WO2020149687A1 - 뷰티 콘텐츠 기반 데이터베이스 구축 방법 및 이를 이용한 뷰티 콘텐츠 또는 뷰티 오브젝트 검색 방법 - Google Patents
뷰티 콘텐츠 기반 데이터베이스 구축 방법 및 이를 이용한 뷰티 콘텐츠 또는 뷰티 오브젝트 검색 방법 Download PDFInfo
- Publication number
- WO2020149687A1 WO2020149687A1 PCT/KR2020/000840 KR2020000840W WO2020149687A1 WO 2020149687 A1 WO2020149687 A1 WO 2020149687A1 KR 2020000840 W KR2020000840 W KR 2020000840W WO 2020149687 A1 WO2020149687 A1 WO 2020149687A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- beauty
- content
- data
- user
- database
- Prior art date
Links
- 230000003796 beauty Effects 0.000 title claims abstract description 294
- 238000000034 method Methods 0.000 title claims abstract description 46
- 230000000007 visual effect Effects 0.000 claims description 23
- 238000000605 extraction Methods 0.000 claims description 18
- 230000036555 skin type Effects 0.000 claims description 10
- 238000010276 construction Methods 0.000 claims description 7
- 239000006071 cream Substances 0.000 description 26
- 230000003020 moisturizing effect Effects 0.000 description 22
- 230000000694 effects Effects 0.000 description 8
- 238000004891 communication Methods 0.000 description 5
- 239000000284 extract Substances 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 239000002453 shampoo Substances 0.000 description 3
- 239000002537 cosmetic Substances 0.000 description 2
- 238000013135 deep learning Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 210000004761 scalp Anatomy 0.000 description 2
- 238000000926 separation method Methods 0.000 description 2
- 241001237961 Amanita rubescens Species 0.000 description 1
- 241000692783 Chylismia claviformis Species 0.000 description 1
- 206010013786 Dry skin Diseases 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 238000013480 data collection Methods 0.000 description 1
- 230000037336 dry skin Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000001151 other effect Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 230000003813 thin hair Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
Definitions
- the present invention relates to a method for building a database based on beauty content and a method for searching for beauty content or beauty objects using the built database.
- viewers have to search directly in order to obtain information about products appearing in beauty content, and if they do not know exactly the product name of the product, there is a problem in that it is difficult to obtain a desired search result.
- the present invention for solving the problems as described above is intended to provide a search result of a beauty object or a beauty object desired by a user only with a search term such as usage or effects when the product name or the like of the product to be searched is not known.
- the present invention is to provide search results of beauty content or beauty objects suitable for a user, including information about products appearing in beauty content, as well as surrounding information about the user.
- a method of building a database according to an embodiment of the present invention for solving the above-described problem includes collecting content data including object data and context data from beauty content; And constructing a database based on the beauty content and the content data, wherein the object data is data related to a beauty object applied to beauty treatment of characters appearing in the beauty content.
- the context data is data related to the beauty content other than the beauty object.
- the content data further includes action data related to the movement or use of the beauty object
- the content data acquisition step includes one movement or use of the beauty object appearing in the beauty content. It is characterized by dividing into the above steps and collecting content data for each step.
- the beauty content is divided into one or more sections according to the divided step, keyframes for each section are set, and content data for each step is collected for each set keyframe. It is characterized by.
- the context data is collected as a preset item according to the collected object data, and includes visual context data or non-visual context data
- the visual context data includes the image of the beauty content. It is data that can be collected by analysis
- the non-visual context data is data that can be collected by analyzing voice or storage information other than the image of the beauty content.
- the visual context data includes the skin tone, skin type, eye color, race, or face shape of the character appearing in the beauty content
- the non-visual context data is generated by the beauty content Region, weather, season or time, place, and situation (TPO) at which the beauty object was used.
- a method for searching for beauty content or a beauty object for solving the above-described problems includes: receiving a search word input by a user from a user client; Collecting user data for the user who entered the search term; Matching the search word, the user data, and the content data based on a database and generating search result information of a beauty content or a beauty object according to the matching result; And transmitting the search result information to the user client, wherein the database is constructed based on content data including beauty content and object data, context data or action data collected from the beauty content.
- the object data is data about a beauty object applied to a beauty treatment of a character appearing in the beauty content
- the context data is collected as a preset item according to the collected object data
- Data related to the beauty content other than the beauty object and the action data is data related to movement or use of the beauty object.
- the step of generating the search result information may match the search word with the object data, context data, or action data, and first extract one or more beauty contents or beauty objects from the database according to the matching result. step; Matching the user data with context data of the first extracted beauty content or beauty object, and secondly extracting one or more beauty content or beauty objects from the first extracted beauty content or beauty object according to the matching result; And generating search result information based on the second extracted beauty content or beauty object.
- the second extraction step is characterized in that the extraction is based on the similarity and weight of the context data of the first extracted beauty content or beauty object matched to the user data, the weight is , Is a preset weight for each context data item according to the first extracted beauty content or object data of the beauty object.
- the second extraction step further considers the probability of purchasing the beauty object of the user calculated based on the search term, the user data, or the object data of the first extracted beauty content or beauty object. Characterized by extraction.
- the method further includes transmitting matching result information to the user client according to a request for providing matching result information received from the user client.
- the beauty content-based database construction and search program according to another embodiment of the present invention is combined with hardware to execute the above-mentioned beauty content-based database construction and search method, and is stored in a medium.
- FIG. 1 is a relationship diagram of a database construction and search method according to an embodiment of the present invention.
- FIG. 2 is a flow chart of a database construction method according to an embodiment of the present invention.
- FIG. 3 is a diagram for explaining an interface for acquiring content data according to an embodiment of the present invention.
- FIG. 4 is a flowchart of a method for searching for beauty content or beauty objects according to an embodiment of the present invention.
- FIG. 5 is a flow chart of a method for searching for beauty content or beauty objects including first and second extraction steps according to an embodiment of the present invention.
- FIG. 6 is a flowchart of a method for searching for beauty content or a beauty object including the step of providing matching result information according to an embodiment of the present invention.
- 'beauty content' is video content in which one or more beauty objects appear, as well as content that directly uses the beauty object, characters using specific objects appear, or beauty objects appear through voice or text. Contains content.
- client' refers to all devices including communication functions that a user can install and use a program (or application).
- the client is a cellular phone, PC phone, mobile terminal of synchronous/asynchronous IMT-2000, palm PC, personal digital assistant, smart phone, swap phone, mobile game machine, tablet PC, smart watch, notebook PC, desktop PC, smart It may include various communication devices such as a camera and a smart TV.
- the client does not basically include a communication function, but may include a device capable of performing communication by combining a memory chip having a communication function.
- 'content provider client' means a client of a content provider that uploads specific beauty content to a server.
- the content provider client is the client of the person uploading the video to the platform.
- 'user client' means a client of a user who performs a search for beauty content or beauty objects by entering a search term.
- FIG. 1 is a relationship diagram of a database construction and search method according to an embodiment of the present invention.
- a server acquires beauty content, collects content data from the acquired beauty content, and builds a database based on the beauty content and the collected content data. To build.
- the beauty content or beauty object search method receives a search term and user data from a user client, and searches search result information generated by matching the search term, user data, and content data included in the database To the client.
- search term and user data from a user client
- search result information generated by matching the search term, user data, and content data included in the database To the client.
- FIG. 2 is a flow chart of a database construction method according to an embodiment of the present invention.
- a method of constructing a database includes collecting content data including object data and context data from beauty content (S200); And constructing a database based on the beauty content and the content data (S400).
- S200 object data and context data from beauty content
- S400 content data
- the server collects content data including object data and context data from the beauty content (S200).
- the object data is data related to a beauty object applied to a beauty treatment of characters appearing in beauty content.
- the object data may include a specific beauty object type, product name, use, effect, or recommended usage, but is not limited thereto.
- the object data for the moisturizing cream faces an appropriate amount of'cream','A','moisturizing', and'twice a day' 'Apply to'.
- the context data is data related to the beauty content other than the beauty object. That is, the context data includes a wide range of surrounding information in addition to the beauty objects appearing in the beauty content.
- the context data includes visual context data and non-visual context data.
- the visual context data is data that can be collected by analyzing an image of the beauty content.
- the visual context data may include the skin tone, skin type, eye color, race, or face shape of the characters appearing in the beauty content, but this is exemplary and is not limited thereto.
- the visual context data of the content is the creator's'light skin tone','dry skin type','brown eyes','Hwang Injong' And the like.
- the non-visual context data is data that can be collected by analyzing voice or storage information other than the image of the beauty content.
- the stored information is information stored about the beauty content itself, and includes, for example, a region, date, and the like of the beauty content.
- the non-visual context data may include environment information on a region, weather, or season in which the beauty content is generated, but this is exemplary and is not limited thereto.
- the video content may include information about a region, weather, season, or region of residence of the creator appearing in the content.
- a beauty object not only the object data for the object itself, but also the visual context data including the user's skin tone, skin type, eye color, and race, and the non-visual context data including environment information such as the weather in the region where the user is located are specified. Since it has an important meaning in the selection of an object, various context data including it are obtained from beauty content.
- the non-visual context data may include time, place, and situation (Time, Place, Occasion; TPO) information in which the beauty object was used.
- time, place, and situation Time, Place, Occasion; TPO
- the non-visual context data of the content may include a'Christmas party'.
- the context data is characterized in that it is collected as a preset item according to the collected object data. That is, the type of context data to be collected may be determined according to object data collected from specific beauty content.
- a'moisturizing cream' information such as a user's'skin type' and a user's region's weather are important. Therefore, when the object data of'moisturizing' and'cream' is collected from the video content related to the moisturizing cream, the'skin' of the creator, the'dry' corresponding to the'weather' item in the region of residence, and the'average below zero' 20 degrees' of context data.
- the content data further includes action data related to movement or use of the beauty object.
- the action data may include a user tool, a sequence, usage, or usage when using a specific beauty object in beauty content, but this is exemplary and is not limited thereto.
- the action data may include'apply','hand', '10ml', etc. have.
- the above-described content data acquisition step (S200) is characterized in that the movement or use of the beauty object appearing in the beauty content is divided into one or more steps, and content data is collected for each step.
- the step classification of the movement or use of the beauty object may be a step classification for the use of one beauty object.
- the moisturizing cream A it can be divided into'steps to apply the appropriate amount of cream to the face' and'steps to apply the removed cream with fingers', and data for each step can be obtained.
- the step separation of the movement or use of the beauty object may be a step separation for each object for use of a plurality of beauty objects.
- the B face oil when used after the use of the A moisturizing cream, it may be divided into'the step of applying the A cream' and'the step of applying the B face oil'.
- the step of acquiring the content data divides the beauty content, which is video content, into one or more sections according to the divided step, sets keyframes for each section, and sets each step of the keyframe content data. It is characterized by collecting.
- the step of obtaining content data from the above-described beauty content (S200) may be obtained by transmitting beauty content to one or more expert clients by the server and receiving content data for the transmitted beauty content.
- FIG. 3 is a view for explaining an interface for acquiring content data according to an embodiment of the present invention.
- an embodiment of a method of obtaining content data by an expert client receiving beauty content from a server will be described with reference to FIG. 3.
- the video content is divided into one or more sections using the section setting bars 22 and 24.
- the waveform 10 of the image of the divided section may be displayed separately as shown in FIG. 3.
- the set keyframe image may display a work image screen 30 and an area 50 for inputting content data for the keyframe.
- the next section may be set again using the section setting bars (22, 24), and then the keyframe setting and content data collection steps may be repeated.
- an area 40 for inputting content data for the entire image, not for each section of the specific beauty content may be displayed separately, and the content data input to the area 40 may be independent of the divided section. It may be input once for the entire video content.
- input of content data may be input in a drop-down box or a subjective form.
- the'content data recognition model' includes an'object data recognition module', a'context data recognition module' or a'action data recognition module', and each recognition module includes a plurality of beauty contents for learning.
- content data may be matched and trained through a machine learning or deep learning learning model.
- a database is constructed based on the plurality of beauty contents and the content data (S400). That is, a database in which a plurality of beauty contents and content data obtained therefrom is matched is constructed.
- a method of searching for beauty content or beauty objects using the database will be described in detail later.
- FIG. 4 is a flowchart of a method for searching for beauty content or beauty objects according to an embodiment of the present invention.
- a method for searching for beauty content or a beauty object includes: receiving a search word input by a user from a user client (S1200); Collecting user data for the user who input the search term (S1400); Matching the search term, the user data, and the content data based on a database and generating search result information of a beauty content or a beauty object according to the matching result (S1600); And transmitting the search result information to the user client (S1800).
- the user inputs a search word input from the user client (S1200).
- the search word may be related to a beauty object, but is not limited thereto and includes all search words.
- the user data may correspond to the context data described above.
- the user data may include the skin tone, skin type, eye color, race, face type, residential area, weather, or season of the user who entered the search word.
- the method for obtaining the user data is not limited to a specific method, and includes all methods that are input from the user or extracted by the server from the user or user client.
- search term Based on the database, the search term, the user data, and content data are matched, and search result information of a beauty content or a beauty object is generated according to the matching result (S1600).
- the content data includes context data and action data as well as object data, even when a beauty object is not specified as a search term (that is, even when the search term does not correspond to the object data) ), if the search term corresponds to action data or context data, it is possible to search for beauty content or beauty objects.
- the server matches the beauty content or the beauty object in the database based on the user data and the action data of the "right” to the user as a search result Can provide.
- the search result information is generated by including priority information between matched beauty content or beauty objects. That is, it may include information on the order of beauty content or beauty objects provided to the user.
- the priority information may be generated based on the similarity of matched search terms, user data and content data, and respective weights. Accordingly, the search result may be different depending on the user even if the search term is the same. Detailed embodiments will be described later.
- FIG. 5 is a flow chart of a method for searching for beauty content or beauty objects including first and second extraction steps according to an embodiment of the present invention.
- the search result information generating step (S1600) matches the search word with the object data, context data or action data, and one or more beauty contents from the database according to the matching result or Extracting a first beauty object (S1620); Matching the user data with the context data of the first extracted beauty content or beauty object, and secondly extracting one or more beauty content or beauty objects from the first extracted beauty content or beauty object according to the matching result ( S1640); And generating search result information based on the second extracted beauty content or beauty object (S1660).
- S1620 Matching the user data with the context data of the first extracted beauty content or beauty object, and secondly extracting one or more beauty content or beauty objects from the first extracted beauty content or beauty object according to the matching result ( S1640);
- S1660 search result information based on the second extracted beauty content or beauty object
- the search term and the object data, context data, or action data are matched, and one or more beauty contents or beauty objects are first extracted from the database according to the matching result (S1620).
- search result information generation step (S1660) described below when the search word does not correspond to any of object data, context data, or action data, in the search result information generation step (S1660) described below, information on no search result may be generated.
- the search term when the search term includes information matching one or more of object data, action data, or context data, the beauty content or beauty object including the object data, action data, or context data matching the search word is selected. First extraction from the database.
- the server extracts beauty content or beauty objects including object data of "moisturizing” or "cream” from the database.
- beauty content including both'moisturizing' and'cream' may be preferentially extracted than beauty content including either.
- the server may extract beauty content or beauty objects that include "20 degrees below zero" context data from the database.
- a suitable search result for each user may be provided by extracting the first extracted beauty content or beauty object based on the search word based on the user data.
- the search term is "moisturizing cream" and the user data is'dry' and'zero below 20 degrees'
- the first extracted beauty content or beauty object including object data of'moisturizing' or'cream' Among them, the beauty content or the beauty object including the context data of'dry' or '20 degrees below zero' may be extracted second.
- search result information to be described later may be different. For example, if user A whose user data is'dry' and '20 degrees below zero' and user B whose'intellect' and'image 40' are searched with the same search term "moisturizing cream", the first extraction Even if the extracted beauty content or beauty object is the same beauty content or beauty object that includes object data of'moisturizing' or'cream', it is extracted to the user A by second extraction based on the user data in the second extraction step.
- Search results including'moisturizing','cream','dry' and '20 degrees below zero' are preferentially provided, and search results including'moisturizing','cream','intelligent' and'video 40 degrees' are provided to user B. Priority may be provided to provide different search results for each user.
- the second extraction step is characterized in that it is extracted based on the similarity and weight between the user data and the context data of the first extracted beauty content or beauty object matched with the user data.
- the degree of similarity between the user data and the context data of the first extracted beauty content or beauty object matched thereto may be scored according to the degree of similarity.
- the skin type context data matched with this may have a high score similarity in the order of dry-neutral-intelligent.
- the weight may be a preset weight for each context data item according to the object data of the first extracted beauty content or beauty object. That is, since the importance of a specific context data item varies according to the characteristics of various beauty objects, a search result can be provided to the user in consideration of this.
- the context data of the'skin type' and the'weather' item may be more important than the context data of the'skin tone' item. Accordingly, in this case, the weight of the context data of the'skin type' and the'weather' item may be set higher than the weight of the context data of the'skin tone' item for object data of'moisturizing' and'cream'.
- the second extraction step further considers the probability of purchasing the beauty object of the user calculated based on the search term, the user data, or the object data of the first extracted beauty content or beauty object. Characterized by extraction.
- Search result information is generated based on the second extracted beauty content or beauty object (S1660).
- the search result information includes priority information between extracted beauty content or beauty objects.
- the similarity between the search word matched in the first extraction step (S1620) and the content data, and the similarity level between the user data and the context data matched in the second extraction step (S1640) are multiplied by the respective weights and added to the score. Accordingly, the priority of each beauty content or beauty object included in the generated search result information may be determined. Accordingly, as described above, even if the same search term is used, the search result may be different depending on the user.
- the search result information is transmitted to the user client (S1800).
- FIG. 6 is a flowchart of a method for searching for beauty content or a beauty object including the step of providing matching result information according to an embodiment of the present invention.
- the matching result information includes detailed information about a matching result between the search word, user data, and content data in the above-described search result information generation step (S1600). For example, it may include matched search terms, specific information about content data or user data, and information about similarity and weight according to matching.
- the method for searching for beauty content or beauty objects further includes receiving a purchase request from a user client.
- the purchase request includes receiving a user's touch input for a specific beauty object input to the user client.
- the user can enter the favorites after the beauty content ends, check the added beauty objects, and decide to purchase.
- the server may further include providing other beauty content associated with the beauty object added to the bookmark to the user client.
- other beauty content associated with the beauty object includes object data for the beauty object.
- context data matching the user data may be further included.
- the user when the user is hesitant to purchase due to the lack of information on the beauty object only by the beauty content viewed by the user, the user provides more information to the user by providing other beauty content in which the corresponding beauty object appeared, to assist in the purchase decision This has the effect of making it.
- the above-described method for constructing and retrieving a database based on beauty contents 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 steps of a method or algorithm described in connection with an embodiment of the present invention may be implemented directly in hardware, a software module executed by hardware, or a combination thereof.
- the software modules may include random access memory (RAM), read only memory (ROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, hard disk, removable disk, CD-ROM, or It may reside on any type of computer readable recording medium well known in the art.
Landscapes
- Business, Economics & Management (AREA)
- Engineering & Computer Science (AREA)
- Accounting & Taxation (AREA)
- Physics & Mathematics (AREA)
- Finance (AREA)
- General Business, Economics & Management (AREA)
- Marketing (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Economics (AREA)
- Tourism & Hospitality (AREA)
- Strategic Management (AREA)
- Human Resources & Organizations (AREA)
- Primary Health Care (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Development Economics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
Description
Claims (11)
- 서버가 동영상 콘텐츠인 뷰티 콘텐츠로부터 콘텐츠 데이터를 수집하여 데이터베이스를 구축하는 방법에 있어서,뷰티 콘텐츠로부터 오브젝트 데이터 및 콘텍스트 데이터를 포함하는 콘텐츠 데이터를 수집하는 단계; 및상기 뷰티 콘텐츠 및 상기 콘텐츠 데이터를 기초로 데이터베이스를 구축하는 단계;를 포함하고,상기 오브젝트 데이터는, 상기 뷰티 콘텐츠에 등장하는 등장인물의 미용 작업(beauty treatment)에 적용되는 뷰티 오브젝트에 관한 데이터이고,상기 콘텍스트 데이터는, 상기 뷰티 오브젝트 외의 상기 뷰티 콘텐츠에 관한 데이터인, 뷰티 콘텐츠 기반 데이터베이스 구축 방법.
- 제1항에 있어서,상기 콘텐츠 데이터는, 상기 뷰티 오브젝트의 움직임 또는 사용에 관한 액션 데이터를 더 포함하고,상기 콘텐츠 데이터 획득 단계는,상기 뷰티 콘텐츠에 등장하는 상기 뷰티 오브젝트의 움직임 또는 사용을 하나 이상의 단계로 구분하고, 각 단계별 콘텐츠 데이터를 수집하는 것을 특징으로 하는, 뷰티 콘텐츠 기반 데이터베이스 구축 방법.
- 제2항에 있어서,상기 콘텐츠 데이터 획득 단계는,상기 구분된 단계에 따라 상기 뷰티 콘텐츠를 하나 이상의 구간으로 분할하여 각 구간별 키프레임을 설정하고, 설정된 키프레임마다 각 단계별 콘텐츠 데이터를 수집하는 것을 특징으로 하는, 뷰티 콘텐츠 기반 데이터베이스 구축 방법.
- 제1항에 있어서,상기 콘텍스트 데이터는, 수집된 상기 오브젝트 데이터에 따라 미리 설정된 항목으로 수집되며, 시각적 콘텍스트 데이터 또는 비시각적 콘텍스트 데이터를 포함하고,상기 시각적 콘텍스트 데이터는, 상기 뷰티 콘텐츠의 이미지를 분석하여 수집 가능한 데이터이고,상기 비시각적 콘텍스트 데이터는, 상기 뷰티 콘텐츠의 이미지 외의 음성 또는 저장정보를 분석하여 수집 가능한 데이터인, 뷰티 콘텐츠 기반 데이터베이스 구축 방법.
- 제4항에 있어서,상기 시각적 콘텍스트 데이터는, 상기 뷰티 콘텐츠에 등장하는 등장인물의 피부톤, 피부타입, 눈동자색, 인종 또는 얼굴형을 포함하고,상기 비시각적 콘텍스트 데이터는, 상기 뷰티 콘텐츠가 생성된 지역, 날씨, 계절 또는 상기 뷰티 오브젝트가 사용된 시간, 장소, 상황(Time, Place, Occasion; TPO)을 포함하는, 뷰티 콘텐츠 기반 데이터베이스 구축 방법.
- 서버가 데이터베이스를 기반으로 사용자에게 뷰티 콘텐츠 또는 뷰티 오브젝트의 검색 결과를 제공하는 방법에 있어서,사용자 클라이언트로부터 사용자가 입력한 검색어를 수신하는 단계;상기 검색어를 입력한 사용자에 대한 사용자 데이터를 수집하는 단계;데이터베이스를 기반으로 상기 검색어, 상기 사용자 데이터 및 콘텐츠 데이터를 매칭하고, 매칭 결과에 따라 뷰티 콘텐츠 또는 뷰티 오브젝트의 검색 결과 정보를 생성하는 단계; 및상기 검색 결과 정보를 상기 사용자 클라이언트에 전송하는 단계;를 포함하고,상기 데이터베이스는, 뷰티 콘텐츠 및 상기 뷰티 콘텐츠로부터 수집한 오브젝트 데이터, 콘텍스트 데이터 또는 액션 데이터를 포함하는 콘텐츠 데이터를 기초로 구축된 것이고,상기 오브젝트 데이터는, 상기 뷰티 콘텐츠에 등장하는 등장인물의 미용 작업(beauty treatment)에 적용되는 뷰티 오브젝트에 관한 데이터이고,상기 콘텍스트 데이터는, 수집된 상기 오브젝트 데이터에 따라 미리 설정된 항목으로 수집되는 것으로, 상기 뷰티 오브젝트 외의 상기 뷰티 콘텐츠에 관한 데이터이고,상기 액션 데이터는, 상기 뷰티 오브젝트의 움직임 또는 사용에 관한 데이터인, 뷰티 콘텐츠 또는 뷰티 오브젝트 검색 방법.
- 제6항에 있어서,상기 검색 결과 정보 생성 단계는,상기 검색어와 상기 오브젝트 데이터, 콘텍스트 데이터 또는 액션 데이터를 매칭하고, 매칭 결과에 따라 상기 데이터베이스로부터 하나 이상의 뷰티 콘텐츠 또는 뷰티 오브젝트를 제1추출하는 단계;상기 사용자 데이터와 상기 제1추출된 뷰티 콘텐츠 또는 뷰티 오브젝트의 콘텍스트 데이터를 매칭하고, 매칭 결과에 따라 상기 제1추출된 뷰티 콘텐츠 또는 뷰티 오브젝트로부터 하나 이상의 뷰티 콘텐츠 또는 뷰티 오브젝트를 제2추출하는 단계; 및상기 제2추출된 뷰티 콘텐츠 또는 뷰티 오브젝트를 기초로 검색 결과 정보를 생성하는 단계;를 포함하는, 뷰티 콘텐츠 또는 뷰티 오브젝트 검색 방법.
- 제7항에 있어서,상기 제2추출단계는,상기 사용자 데이터와 이에 매칭된 상기 제1추출된 뷰티 콘텐츠 또는 뷰티 오브젝트의 콘텍스트 데이터 간의 유사도 및 가중치를 기초로 추출하는 것을 특징으로 하고,상기 가중치는, 상기 제1추출된 뷰티 콘텐츠 또는 뷰티 오브젝트의 오브젝트 데이터에 따라 콘텍스트 데이터 항목별 미리 설정된 가중치인, 뷰티 콘텐츠 또는 뷰티 오브젝트 검색 방법.
- 제7항에 있어서,상기 제2추출단계는,상기 검색어, 상기 사용자 데이터 또는 상기 제1추출된 뷰티 콘텐츠 또는 뷰티 오브젝트의 오브젝트 데이터를 기초로 산출된 상기 사용자의 뷰티 오브젝트 구매확률을 더 고려하여 추출하는 것을 특징으로 하는, 뷰티 콘텐츠 또는 뷰티 오브젝트 검색 방법.
- 제6항에 있어서,상기 사용자 클라이언트로부터 수신한 매칭 결과 정보 제공요청에 따라, 상기 사용자 클라이언트에 매칭 결과 정보를 전송하는 단계;를 더 포함하는, 뷰티 콘텐츠 또는 뷰티 오브젝트 검색 방법.
- 하드웨어인 컴퓨터와 결합되어, 제1항 내지 제10항 중 어느 한 항의 방법을 실행시키기 위해 매체에 저장된, 뷰티 콘텐츠 기반 데이터베이스 구축 및 검색 프로그램.
Applications Claiming Priority (4)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR10-2019-0006232 | 2019-01-17 | ||
KR20190006232 | 2019-01-17 | ||
KR10-2020-0006153 | 2020-01-16 | ||
KR1020200006153A KR102448617B1 (ko) | 2019-01-17 | 2020-01-16 | 뷰티 콘텐츠 기반 데이터베이스 구축 방법 및 이를 이용한 뷰티 콘텐츠 또는 뷰티 오브젝트 검색 방법 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2020149687A1 true WO2020149687A1 (ko) | 2020-07-23 |
Family
ID=71613825
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/KR2020/000840 WO2020149687A1 (ko) | 2019-01-17 | 2020-01-17 | 뷰티 콘텐츠 기반 데이터베이스 구축 방법 및 이를 이용한 뷰티 콘텐츠 또는 뷰티 오브젝트 검색 방법 |
Country Status (1)
Country | Link |
---|---|
WO (1) | WO2020149687A1 (ko) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116228459A (zh) * | 2023-05-06 | 2023-06-06 | 南京恒奥源医疗器械有限公司 | 用于面部注射组合物各成分配比数据分析系统 |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001104050A (ja) * | 1999-10-06 | 2001-04-17 | Seiko Epson Corp | 画像処理を用いた化粧支援方法および化粧台ならびに画像処理を用いた化粧支援処理プログラムを記録した記録媒体 |
KR20060077988A (ko) * | 2004-12-30 | 2006-07-05 | 삼성전자주식회사 | 멀티미디어 통신 시스템에서 컨텍스트 추출 및 이를 통한정보 제공 시스템 및 방법 |
KR20100023786A (ko) * | 2008-08-22 | 2010-03-04 | 정태우 | 영상에 포함된 객체를 색인하는 방법 |
KR20120051342A (ko) * | 2010-11-12 | 2012-05-22 | 한국전자통신연구원 | 사용자 색감 기반의 감성 화장 추천 시스템 및 방법 |
KR20170096904A (ko) * | 2016-02-17 | 2017-08-25 | 삼성전자주식회사 | 사용자의 피부 유형에 따른 콘텐트를 제공하기 위한 전자 장치 및 방법 |
-
2020
- 2020-01-17 WO PCT/KR2020/000840 patent/WO2020149687A1/ko active Application Filing
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2001104050A (ja) * | 1999-10-06 | 2001-04-17 | Seiko Epson Corp | 画像処理を用いた化粧支援方法および化粧台ならびに画像処理を用いた化粧支援処理プログラムを記録した記録媒体 |
KR20060077988A (ko) * | 2004-12-30 | 2006-07-05 | 삼성전자주식회사 | 멀티미디어 통신 시스템에서 컨텍스트 추출 및 이를 통한정보 제공 시스템 및 방법 |
KR20100023786A (ko) * | 2008-08-22 | 2010-03-04 | 정태우 | 영상에 포함된 객체를 색인하는 방법 |
KR20120051342A (ko) * | 2010-11-12 | 2012-05-22 | 한국전자통신연구원 | 사용자 색감 기반의 감성 화장 추천 시스템 및 방법 |
KR20170096904A (ko) * | 2016-02-17 | 2017-08-25 | 삼성전자주식회사 | 사용자의 피부 유형에 따른 콘텐트를 제공하기 위한 전자 장치 및 방법 |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN116228459A (zh) * | 2023-05-06 | 2023-06-06 | 南京恒奥源医疗器械有限公司 | 用于面部注射组合物各成分配比数据分析系统 |
CN116228459B (zh) * | 2023-05-06 | 2023-07-14 | 南京恒奥源医疗器械有限公司 | 用于面部注射组合物各成分配比数据分析系统 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2020251233A1 (ko) | 영상데이터의 추상적특성 획득 방법, 장치 및 프로그램 | |
WO2012060537A2 (ko) | 얼굴 및 스타일 인식 기반의 추천 시스템 및 그 방법 | |
WO2017213396A1 (ko) | 도서 추천 서비스 제공 장치 및 방법 | |
WO2015167074A1 (ko) | 토픽을 추출하고, 추출된 토픽의 적합성을 평가하는 방법 및 서버 | |
WO2012070840A2 (ko) | 컨센서스 검색 장치 및 방법 | |
WO2015005679A1 (ko) | 음성 인식 방법, 장치 및 시스템 | |
WO2021132802A1 (en) | Apparatus for video searching using multi-modal criteria and method thereof | |
WO2020085663A1 (ko) | 인공지능 기반의 자동 로고생성 시스템 및 이를 이용한 로고생성 서비스 방법 | |
WO2017032187A1 (zh) | 自动捕捉目标物的方法、装置及存储介质 | |
WO2011136425A1 (ko) | 개체명 사전 및 마이닝 규칙이 결합된 온톨로지 스키마를 이용한 리소스 기술 프레임워크 네트워크 구축 장치 및 방법 | |
Jaimes et al. | Learning personalized video highlights from detailed MPEG-7 metadata | |
WO2020149687A1 (ko) | 뷰티 콘텐츠 기반 데이터베이스 구축 방법 및 이를 이용한 뷰티 콘텐츠 또는 뷰티 오브젝트 검색 방법 | |
EP3545487A1 (en) | Electronic apparatus, controlling method of thereof and non-transitory computer readable recording medium | |
CN110781328A (zh) | 基于语音识别的视频生成方法、系统、装置和存储介质 | |
WO2012046906A1 (ko) | 다중 자원을 통합한 지식베이스를 이용하여 연구 주체간의 상관관계가 표시된 자원검색 정보 제공 장치 및 방법 | |
WO2021107449A1 (ko) | 음역 전환 신조어를 이용한 지식 그래프 기반 마케팅 정보 분석 서비스 제공 방법 및 그 장치 | |
WO2015133856A1 (ko) | 정답 키워드 제공 방법 및 장치 | |
CN113380271A (zh) | 情绪识别方法、系统、设备及介质 | |
Muehlbradt et al. | What's in an ALT Tag? Exploring Caption Content Priorities through Collaborative Captioning | |
Park et al. | Achieving real-time sign language translation using a smartphone's true depth images | |
WO2024085535A1 (ko) | 인공지능을 이용하여 이미지를 합성하는 방법 및 이미지 합성에 기반하여 헤어디자이너를 매칭하는 방법 | |
WO2021107445A1 (ko) | 지식 그래프 및 국가별 음역 전환 기반 신조어 정보 서비스 제공 방법 및 그 장치 | |
WO2020116703A1 (ko) | 개인 맞춤형 화장품 정보 제공 시스템 및 방법 | |
CN110008314B (zh) | 一种意图解析方法及装置 | |
CN111800650A (zh) | 视频配乐方法、装置、电子设备及计算机可读介质 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 20742088 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 20742088 Country of ref document: EP Kind code of ref document: A1 |
|
32PN | Ep: public notification in the ep bulletin as address of the adressee cannot be established |
Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 25.04.2022) |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 20742088 Country of ref document: EP Kind code of ref document: A1 |