data into the appearance characteristic recognition model 200 by the server.

2 is a block diagram of an appearance characteristic recognition model according to an embodiment of the present invention.

Referring to FIG. 2, in one embodiment, the appearance characteristic recognition model 200 includes a plurality of individual characteristic recognition modules 210 for determining different appearance classification criteria. That is, the appearance characteristic recognition model includes a plurality of individual characteristic recognition modules specialized for recognizing respective appearance classification criteria. The more the classification criteria of a specific object, the more the server includes a plurality of individual characteristic recognition modules in the appearance characteristic recognition model. The individual characteristic recognition module calculates individual external characteristics included in a specific external classification criteria of image data.

In addition, in one embodiment, the calculating of the individual external characteristics (S4200) may include calculating the plurality of individual external characteristics by inputting the image data in a specialized external characteristic recognition model corresponding to object type information.

To this end, in one embodiment, prior to calculating individual appearance characteristics for specific image data, obtaining type information of the object may further include. Acquiring the type information of the object includes, but is not limited to, acquiring image data by inputting it to the object type recognition model 100.

In one embodiment, the specialized external characteristic recognition model includes a plurality of external characteristic recognition modules preset in advance to be applied according to specific object type information. That is, the type of the appearance classification criterion applied according to object type information calculated for specific image data may be determined.

Specifically, referring to FIG. 2, specialized appearance characteristic recognition in which a combination of different appearance classification criteria (that is, combination of individual characteristic recognition modules) is set according to specific object type information (object 1, object 2, object 3) Models can be generated, and input the image data to each of a plurality of individual characteristic recognition modules in a specialized appearance characteristic recognition model of the corresponding type information according to object type information calculated for specific image data to calculate individual appearance characteristics can do. Also, it goes without saying that the combination of individual characteristic recognition modules in the specialized external characteristic recognition model of different object type information may be the same.

In addition, in one embodiment, as shown in FIG. 2, the external classification criteria include specialized external classification criteria applied only to a specific type of object and universal external classification criteria applied to all types of objects, and include a plurality of object type information. Each specialized appearance characteristic recognition model for Korea can be used by sharing a general purpose individual characteristic recognition module.

For example, the external classification criteria such as'color','pattern', and'texture' can be applied regardless of the object type (individual appearance characteristics are calculated). May correspond to

The plurality of individual appearance characteristics calculated by the above-described method are input to the abstract characteristic recognition model 300 to calculate the abstract characteristics.

In one embodiment, the server may input a plurality of individual external characteristics as well as object type information to the abstract characteristic recognition model, and input the individual external characteristics or object type information in the form of external descriptive data combined. Can be.

That is, according to the present invention, as the abstract characteristics are calculated based on the external description data calculated from the image data, not the image data itself such as an image, efficient processing of data and objective abstract characteristic calculation are possible.

In one embodiment, the abstract characteristic includes a first emotional characteristic. The first emotional characteristic is a perceived emotional characteristic based on the appearance of a specific object, and includes a plurality of first individual emotional characteristics that are specific emotional characteristics.

In one embodiment, the definition and number of each first individual emotional characteristic included in the first emotional characteristic may be set by the server, and may be added or changed. The first emotional characteristic, which is an emotional characteristic for the appearance of the object, may be defined differently according to the time zone or region, so that it can be variously changed accordingly.

For example, each first individual emotion characteristic may be set to'Cute','soft','modern', etc., and also to each first individual emotion characteristic. It may be set to further include'Elegant','Wild','Classic', etc., which are contrasting first individual emotional characteristics. However, the first individual emotional characteristic of the present invention is not limited to this and can be variously set.

Referring to FIG. 3, in one embodiment, the abstract characteristic recognition model 300 includes a first emotional characteristic recognition model 310 that receives an individual external characteristic and calculates a first emotional characteristic.

In one embodiment, the first emotional characteristic recognition model, each of the first individual emotions set in the plurality of individual appearance characteristics, as the scores for each of the plurality of first individual emotion characteristics are set for each individual appearance characteristic The first emotional characteristic may be calculated by summing the scores for the characteristic.

For example, as shown in FIG. 4, for each of the plurality of individual appearance characteristics included in each appearance classification criterion, a score for each of the first individual emotion characteristics may be set. In FIG. 4, each score is indicated as 0 or 1, but the present invention is not limited thereto, and the score may be set in various ways such as a number between 0 and 1 or a negative number.

In one embodiment, a score table in which a score for each individual appearance characteristic is set for each individual appearance characteristic, as well as one score table, a plurality of different score tables may be generated. For example, the score table may be different for each country or region, or a personalized score table for each user may be generated, and the score table can be freely changed by a server.

Specifically, since the type or degree of the first individual emotion characteristic perceived by each user for the same individual appearance characteristic may be different, when data for each user is accumulated, the score value, weight, etc. of each user's score table are adjusted. By updating by, it is possible to calculate the first individual emotion characteristics optimized for each individual.

Subsequently, the first emotion characteristics are calculated by summing each first individual emotion characteristic score based on the inputted plurality of individual external characteristics and respective first individual emotional characteristic scores set in the individual external characteristics.

For example, in the example of FIG. 4, when the individual appearance characteristics of'V-neck' and'red' calculated from specific image data are input, the sum of the scores of each first individual emotional characteristic is'cute: 1'. ,'Elegant: 2','Soft: 0','Rough: 1','Modern: 0','Classic: 1', and based on this, the first emotional characteristics can be calculated.

In one embodiment, the first emotional characteristic may be calculated by including a ratio of the total score of each first individual emotional characteristic score. For example, in the example described above, since the overall score is 5, the first emotional characteristic includes a ratio of each first individual emotional characteristic score, such as "cute: 0.2, elegant: 0.4, soft: 0, rough: 0.2, modern It can be calculated as Red: 0, Classic: 0.2".

In another embodiment, the first emotional characteristic may be calculated by including each first individual emotional characteristic score. For example, in the example above, the first sentiment characteristic is "cute: 1, elegant: 2, soft: 0, rough: 1, modern: 0, classic: 1 to include each first individual sentiment score. ".

In another embodiment, the first emotional characteristic may be to calculate only the first individual emotional characteristic in which each first individual emotional characteristic score is equal to or greater than a preset threshold. For example, in the above-described example, when the threshold value is 2 (or a ratio of 0.4), only the first individual emotion characteristic of'elegant' may be calculated as the first emotion characteristic. However, the calculation of the first emotional characteristic is not limited to the above-described example and can be calculated by various algorithms.

In one embodiment, since the first emotional characteristic is an emotional characteristic that is felt with respect to the appearance of the object, definitions perceived by each user for each first individual emotional characteristic may be different. However, since the first emotional characteristic is calculated by inputting the individual external characteristics to the standardized first emotional characteristic recognition model 320, the same first emotional characteristics are calculated for the same image data (ie, the same individual external characteristics). Can be.

For example, when a specific user recognizes that the user A is'cute' with respect to the image data related to the'elegant' jacket of the general definition, and inputs the image data to search for similar image data, the server may use Even if the first individual emotion characteristics of'elegant (defined as the standardized first emotion characteristic recognition model)' are calculated, the server provides the user with image data including the first individual emotion characteristics as similar image data. Since the provided image data will include the first emotion characteristic of'cute for user A (ie, the same as'elegant' of general definition)), there is no problem in the user's search.

In addition, in one embodiment, the method for acquiring abstract characteristics of image data according to an embodiment of the present invention may further include setting a personalized abstract characteristic for a user.

That is, as described above, the definition of the first individual emotion characteristic defined in the recognition model of the first emotional characteristic and the definition recognized by a specific user with respect to the first individual emotional characteristic may be different. The definition of the first emotional characteristic recognition model for a characteristic may be matched with the definition that the user thinks. Through this, when the user searches for image data using a specific search keyword, as described later, the personalized abstract characteristic can be used.

For example, if a specific user's'cute' first individual emotional characteristic of the first emotional characteristic recognition model is matched to be recognized as'elegant', the user uses the search keyword of "elegant jacket" to search for the video. When searching for data, the server calculates the first individual emotional characteristic of'cute' instead of'elegant' on the first emotional characteristic recognition model from the search keyword and provides the search result including the result according to the matching result. , Since the user will recognize the search result as'elegant', there is no problem in the user's search.

Also, in one embodiment, a definition recognized by a specific user with respect to the first individual emotional characteristic may be obtained in various ways. For example, a plurality of image data from which the first emotional characteristic is calculated may be provided to the user to receive and match the meaning felt by the user, but is not limited thereto.

In one embodiment, the abstract characteristic includes a second emotional characteristic. The second emotional characteristic is an emotional characteristic recognized based on information given to a product of a specific object, and includes a second individual emotional characteristic that is various emotional characteristics felt for different types of product information.

For example, the second individual emotion characteristics of'cheap','expensive' felt about product information of'price', or the second individual emotion characteristics of'fast','slow' felt about product information of'delivery time' It may include. That is, in regard to the user's preference for a specific object, information given to the product of the corresponding image data, such as price and delivery time, as well as appearance (design) may be an important consideration factor, and thus emotional characteristics of the user are additionally calculated .

Referring to FIG. 3, in one embodiment, the abstract characteristic recognition model 300 includes a second emotional characteristic recognition model 320 that receives product information about an object of image data and calculates a second emotional characteristic. do.

In addition, in one embodiment, the second emotional characteristic recognition model 320 includes a plurality of second individual emotional characteristic recognition modules for determining emotional characteristics for different product information. The second individual emotional characteristic recognition module calculates each second individual emotional characteristic for specific product information of an object of image data.

In addition, in one embodiment, the second emotional characteristic may be calculated in consideration of various information such as individual appearance characteristics, object type information, or user information as well as the product information.

For example, in the case of the second individual emotion characteristic of'price' among product information (for example,'expensive' or'cheap'), the criterion for determining the second individual emotion characteristic according to the object type, brand, or user type Since this may be different, the second emotional characteristic is calculated in consideration of various information including the same.

Subsequently, the server generates external descriptive data by combining a plurality of individual external characteristics and abstract characteristics calculated for the image data (S4200).

In one embodiment, the external description data generation step (S4200) extracts a code value corresponding to each of a plurality of individual appearance characteristics and abstract characteristics of the image data, and the code string form in which the plurality of code values are combined And generating external description data. That is, as the server codes the individual external characteristics and the abstract characteristics, external description data can be generated as a code string, thereby enabling efficient processing of external description data.

In one embodiment, when the abstract characteristics are calculated by inputting the outline description data combined with a plurality of individual appearance characteristics into the abstract characteristic recognition model, the code values corresponding to the abstract characteristics are extracted and added to the existing outline description data. You can simply update the appearance description data.

In addition, in one embodiment, the code value corresponding to the first emotional characteristic may include information about the summed score for each of the first individual emotional characteristics.

Specifically, in the above-described example, "Cute: 0.2, Elegant: 0.4, Soft: 0, Coarse: 0.2, Modern: 0, Classical: 0.2", the first emotional characteristics are calculated, and each first individual emotional characteristic If the code values corresponding to "Cute: Aa, Elegant: Ac, Soft: Ad, Coarse: Af, Modern: Ai, Classic: Ap", then the code descriptive data for the first emotional characteristic is "Aa20, Ac40, Ad00, Af20, Ai00, Ap20". In addition, if the code values corresponding to the individual appearance characteristics of'red' and'V-neck' are'Oa02 and'Bb02', the external description data of the video data is a combination of "Aa20, Ac40, Ad00, Af20, Ai00, Ap20" , Bb02, Oa02". However, the code system of the present invention is not limited to this and can be constructed in various ways.

Referring to FIG. 5, the method for acquiring abstract characteristics of image data according to an embodiment of the present invention further includes (S4300) the server generating recommended image data information based on appearance description data.

In one embodiment, the recommended image data information refers to information about recommended image data, which is image data that can be grouped with the image data because it is similar or related to specific image data.

For example, the recommended image data of the image data related to a specific jacket may be grouped with specific image data such as image data related to other jackets similar to the jacket or image data related to bottoms, accessories or interiors matching the jacket. Contains various image data.

In one embodiment, in the step S4300 of generating the recommended image data information, the server calculates the recommended outline description data matching the outline description data of the specific image data, and generates the recommended image data information for the image data including the same. It may be. That is, the recommended image data is calculated based on the calculated appearance description data, not the specific image data itself.

In one embodiment, the calculation of the recommended outline description data is based on a correlation between the calculated first individual emotion characteristics and each first individual emotion characteristics, as well as a plurality of individual appearance characteristics included in the outline description data. Recommended external descriptive data can be calculated.

In one embodiment, the degree of association between the first individual emotional characteristics includes similarity and dissimilarity between each of the first individual emotional characteristics. That is, the degree of association between the first individual emotional characteristics may refer to a degree of matching or disposition of other emotional characteristics with respect to specific emotional characteristics.

For example, the first individual emotional characteristics of'cute' match the first individual emotional characteristics of'pure','soft', and'natural', but the first individual emotional characteristics of'sexy' and'rough' It can be out of place or can be deployed. Accordingly, an association degree between each first individual emotional characteristic is established, and based on this, the first individual emotional characteristic similar to or matching the first individual emotional characteristic of the input image data is included, and the arranged first individual emotional characteristic is included. In order not to do so, it is possible to calculate the recommended external description data.

In addition, in one embodiment, the calculation of the recommended appearance description data may be calculated by further considering preference information for each of the individual appearance characteristics, the first individual emotion characteristics, or the second individual emotion characteristics of the specific user of the server. .

That is, in one embodiment, when preference information of each specific user as well as external description data of specific image data is input to the server, the server calculates the recommended external description data for the image data, and the user's personal preference. In consideration of the information, appropriate recommendation appearance data can be calculated for each input user, and based on this, personalized personalized recommendation image data information can be generated. That is, recommended image data information may be different according to a user for the same image data.

Referring to FIG. 6, the method for acquiring abstract characteristics of image data according to an embodiment of the present invention further includes a step (S4400) in which the server matches one or more first individual emotion characteristics with keywords; That is, one or more first individual emotional characteristics may be matched to a specific word.

For example, match the first individual sentiment characteristics of'Modern' and'Simple' to the keyword "Dandy", or'Sexy' or'Luxury' to the keyword "Party" Luxe' can match the first individual sentiment characteristics.

In addition, in one embodiment, when a new keyword is input to the server, one or more first individual emotion characteristics may be matched to the new keyword based on the first individual emotion characteristic matched to an existing keyword similar to the new keyword. have.

In addition, in one embodiment, the server transmits the image data corresponding to the search keyword received from the user client to the user client (S4500); further includes. That is, when a user inputs a search keyword to search for image data, image data corresponding to the search keyword may be extracted and transmitted to the user client as a search result.

In one embodiment, extraction of image data corresponding to the search keyword extracts object type information, individual appearance characteristics, first individual emotion characteristics, or second individual emotion characteristics that match the search keyword, and extracts the extracted characteristics. It may be to extract the image data of the external descriptive data to include.

For example, when a user inputs "cheap V-neck shirt suitable for a party" as a search keyword,'shirt' as the object type information from the search keyword,'V-neck' as the individual appearance characteristic, and'cheap' as the second individual emotional characteristic ', and'Sexy' and'Luxurious', which are the first individual emotion characteristics matched in step S4400 with respect to the keyword "party", may be extracted. Accordingly, the server can transmit image data having external descriptive data including the extracted characteristics to the user client as a search result.

For another example, even if the user inputs only the search keyword of "party", the user extracts the first individual emotional characteristics of'sexy' and'luxury' as described above, and then uses the image data having external descriptive data including it. It can be sent to the client.

That is, by matching the first individual emotional characteristics to the keyword in step S4400, even if it is not a search keyword that directly expresses information on the type or appearance of a specific object, it is possible to calculate appropriate image data as a search result. There is.

In addition, in one embodiment, when extracting the image data corresponding to the search keyword, it may be extracted by further considering preference information for each of the user's individual appearance characteristics, first individual emotion characteristics, or second individual emotion characteristics. . In this case, even if the same search keyword is input, the search result may be different depending on the user inputting the search keyword.

Referring to FIG. 7, in the method of acquiring abstract characteristics of image data according to an embodiment of the present invention, the server inputs the calculated first emotional characteristics into a style recognition model 500, thereby inputting style information. And calculating (S4600).

In one embodiment, the style information is information about a style, look, or TPO (Time, Place, Occasion) for a specific form or method in fashion, for example, casual (Casual). , Party (Party), Dandy (Dandy), Girlish (Girlish) style (look). Each style type or definition can be set by the server, and can be freely changed or added.

In one embodiment, the style recognition model 500, the first sentiment characteristics (one or more first individual sentiment characteristics), as the area of each of a plurality of styles are set in the first emotional characteristic spatial map 520 It may be to calculate the style information of the location.

Referring to FIG. 8, in one embodiment, the first emotional characteristic spatial map 520 is based on an association between a plurality of first individual emotional characteristics, each of the first individual emotional characteristics, as shown in FIG. 8(a). Means a spatial map placed on a flat surface. That is, when each first individual emotional characteristic is input, a position on the first emotional characteristic spatial map of each first individual emotional characteristic may be determined.

For example, the input first sentiment is "Elegant: 2, Practical: 1, Cheerful: 2, Light: 4, Soft: 1, Lovely: 3, Feminine: 3, Gorgeous: 1, Sexy: 2, Delicate: 2, Cool: 3, Natural: 1, Mild: 2", when calculated to include scores of each first individual emotion characteristic, as shown in FIG. 8(a), the position of each first individual emotional characteristic And a point 530 on the first emotional characteristic spatial map based on the score.

In addition, in one embodiment, as shown in FIG. 8( b), a region 540 of each of a plurality of styles may be set in the first emotional characteristic spatial map. That is, the style region may be preset in consideration of the degree of association of individual first emotional characteristics for each style, and each style region may include overlapping portions.

That is, in one embodiment, a plurality of styles of each region 540 is set in the first emotional characteristic spatial map 520 in the style recognition model 500, and the first emotional characteristic calculated from specific image data is the above. When each first individual emotion characteristic is displayed on the first emotion characteristic spatial map by being input to the style recognition model, style information for the image data may be calculated based on the displayed first individual emotion characteristic. .

Specifically, referring to FIG. 8(b), when the first emotional characteristic of the above-described example is input and a plurality of style areas are displayed on the set first emotional characteristic spatial map, the most first individual emotional characteristic 530 The'Romantic style', which is the style of the located style area 543, may be calculated as style information.

In addition, in one embodiment, a plurality of style information may be calculated, and in this case, a plurality of style information may be set by setting a priority based on the number of first individual emotion characteristics 530 located in each style area 540. Can be calculated.

Specifically, in the case of calculating a plurality of style information in the above-described example, based on the number of first individual emotion characteristics located in each style area,'Romantic style 543','Elegance style 544 )'and'Casual (541)' in order of order to set a priority order to calculate a plurality of style information.

In one embodiment, the calculated style information may be used to generate recommended image data information in step S4300 described above. That is, when style information is calculated based on appearance description data of specific image data, image data having the same or similar style information as the style information may be generated as recommended image data information.

In addition, in one embodiment, the region set on the first emotional characteristic spatial map includes not only the above-described style region, but also regions of various concepts related to the first emotional characteristic, which is the perceived emotional characteristic of the object's appearance. Can be set.

In one embodiment, when the video data is video data including a plurality of frames, the abstract characteristic calculation step (S4100) may be characterized in that it is performed for each frame in the video data, and external appearance The data generation step S4200 may be characterized by sequentially generating and generating a plurality of individual external characteristics and abstract characteristics for each frame.

An apparatus for acquiring abstract characteristics of image data according to another embodiment of the present invention includes at least one computer, and performs the above-mentioned abstract characteristic acquisition method.

Referring to FIG. 9, the server 10 for obtaining abstract characteristics of image data according to an embodiment of the present invention includes an abstract characteristic recognition model 300, an external description data generation unit 600, and a database 800. And performs the above-described abstract characteristic acquisition method.

In addition, in another embodiment, the server device 10 may include an object type recognition model 100, an appearance characteristic recognition model 200, a detailed type recognition model 400, a style recognition model 500, or a recommended image data generation unit. One or more of 700 may be further included.

In addition, the method for obtaining abstract characteristics of image data according to an embodiment of the present invention described above may be implemented as a program (or application) to be executed in combination with a computer that is hardware, and stored in a medium.

The above-described program is C, C++, JAVA, machine language, etc., in which a processor (CPU) of the computer can be read through a device interface of the computer in order for the computer to read the program and execute the methods implemented as a program. It may include a code (Code) coded in the computer language of. Such code may include functional code related to a function or the like that defines necessary functions for executing the above methods, and control code related to an execution procedure necessary for the processor of the computer to execute the functions according to a predetermined procedure. can do. In addition, the code may further include a memory reference-related code as to which location (address address) of the internal or external memory of the computer should be referred to additional information or media necessary for the computer's processor to perform the functions. have. In addition, when the processor of the computer needs to communicate with any other computer or server 10 in the remote in order to execute the functions, the code uses the communication module of the computer to remotely operate any other computer. The server 10 may further include a communication-related code on how to communicate, and what information or media should be transmitted/received during communication.

The storage medium refers to a medium that stores data semi-permanently and that can be read by a device, rather than a medium that stores data for a short moment, such as a register, cache, memory, or the like. Specifically, examples of the medium to be stored include, but are not limited to, ROM, RAM, CD-ROM, magnetic tape, floppy disk, and optical data storage device. That is, the program can be stored in various recording media on various servers or various recording media on the user's computer that the computer can access. In addition, the medium is distributed in a computer system connected by a network, so that the computer readable code may be stored in a distributed manner.

The embodiments of the present invention have been described with reference to the accompanying drawings, but a person skilled in the art to which the present invention pertains may implement the present invention in other specific forms without changing its technical spirit or essential features. You will understand. Therefore, it should be understood that the above-described embodiments are illustrative in all respects and not restrictive.

Claims (13)

A step of the server inputting a plurality of individual appearance characteristics calculated on the image data into the abstract characteristic recognition model, thereby calculating an abstract characteristic including the first emotional characteristic;
Generating an external descriptive data by combining the plurality of individual external characteristics and the abstract characteristics calculated by the server; And
And the server setting personalized abstract characteristics for a specific user,
The individual external characteristics are to express various external characteristics within the external classification criteria, which are specific classification criteria for describing the external shape of the subject,
The first emotional characteristic includes a plurality of first individual emotional characteristics, which are specific emotional characteristics recognized based on the appearance of the object,
The step of setting the personalized abstract characteristics,
A method for acquiring abstract characteristics of image data, characterized in that, by receiving a definition recognized by the user for a specific first individual emotional characteristic, matching with the definition of the specific first individual emotional characteristic set on the abstract characteristic recognition model . According to claim 1,
The individual external characteristics are calculated by inputting the image data into each of the plurality of individual characteristic recognition modules for determining different external classification criteria in the external characteristic recognition model,
The individual characteristic recognition module calculates the individual external characteristics included in the specific external classification criteria of the image data. According to claim 1,
The appearance classification criteria include a specialized appearance classification criterion applied only to a specific type of object and a general-purpose appearance classification criterion applied to all types of objects. According to claim 1,
The abstract characteristic recognition model includes a first emotional characteristic recognition model,
The first emotional characteristics recognition model,
As the scores for each of the plurality of first individual emotional characteristics are set for each individual external characteristic, the first emotional characteristics are calculated by adding up the scores for each of the first individual emotional characteristics set in the input multiple individual external characteristics. The method of obtaining abstract characteristics of image data. The method of claim 4,
The external descriptive data generation step,
By extracting a code value corresponding to each of the plurality of individual appearance characteristics and the first emotional characteristics for the image data, to generate the appearance description data in the form of a code string in which the plurality of code values are combined,
The code value corresponding to the first emotional characteristic,
A method for acquiring abstract characteristics of image data, characterized in that it comprises information on the sum of the scores for each of the first individual emotional characteristics. delete According to claim 1,
And generating, by the server, recommendation image data information that is information on one or more image data including the recommendation outline description data by calculating recommendation outline description data matching the outline description data. How to acquire the abstract characteristics of. The method of claim 7,
The recommended image data information generating step,
Characterized in that the server calculates recommended appearance description data based on the correlation between the appearance description data and the first individual emotional characteristics, and generates recommended image data information,
The method of obtaining abstract characteristics of image data, wherein the degree of association between the first individual emotional characteristics includes similarity and dissimilarity between each of the first individual emotional characteristics. The method of claim 8,
The recommended image data information generating step,
Characterized in that the server calculates recommended external description data based on the external description data, the correlation between the first individual emotional characteristics, and user preference information, and generates recommended image data information,
The user preference information is a preference information for each of the first individual emotional characteristics of the user, a method for obtaining abstract characteristics of image data. According to claim 1,
The server matching one or more first individual sentiment characteristics with keywords; And
Further comprising the step of extracting a first individual sentiment characteristic matching the search keyword received from the user client by the server, and transmitting the image data including the first individual sentiment characteristic to the user client. Method of obtaining abstract characteristics. According to claim 1,
The server further comprises the step of calculating style information by inputting the calculated first emotional characteristics into a style recognition model.
The style recognition model,
As each area of a plurality of styles is set in the first emotional property spatial map, the style information of the area in which the input first emotional property is located is calculated,
The first emotional characteristic spatial map is a spatial map in which a plurality of first individual emotional characteristics are arranged on a plane based on a correlation between each of the first individual emotional characteristics, and the method for obtaining abstract characteristics of image data. A server apparatus comprising at least one computer and executing the method of any one of claims 1 to 5 or 7 to 11. A program for acquiring abstract characteristics of image data stored in a medium to execute the method of any one of claims 1 to 5, 7 to 11, in combination with a computer that is hardware.

Images