KR20170070669A - Apparatus and method for recommending segment image for minimizing heterogeneity - Google Patents

Apparatus and method for recommending segment image for minimizing heterogeneity Download PDF

Info

Publication number
KR20170070669A
KR20170070669A KR1020150178477A KR20150178477A KR20170070669A KR 20170070669 A KR20170070669 A KR 20170070669A KR 1020150178477 A KR1020150178477 A KR 1020150178477A KR 20150178477 A KR20150178477 A KR 20150178477A KR 20170070669 A KR20170070669 A KR 20170070669A
Authority
KR
South Korea
Prior art keywords
image
user
images
module
recommendation
Prior art date
Application number
KR1020150178477A
Other languages
Korean (ko)
Inventor
박윤경
문경덕
노경주
Original Assignee
한국전자통신연구원
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 한국전자통신연구원 filed Critical 한국전자통신연구원
Priority to KR1020150178477A priority Critical patent/KR20170070669A/en
Publication of KR20170070669A publication Critical patent/KR20170070669A/en

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/80—Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
    • H04N21/85—Assembly of content; Generation of multimedia applications
    • H04N21/854—Content authoring
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
    • H04N21/4508—Management of client data or end-user data
    • H04N21/4532—Management of client data or end-user data involving end-user characteristics, e.g. viewer profile, preferences
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
    • H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
    • H04N21/4668—Learning process for intelligent management, e.g. learning user preferences for recommending movies for recommending content, e.g. movies
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
    • H04N21/47—End-user applications
    • H04N21/475—End-user interface for inputting end-user data, e.g. personal identification number [PIN], preference data
    • H04N21/4755—End-user interface for inputting end-user data, e.g. personal identification number [PIN], preference data for defining user preferences, e.g. favourite actors or genre
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
    • H04N21/80—Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
    • H04N21/81—Monomedia components thereof
    • H04N21/8126—Monomedia components thereof involving additional data, e.g. news, sports, stocks, weather forecasts
    • H04N21/8133—Monomedia components thereof involving additional data, e.g. news, sports, stocks, weather forecasts specifically related to the content, e.g. biography of the actors in a movie, detailed information about an article seen in a video program

Abstract

A sculpture image recommendation apparatus and method therefor which minimizes heterogeneity between images are disclosed. The image recommendation apparatus includes a user application module that receives a piece of video request for creating a combined image from a user together with a story, a search module that searches for pieces of images related to the inputted story, A recommendation module for recommending pieces of images to the user in consideration of the accuracy of the retrieved piece images, the degree of association between the front and rear images to match the story flow, and the user's intention.

Figure P1020150178477

Description

Technical Field [0001] The present invention relates to a sculpture image recommendation apparatus and a method thereof for minimizing heterogeneity between images,

The present invention relates to image editing and recommendation techniques.

As the use of smartphones explosively increases and the culture of creating content by oneself spreads, vast amount of image information is accumulating. However, the vast amount of image information is mostly consumed as one-time consumer goods that transmit simple messages, and the potential value of image data as an information carrier is not fully utilized.

Since the image data has a faster understanding and recognition speed than the text data, it can be utilized as a more intuitive information transmission means. In order to utilize the image data as a new information transmission means, first, the image should be divided and stored and managed. At this time, the image should be divided into meaning units, and the meaning of each divided image should be stored and managed together.

When a user requests a video having a predetermined meaning when a user wants to create a new video for information transmission, the service provider finds a necessary piece of video from the repository and supports the user to produce a video. The user can create a new image by collecting a plurality of pieces of sculptural images thus provided, thereby creating added value from the previously produced image.

According to an embodiment of the present invention, a sculpture image recommendation apparatus and a method thereof are provided which are adapted to a user's intention and minimized heterogeneity between images.

The image recommendation apparatus includes a user application module that receives a piece of video request for creating a combined image from a user together with a story, a search module that searches for pieces of images related to the inputted story, A recommendation module for recommending pieces of images to the user in consideration of the accuracy of the retrieved piece images, the degree of association between the front and rear images to match the story flow, and the user's intention.

The image has higher value as an information transferring material because it can transmit intuitive information compared to other information transferring means. The existing image production environment requires expert and expensive equipments, so there is a limit to anyone to easily produce images and transmit information. However, according to the present invention, when a user combines sculptured images to produce a new image, it is possible to recommend sculptured images with minimized heterogeneity between sculptured images. In order to minimize heterogeneity, we recommend sculptured images considering search accuracy, user 's intention, and similarity between before and after images, so that anyone can easily create new images from recommended sculptured images. Furthermore, it enables the activation of a new industry called image recycling.

1 is a configuration diagram of a video recommendation apparatus according to an embodiment of the present invention;
FIG. 2 is a detailed configuration diagram of the recommendation module of FIG. 1 according to an embodiment of the present invention;
3 is a diagram illustrating a combined image generated by combining a story created by a user, a retrieved piece image, and a recommended piece image according to an exemplary embodiment of the present invention.
4A and 4B are flowcharts illustrating a sculptural image recommendation method according to an exemplary embodiment of the present invention.

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings. In the following description of the present invention, a detailed description of known functions and configurations incorporated herein will be omitted when it may make the subject matter of the present invention rather unclear. In addition, the terms described below are defined in consideration of the functions of the present invention, which may vary depending on the intention of the user, the operator, or the like. Therefore, the definition should be based on the contents throughout this specification.

When a user requests a service provider to recommend a sculpture image in order to create a new image by combining sculptured images, there is a method of using a content-based, collaborative-based, knowledge-based, and situation-based filtering for recommendation. However, these methods analyze user 's usage history to recommend items that are similar to those used mainly, or recommend methods using preferences of people who have similar tendencies to users. This recommendation method is suitable for recommending individual items, books, or movies, but is not suitable for recommending sculptural images for producing new images. When a new image is produced using a sculptured image, it is necessary to combine sculptured images taken from different originals, so that it is necessary to minimize heterogeneity between sculptured images.

The present invention proposes a device and a method for recommending a sculptured image in which not only prediction of what a user likes and recommends but also disconnection of a story and minimization of heterogeneity between sculptured images is proposed. When a service provider recommends a sculptured image to a user by a recommendation of a user, the service provider does not provide all of a large number of images to be searched, but provides a sculpture with a minimized heterogeneity to suit the intention of the user and the flow of the already- It is recommended to select images. Accordingly, the user can combine pieces of sculptured images that are fit for the user's intention and have high similarity. Hereinafter, an image recommendation apparatus and a method thereof will be described in detail with reference to the following drawings.

1 is a configuration diagram of a video recommendation apparatus according to an embodiment of the present invention.

Referring to FIG. 1, the image recommendation apparatus 1 includes a pre-processing unit 10, a post-processing unit 12, and a storage unit 14. The pre-processing unit 10 includes an analysis module 100 and a management module 102. The post-processing unit 12 includes a user application module 120, a search module 122, a recommendation module 124, ).

The video recommendation apparatus 1 according to one embodiment is a user terminal. As another example, the video recommendation apparatus 1 may be a server connected to a user terminal via a network. In this case, a user can be connected to the server by executing an app installed or downloaded to the user terminal. Although the preprocessing unit 10, the post-processing unit 12 and the storage unit 14 are illustrated as being located in the video recommendation apparatus 1 in FIG. 1, the components may be physically separated or integrated according to their functions . The pre-processing unit 10 and the post-processing unit 10 may be implemented by a processor, and the storage unit 14 may be implemented by a memory.

The analysis module 100 of the preprocessing unit 10 analyzes the image data and divides the image data into fragmented images that are the minimum units of meaning that can be recycled. The management module 102 stores fragment images divided through the analysis module 100 in the storage unit 14 and saves and manages the semantic data in the storage unit 14 with respect to the fragment images. The analysis module 100 divides the image data into the minimum units meaningful so that the user can combine the fragmented images to produce a new image and the semantic data divided into the minimum units are stored in the storage unit 14 . The image can be reused as an information representation and transmission means by allowing only the portion required by the user among the semantic data stored in the storage unit 14 to be reusable.

The user application module 120 of the post-processing unit 12 receives a story for producing a combined image from a user and provides sculptured images reproduced through the recommendation module 124 to a user. The search module 122 searches for pieces of images related to a story input from a user. The recommendation module 124 recommends to the user a sculptured image suited to the story flow among the sculptural images retrieved through the search module 122. [ The reproduction module 126 combines and reproduces the recommended piece images through the recommendation module 124 in a desired order.

2 is a detailed block diagram of the recommendation module of FIG. 1 according to an embodiment of the present invention.

Referring to FIGS. 1 and 2, the recommendation module 124 includes a user information management unit 1240, a log management unit 1242, a search management unit 1244, and a recommendation engine 1246.

The user information management unit 1240 receives user information and designates an explicit request of the user. The user information includes a user preference model, a user profile, and the like. The log management unit 1242 stores and manages the search word and the selected piece image information generated in the process of producing one story for determining the similarity between the selected front and back piece images. The search management unit 1244 delivers the search word of the user to the search engine and receives the search result from the search engine. The search term may be composed of a query describing the image so as to search for an image necessary for producing a new image. The recommendation engine 1246 identifies a sculptural image to be recommended to the user among the retrieved sculptural images.

FIG. 3 is a structure diagram of a combined image generated by combining a story created by a user, a searched piece image, and a recommended piece image according to an embodiment of the present invention.

Referring to FIG. 3, a user creates a story for video production. The user creation story 301 is composed of queries (Q1, Q2, ..., Qn) depicting the images so as to be able to retrieve the images required for the new image production. When the search engine searches the fragment images according to the query created by the user and provides the fragment image list 302, the recommendation engine determines the order of the fragment images in the retrieved fragment image list 302 and recommends the fragment images in order. The sculpture image recommendation process of the recommendation engine will be described later with reference to Figs. 4A and 4B.

Among the recommended pieces of sculptural images, the user selects the sculptural images desired by him / herself and constructs the combined image 303. For example, m pieces of sculptured images (R n - 11 , R n -12 , ..., R n -1 m ) 305 ranked for the n-1th query Q n -1 304 of the user- And the piece image selected by the user among them is the ( n-1 ) th piece (S n-1 ) 306 of the combined image.

4A and 4B are flowcharts illustrating a sculptural image recommendation method according to an exemplary embodiment of the present invention.

Referring to FIG. 4A, the image recommendation apparatus according to an exemplary embodiment checks whether or not the retrieved piece image list (R n1 , ..., R nm ) 400 matches a search that the user intends to search for sculpture image recommendation . First, in step 401, it is determined whether the user has already selected a piece image in the retrieved piece image list (R n1 , ..., R nm ) 400. If there is a piece image already selected, it is removed from the recommendation list (402), thereby preventing the same piece image from being repeatedly used in one combined image.

If there is no piece image selected by the user, the accuracy is calculated using the pre-distance of the object (SPO) included in the user's query and the object (SPO) of the retrieved piece image (403). (Step 404). If all the pieces of video are processed (step 404), the order of the pieces of video is determined according to the accuracy, (405) to generate a classified piece image list (406).

Referring to FIG. 4B, the image recommendation apparatus recommends a piece image having the highest accuracy in the classified image list 406 to the user (407). At this time, it is determined whether or not there is a piece image having the same accuracy (Step 408). If there is the same piece image, it is confirmed whether the image Si previously selected by the user and the retrieved piece image Rni are pieces of the same image file ). As a result of checking, the corresponding piece image is recommended (410). In contrast, if the fragment is not a fragment derived from the same image, the similarity is calculated (411) using an attribute learning unit or the like of the image, and then the fragmented image having the image attribute most similar to the previously selected image is selected and recommended ).

The embodiments of the present invention have been described above. It will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims. Therefore, the disclosed embodiments should be considered in an illustrative rather than a restrictive sense. The scope of the present invention is defined by the appended claims rather than by the foregoing description, and all differences within the scope of equivalents thereof should be construed as being included in the present invention.

1: video recommendation apparatus 10:
12: post-processor 14:
100: Analysis module 102: Management module
120: user application module 122: search module
124: Recommendation module 126: Playback module
1240: User information management unit 1242:
1244: Search management part 1246: Recommendation engine

Claims (1)

A user application module for receiving a piece image request necessary for producing a combined image from a user together with a story;
A retrieval module for retrieving pieces of images related to an inputted story; And
A recommendation module for recommending pieces of images to the user in consideration of the accuracy of the retrieved piece images, the degree of association between the forward and backward images and the intention of the user among the pieces of image retrieved through the retrieval module;
And a motion estimation unit for estimating a motion of the image.
KR1020150178477A 2015-12-14 2015-12-14 Apparatus and method for recommending segment image for minimizing heterogeneity KR20170070669A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
KR1020150178477A KR20170070669A (en) 2015-12-14 2015-12-14 Apparatus and method for recommending segment image for minimizing heterogeneity

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
KR1020150178477A KR20170070669A (en) 2015-12-14 2015-12-14 Apparatus and method for recommending segment image for minimizing heterogeneity

Publications (1)

Publication Number Publication Date
KR20170070669A true KR20170070669A (en) 2017-06-22

Family

ID=59282924

Family Applications (1)

Application Number Title Priority Date Filing Date
KR1020150178477A KR20170070669A (en) 2015-12-14 2015-12-14 Apparatus and method for recommending segment image for minimizing heterogeneity

Country Status (1)

Country Link
KR (1) KR20170070669A (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20200044435A (en) * 2018-10-19 2020-04-29 인하대학교 산학협력단 Customized image recommendation system using shot classification of images

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR20200044435A (en) * 2018-10-19 2020-04-29 인하대학교 산학협력단 Customized image recommendation system using shot classification of images

Similar Documents

Publication Publication Date Title
US11354356B1 (en) Video segments for a video related to a task
US11709901B2 (en) Personalized search filter and notification system
US8370358B2 (en) Tagging content with metadata pre-filtered by context
US10585927B1 (en) Determining a set of steps responsive to a how-to query
US20120246302A1 (en) System and methodology for creating and using contextual user profiles
US20110072047A1 (en) Interest Learning from an Image Collection for Advertising
KR20170093713A (en) Method and device for mobile searching based on artificial intelligence
US20230229718A1 (en) Shared User Driven Clipping of Multiple Web Pages
JP2011175362A (en) Information processing apparatus, importance level calculation method, and program
KR20160107187A (en) Coherent question answering in search results
CN103164449A (en) Search result showing method and search result showing device
US11609942B2 (en) Expanding search engine capabilities using AI model recommendations
JP2014505945A (en) Providing search information
WO2014107193A1 (en) Efficiently identifying images, videos, songs or documents most relevant to the user based on attribute feedback
Garcia del Molino et al. Phd-gifs: personalized highlight detection for automatic gif creation
US20170193531A1 (en) Intelligent Digital Media Content Creator Influence Assessment
US20190082236A1 (en) Determining Representative Content to be Used in Representing a Video
CN106708886B (en) Display method and device for in-site search words
JP2015005174A (en) Content retrieval system, method, and program
CN112328889A (en) Method and device for determining recommended search terms, readable medium and electronic equipment
US20200073925A1 (en) Method and system for generating a website from collected content
KR102164409B1 (en) Crowdsourcing user-provided identifiers and associating them with brand identities
US9213745B1 (en) Methods, systems, and media for ranking content items using topics
US10521461B2 (en) System and method for augmenting a search query
US20130262970A1 (en) Identifying picture files of a picture file storage system having relevance to a first file