CN106777030A - Information-pushing method and device - Google Patents
Information-pushing method and device Download PDFInfo
- Publication number
- CN106777030A CN106777030A CN201611124599.5A CN201611124599A CN106777030A CN 106777030 A CN106777030 A CN 106777030A CN 201611124599 A CN201611124599 A CN 201611124599A CN 106777030 A CN106777030 A CN 106777030A
- Authority
- CN
- China
- Prior art keywords
- person
- taken
- image
- information
- similarity
- Prior art date
- Legal status (The legal status 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 status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- 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—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/01—Social networking
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/30—Scenes; Scene-specific elements in albums, collections or shared content, e.g. social network photos or video
Abstract
The disclosure is directed to a kind of information-pushing method and device, described information recommendation method includes：First object image is obtained, the first object image includes the image of first person of being taken；Image recognition is carried out by the first object image, it is determined that second person of being taken in addition to first person of being taken；Obtain the corresponding destinations traffic information of second person of being taken；According to the destinations traffic information, second person of being taken is recommended to first person of being taken.The disclosure carries out information pushing push by image recognition, can improve the accuracy and specific aim of pushed information；For publisher provides more social possible, Consumer's Experience is improved.
Description
Technical field
This disclosure relates to communication technical field, more particularly to a kind of information-pushing method and device.
Background technology
With becoming increasingly popular for the intelligent communications terminals such as mobile phone, panel computer, PC, increasing people passes through society
Network application is handed over to be linked up and exchanged.Meanwhile, the number of users with social networking application constantly increases, and user is in social network
Suitable good friend is found in network becomes more and more difficult.
In correlation technique, social network sites or system can be according to the log-on message of user, affiliated group, user present positions etc.
Information, good friend's push is carried out to user.For example, the alumnus for belonging to an educational institution is pushed to user, or according to user's
Selection good friend's querying condition, is the modes such as its matching good friend and push.
The content of the invention
To overcome problem present in correlation technique, the disclosure to provide a kind of information-pushing method and device.
According to the first aspect of the embodiment of the present disclosure, there is provided a kind of information-pushing method, including：
First object image is obtained, the first object image includes the image of first person of being taken；
Image recognition is carried out by the first object image, it is determined that in addition to first person of being taken second is clapped
The person of taking the photograph；
Obtain the corresponding destinations traffic information of second person of being taken；
According to the destinations traffic information, second person of being taken is recommended to first person of being taken.
Alternatively, it is described to carry out image recognition by the first object image, it is determined that removing first person of being taken
The second outer person of being taken, including：
Obtain first character features of first person of being taken；
Obtain second character features of second person of being taken；
Determine that similarity is higher than predetermined similarity and first character features between in second character features
Candidate's character features；
The corresponding person of being taken of candidate's character features is defined as second person of being taken.
Alternatively, first character features or second character features indicate it is following at least one：Age, sex,
Interest.
Alternatively, the corresponding destinations traffic information of acquisition second person of being taken, including：
Obtain third party's thing feature of second person of being taken；
Communications database is obtained, the communications database includes the corresponding relation between character features and the communication information；
Destinations traffic information corresponding with third party's thing characteristic information is obtained according to the communications database；
Wherein, third party's thing feature indicate it is following at least one：Facial characteristics, resemblance.
Alternatively, methods described also includes：
According to the 4th character features of second person of being taken, acquisition includes second target of second person of being taken
Image；
To the first object image and second target image, image similarity calculating is carried out；
According to image similarity, information pushing is carried out to first person of being taken and second person of being taken, wherein,
First person of being taken is the publisher of the first object image, and second person of being taken is second target image
Publisher.
Alternatively, to the first object image and second target image, the step of carrying out image similarity and calculate
Including：
Background separation is carried out to the first object image and second target image respectively, to obtain the first background area
Domain and the second background area；
When the image similarity of first background area and second background area meets pre-conditioned, institute is obtained
State the second target area in the first object region and second target image in first object image, the first object
Region and second target area include first person of being taken and second person of being taken；
According to the first object region and the image similarity of second target area, the first object figure is determined
The image similarity of picture and second target image.
Alternatively, it is described according to the destinations traffic information, recommend described second to be taken to first person of being taken
Person, including：
Good friend's addition request is sent to the corresponding terminal of second person of being taken according to the destinations traffic information.
A kind of second aspect, there is provided information push-delivery apparatus, including：
Target image acquisition module, is configured as obtaining first object image, and the first object image includes the first quilt
The image of photographer；
The person's of being taken determining module, is configured to carry out image recognition to the first object image, it is determined that removing institute
State second person of being taken outside first person of being taken；
Communication information acquisition module, is configured as obtaining the corresponding destinations traffic information of second person of being taken；
Recommending module, is configured as according to the destinations traffic information, and described second is recommended to first person of being taken
The person of being taken.
Alternatively, the person's of being taken determining module includes：
First acquisition submodule, is configured as obtaining first character features of first person of being taken；
Second acquisition submodule, is configured as obtaining second character features of second person of being taken；
Similarity determination sub-module, be configured as determining in second character features with first character features it
Between similarity higher than predetermined similarity candidate's character features；
Determination sub-module, is configured as the corresponding person of being taken of candidate's character features being defined as described second being clapped
The person of taking the photograph.
Alternatively, the communication information acquisition module includes：
3rd acquisition submodule, is configured as obtaining third party's thing feature of second person of being taken；
Communications database acquisition submodule, is configured as obtaining communications database, and the communications database includes that personage is special
Levy the corresponding relation between the communication information；
Destinations traffic acquisition of information submodule, is configured as obtaining special with third party's thing according to the communications database
Reference ceases corresponding destinations traffic information；
Wherein, third party's thing feature indicate it is following at least one：Facial characteristics, resemblance.
Alternatively, described device also includes：
Second target image acquisition module, is configured as the 4th character features according to second person of being taken, and obtains
The second target image including second person of being taken；
Image similarity acquisition module, is configured as, to the first object image and second target image, carrying out
Image similarity is calculated；
Info push module, is configured as according to image similarity, is clapped to first person of being taken and described second
The person of taking the photograph carries out information pushing, wherein, first person of being taken is the publisher of the first object image, and described second is clapped
The person of taking the photograph is the publisher of second target image.
Alternatively, image similarity acquisition module includes：
Background separation submodule, is configured to carry on the back the first object image and second target image
Scape is separated, to obtain the first background area and the second background area；
Target area acquisition submodule, is configured as the image of first background area and second background area
When similarity meets pre-conditioned, in the first object region and second target image in the acquisition first object image
The second target area, the first object region and second target area include first person of being taken and described
Second person of being taken；
Image similarity determination sub-module, is configured as according to the first object region and second target area
Image similarity, determines the image similarity of the first object image and second target image.
A kind of third aspect, there is provided information push-delivery apparatus, including：
Processor；
Memory for storing processor-executable instruction；
Wherein, the processor is configured as：
First object image is obtained, the first object image includes the image of first person of being taken；By to described
One target image carries out image recognition, it is determined that second person of being taken in addition to first person of being taken；Obtain second quilt
The corresponding destinations traffic information of photographer；According to the destinations traffic information, described second is recommended to first person of being taken
The person of being taken.
The technical scheme provided by this disclosed embodiment can include the following benefits：Row information is entered by image recognition
Push, the accuracy and specific aim of pushed information can be improved；For publisher provides more social possible, Consumer's Experience is improved,
The life of abundant publisher；The information for appearing in the stranger in oneself image can be obtained, there is provided more social activity chances.
It should be appreciated that the general description of the above and detailed description hereinafter are only exemplary and explanatory, not
The disclosure can be limited.
Brief description of the drawings
Accompanying drawing herein is merged in specification and constitutes the part of this specification, shows the implementation for meeting the disclosure
Example, and it is used to explain the principle of the disclosure together with specification.
Fig. 1 is the configuration diagram of the social networks according to an exemplary embodiment.
Fig. 2 is the schematic flow sheet of the information-pushing method according to an exemplary embodiment.
Fig. 3 is the flow of second person of being taken of determination according to an exemplary embodiment in addition to first person of being taken
Schematic diagram.
Fig. 4 is the schematic flow sheet of the information-pushing method according to another exemplary embodiment.
Fig. 5 is the schematic flow sheet of the information-pushing method according to further example embodiment.
Fig. 6 be according to an exemplary embodiment to carry out image to first object image and the second target image similar
Spend the schematic flow sheet for calculating.
Fig. 7 a are the first image schematic diagrames of the first publisher issue of an exemplary embodiment.
Fig. 7 b are the second image schematic diagrames of the second publisher issue of an exemplary embodiment.
Fig. 8 is the second image schematic diagram of the second publisher issue of another exemplary embodiment.
Fig. 9 is the schematic flow sheet of the information-pushing method according to another exemplary embodiment.
Figure 10 a- Figure 10 d are the information pushing effect diagrams obtained according to an exemplary embodiment.
Figure 11 is a kind of block diagram of the information push-delivery apparatus according to an exemplary embodiment.
Figure 12 is a kind of block diagram of the device for information-pushing method according to an exemplary embodiment.
Specific embodiment
Here exemplary embodiment will be illustrated in detail, its example is illustrated in the accompanying drawings.Following description is related to
During accompanying drawing, unless otherwise indicated, the same numbers in different accompanying drawings represent same or analogous key element.Following exemplary embodiment
Described in implementation method do not represent all implementation methods consistent with the disclosure.Conversely, they be only with it is such as appended
The example of the consistent apparatus and method of some aspects described in detail in claims, the disclosure.
It is the configuration diagram of the social networks according to an exemplary embodiment referring to Fig. 1.Social networks 100 is wrapped
Include：One or more user equipmenies 110, social networking system 120 and network 130.
User equipment 110 can receive the input from publisher and can send and receive data via network 130.User
Equipment 110 can be the equipment such as smart mobile phone, panel computer, personal digital assistant, mobile phone.In certain embodiments, use
The programs such as the executable browser application interacted with social networking system 120 of family equipment 110.In further embodiments,
User equipment 110 can be by the operation on native operating sys-tern (for example, android operating systems or ios operating systems etc.)
Application program (APP) is interacted with social networking system 120.User equipment 110 can be used the mode for wiredly and/or wirelessly communicating,
Communicated with social networking system 120 via network 130.
Network 130 may include any combination of LAN and/or wide area network.In one embodiment, network 130 may include
Use such as Ethernet, 802.11,3G, 4G, 5G, GSM, digital subscriber line (DSL), micro-wave access global inter communication (WiMAX)
Etc. the link of technology.
During social networking system 120 includes the multiple users of storage and enables user in social networks and social networks
Other users communicate and interaction social networks one or more computing devices (for example, server).
By user equipment 110 and the communication interaction of social networking system 120, publisher can be by user profile, image (example
Such as, photo or any kind of accompanying drawing) and individualized content etc. be published to social networking system 120, other users can pass through
The content for accessing 120 pairs of issues of social networking system is browsed.
The embodiment of the present disclosure is published to the image of social networking system 120 using publisher, carries out image recognition, thus,
Information pushing is carried out, the accuracy and specific aim of information pushing can be improved.
It is the schematic flow sheet of the information-pushing method according to an exemplary embodiment referring to Fig. 2.The method for pushing
Including：
In step 201, first object image is obtained, first object image includes the image of first person of being taken.
In an embodiment of the disclosure, first object image can be the image that a publisher is published in social networks,
Or the image collected by image collecting device (for example, camera).One first quilt is included in first object image
Photographer.
In step 202., image recognition is carried out by first object image, it is determined that in addition to first person of being taken second
The person of being taken.
Referring to Fig. 3, in an embodiment of the disclosure, second in addition to first person of being taken can be in the following manner determined
The person of being taken：
In step 301, first character features of first person of being taken are obtained.
In step 302, second character features of second person of being taken are obtained.
In one embodiment, recognition of face can be carried out to first person of being taken for identifying and second person of being taken, is obtained
Obtain the information of first person of being taken and second person of being taken.Thus, can be obtained in the registration information database of social networking system
Take first character features of first person of being taken, and second person of being taken the second character features.
In one embodiment, the first character features or the second character features indicate it is following at least one：Age, sex,
Interest.
In step 303, determine in the second character features similar higher than predetermined to similarity between the first character features
Candidate's character features of degree.
In step 304, the corresponding person of being taken of candidate's character features is defined as second person of being taken.
In embodiment of the disclosure, in order to improve the specific aim of information recommendation, determine and the in the second character features
Similarity is higher than candidate's character features of predetermined similarity between one character features, and determines that second is taken according to similarity
Person.
In step 203, the corresponding destinations traffic information of second person of being taken is obtained.
In an embodiment of the disclosure, the destinations traffic information of second person of being taken is obtained in the following manner：Obtain
Third party's thing feature of second person of being taken；Obtain communications database, communications database include character features and the communication information it
Between corresponding relation；Destinations traffic information corresponding with third party's thing characteristic information is obtained according to communications database；Wherein, the 3rd
Character features indicate it is following at least one：Facial characteristics, resemblance.
Destinations traffic information can be account information, contact method etc..Destinations traffic information can be according to second person of being taken
Third party's thing feature, obtained from social networking system.
In step 204, according to destinations traffic information, second person of being taken is recommended to first person of being taken.
After getting the destinations traffic information of second person of being taken, second person of being taken can be recommended to first person of being taken.
In one embodiment, good friend's addition request is sent to the corresponding terminal of second person of being taken according to destinations traffic information.Thus,
The information-pushing method that the disclosure is implemented, the person of being taken in image recognition carries out information pushing, can improve pushed information
Accuracy and specific aim；Provide the user more social possible, raising Consumer's Experience.
It is the schematic flow sheet of the information-pushing method according to another exemplary embodiment referring to Fig. 4.The push side
Method includes：
In step 401, the image of one or more publishers issue is obtained.
In embodiment of the disclosure, publisher is published to social networking system 120, will be stored in social networking system
In 120 storage region, and it is pushed to booking reader or is pushed to default website etc..
The image of publisher's issue can be to shoot the image for obtaining by capture apparatus (for example, mobile terminal, camera etc.),
Or by way of Wireless/wired transmission, the image got from webserver download or other user equipmenies.
In some embodiments of the present disclosure, image can be video image, picture etc..
In an embodiment of the disclosure, the time that can be issued with image is according to the acquisition for carrying out image.For example, to pre-
If the time period is (for example, 07:00 to 21:00 this time period) in image obtained.In further embodiments, can also set
Set the goal crowd, target location or keyword etc., so as to target group can be obtained (for example, the VIP registration meetings of certain social networking application
Member) issue image, positioned at target location (for example, sight spot A) place issue image and setting keyword (for example, the performance of M
Meeting) associated image etc..
It should be understood that in embodiment of the disclosure, user is when image is issued, it will usually adds text information and is retouched
State, thus, the keyword message of setting can be obtained from text information, with the associated image of the keyword for obtaining and setting.
For example, obtaining all images comprising " concert of M " in text information.
In step 402, according to the image for obtaining, image similarity calculating is carried out.
In embodiment of the disclosure, image similarity is counted primarily with respect to the similarity degree of the picture material for obtaining
Calculate, draw an image similarity value, the image similarity value is higher, illustrates that the content of this two width picture is more similar.Image phase
Can be calculated by the visual signature of two width pictures like degree, visual signature is specifically as follows color RGB (Red Green
Blue, three primary colors) feature, textural characteristics and histogram feature and SIFT (Scale-invariant feature
Transform, scale invariant feature conversion) feature etc., due to calculating the image between picture and picture by visual signature
Similarity belongs to well known to a person skilled in the art prior art, so will not be repeated here.
In embodiment of the disclosure, the calculating of image similarity can be carried out using following two modes：
Mode one：Contrast one by one
Select a target image, the image of acquisition contrasted with the target image successively, obtain every image and
The image similarity of the target image.
Mode two：Overall contrast
The image of acquisition is integrally carried out the calculating of image similarity, the similarity of all images is obtained.
In step 403, according to image similarity, information pushing is carried out to one or more publishers.
Situations such as image similarity can embody the common interest of user, social circle, thus, in the embodiment of the present disclosure, leads to
Acquisition image similarity is crossed, enters the push of row information, more accurately information pushing can be provided the user.
In embodiment of the disclosure, information pushing at least includes one of：Friend information is pushed, people information
Push and default relevant information is pushed.
When image similarity calculating is carried out using above-mentioned first way, the image similarity that will can be obtained is arranged
Sequence, and the information of the corresponding publisher of image using image similarity within a preset range is pushed to target figure as friend information
As corresponding publisher.In one embodiment, the personage that image similarity image within a preset range is included can be obtained
Information is simultaneously pushed to the corresponding publisher of target image.
When image similarity calculating is carried out using the above-mentioned second way, can be full in the image similarity value of all images
During sufficient predetermined threshold value, can using the information of remaining publisher in addition to publisher as friend information to the publisher
Pushed.For example, the image of publisher A issues, the image of publisher B issues, the image and publisher D of publisher C issues
The image similarity of issue meets predetermined threshold value, then make the information of the information of publisher B, the information of publisher C and publisher D
Publisher A is pushed to for friend information；Using the information of the information of publisher A, the information of publisher C and publisher D as good
Friendly information is pushed to publisher B；Believe the information of the information of publisher A, the information of publisher B and publisher D as good friend
Breath is pushed to publisher C；Using the information of the information of publisher A, the information of publisher B and publisher C as friend information
It is pushed to publisher D.
In one embodiment, the information of publisher may include at least one of following information：Account information, sex, year
Age, contact method etc..It is stored with social networking system 120 when each publisher is registered, fills in relevant information and is provided
Above- mentioned information.
In one embodiment, people information may include at least one of following information：Account information, sex, age, connection
It is mode.In an embodiment of the disclosure, in order to improve security, the personage that image includes is obtained by reasonable approach
, for example, the user corresponding to related person passes through registration, fills in the information of the offers such as related form.
In one embodiment, default relevant information can be Domestic News information, merchandise news or race information etc..
When relevant information is pushed for publisher, using a publisher as target publisher, and target publisher issue can be recognized
Characteristic information in image, and information matches are carried out according to characteristic information, obtain default relevant information.And be the image of issue
Similarity publisher high pushes the default relevant information of matching.Thus, can be that image similarity publisher's push high is identical
Default relevant information, improve information pushing efficiency.
Thus, the information-pushing method that the disclosure is implemented, information pushing is carried out according to the image similarity that publisher issues,
The accuracy and specific aim of pushed information can be improved；For publisher provides more social possible, Consumer's Experience is improved, enrich hair
The life of cloth person.
It is the schematic flow sheet of the information-pushing method according to further example embodiment referring to Fig. 5.The embodiment
On the basis of the embodiment shown in above-mentioned Fig. 2 and Fig. 4, the 4th character features according to second person of being taken, acquisition includes the
Second target image of two persons of being taken, and when second person of being taken is publisher's (i.e. second issue of the second target image
Person), when first person of being taken is publisher (i.e. first publisher) of first object image, the method for pushing of the embodiment includes
Following steps：
In step 501, to first object image and the second target image, image similarity calculating is carried out.
Referring to Fig. 6, in an embodiment of the disclosure, to first object image and the second target image, image phase is carried out
Comprised the following steps like degree calculating：
In step 601, background separation is carried out to first object image and the second target image respectively, to obtain first back of the body
Scene area and the second background area.
In embodiment of the disclosure, image can be video image or picture.For video image, Surendra can be used
Algorithm extracts background, obtains background area.For picture, background area can be carried out using modes such as image segmentation, edge extractings
Acquisition.
By obtaining the image similarity of image background regions, the situation of the background area of image can be obtained.In reality
In, in identical position, (for example, same sight spot A), identical seat/setting (for example, party, concert, dinner party etc.) etc. make
It is the image that background is obtained, the image similarity of its background area is higher.
In step 602, when the image similarity of the first background area and the second background area meets pre-conditioned, obtain
Take the image similarity of the first object region of first object image and the second target area of the second target image.
In an embodiment of the disclosure, can be set to pre-conditioned：Image similarity is more than 50%.Referring to Fig. 7 a and
Fig. 7 b, in one embodiment, Fig. 7 a are the first object image of the first publisher issue, and Fig. 7 b are the second publisher issue
Second target image.The image similarity of the background area of the two images is to meet pre-conditioned.
In embodiment of the disclosure, when the image similarity of the first background area and the second background area meets default bar
During part, illustrate that first object image and the second target image have similar background, thus, can further deduce, the first issue
Person and the second publisher are interested in identical scene, or identical scene carried out activity, take pictures.
First object region can be the foreground area or people's object area of first object image.The second same target area can
It is the foreground area or people's object area of the second target image.Foreground area can be obtained by reducing background area.People's object area can
Obtained by modes such as feature extraction, rim detections.
In step 502, according to image similarity, information pushing is carried out to first person of being taken and second person of being taken.
In an embodiment of the disclosure, first object region is the area at least including the first personage and the second personage
Domain.Second target area is the region at least including third party's thing and the 4th personage.Carrying out first object region and second
When the image similarity of target area is calculated, it is right respectively to carry out the first personage, the second personage, third party's thing and the 4th personage
Than.
In an embodiment of the disclosure, to the character features of the first personage, the second personage, third party's thing and the 4th personage
Extracted.Character features at least include one of：Hair style, clothing and face characteristic.When being contrasted, can
The character features of the first personage are contrasted with the character features of third party's thing and the character features of the 4th personage respectively, to sentence
Disconnected first personage and third party's thing or the similarity of the 4th personage.And by the character features of the second personage respectively with third party's thing
Character features and the character features of the 4th personage contrasted, to judge the second personage with third party's thing or the phase of the 4th personage
Like degree.
Referring to Fig. 7 a and Fig. 7 b, in one embodiment, Fig. 7 a show the first image of the first publisher issue；Fig. 7 b
It is the second image of the second publisher issue.Wherein, A is the first personage in the first image, and B is the second people in the first image
Thing, A ' is the third party's thing in the second image, and B ' is the 4th personage in the second image.By to the first personage, the second personage,
The character features of third party's thing and the 4th personage are contrasted, and available first personage A is similar to third party's thing A ', the second personage
B is similar to the 4th personage B '.
Thus, contrasted by by the personage of first object image and the second target image, can obtain first object figure
The similarity of the personage of picture and the second target image.When the people-similarity of first object image and the second target image meets pre-
If during condition, the push of row information can be entered.
In embodiment of the disclosure, the push of information at least includes one of：Friend information is pushed, Ren Wuxin
Breath is pushed and default relevant information is pushed.
In one embodiment, can in the following manner carry out confirming that the personage in the image of publisher's issue (is clapped
The person of taking the photograph) and publisher relation：
When publisher is registered in social networking system 120, there is provided the real individual's picture of publisher.Thus, may be used
The publisher's individual's picture provided during by the personage in the image by issue with registration carries out character features extraction and compares, with
It is determined that whether the personage in the image of issue is same people with publisher.
In one embodiment, when the first object region and second target area of the second image of first object image
Image similarity meets pre-conditioned, and the first publisher of first object image is the first personage or third party's thing (is clapped
The person of taking the photograph), when the second publisher of the second image is the second personage or the 4th personage (person of being taken), to the first publisher (i.e.
First person of being taken) information of information including the second publisher (i.e. second person of being taken) that pushes, pushed to the second publisher
Information including the first publisher information.Thus, because the first publisher and the second publisher have appeared in identical scene
In, show that two people may have common hobby (for example, liking identical sight spot A, liking same type of party etc.), Huo Zheke
There can be common acquaintance's (for example, participating in same dinner party), therefore, the information of the second publisher is pushed to the first publisher, give
Second publisher pushes the information of the first publisher.Thus, the specific aim and accuracy of pushed information can be improved so that publisher
Pushed information interested can be got, the possibility be further familiar with, understand, associate between two publishers can be greatly improved
Property.
The information of the first publisher and the information of the second publisher, can obtain from log-on message, or believe from historical record
Obtained in breath (for example, authentication information of publisher's input etc.).
In further embodiments, first object region or the second target area can only include a personage.Referring to Fig. 7 a
And Fig. 8, personage B ' is only included in the second target area in the second target image.Then carrying out first object region and second
When the image similarity of target area is calculated, only to the personage in the second target area respectively by the personage with first object region
Contrasted.Thus, as long as the similarity of any personage in personage in the second target area and first object region
Meet preset requirement, you can carry out information pushing.
In one embodiment, the information-pushing method of the embodiment also includes：Issued to the first publisher and second
When person carries out information pushing, while pushing reason.
Pushing reason may include：Associated picture and/or related text explanation.Referring to Fig. 7 a and Fig. 7 b, in one embodiment
In, when the information of the second publisher is pushed to the first publisher, can be managed the second target image shown in Fig. 7 b as push
By being pushed to the first publisher simultaneously.Can also be by " you once take pictures in identical place, and appear in the camera lens of other side " etc.
Explanatory note is pushed as push reason simultaneously.
Referring to Fig. 9, on the basis of the method for pushing shown in Fig. 6, the information-pushing method of the another embodiment of the disclosure is also
Including：
In step 901, detect that the first personage and the second personage account for the first area ratio of first object image respectively.
In one embodiment, people's object plane of the first personage and the second personage by edge detection algorithm, can respectively be obtained
Product, and it is compared with the gross area of first object image, respectively obtain the ratio of both shared first images.
In step 902, detect that third party's thing and the 4th personage account for the ratio between second area of the second target image respectively.
With above-mentioned steps 901, the ratio of third party's thing and the second target image shared by the 4th personage can be respectively obtained.
In step 903, according to the ratio between the first area ratio and second area, pushed information is determined.
In a kind of scene of embodiment of the disclosure, the second personage is in first object image only as background people
Thing, i.e. the first personage are only leading role, and it is due to accidental, or due to a variety of causes that the second personage is appeared in first object image
In the image that situations such as (for example, people is more, just passing by) stolen shadow shoots to other people, in this case, shared by the second personage
Ratio is by less than the ratio shared by the first personage.And in the second target image, then third party's thing similar to the first personage is
It is that now, the ratio shared by third party's thing will be less than as leading role as the 4th personage as background characters, with the second figure picture
Ratio shared by 4th personage.Thus, it is possible to the second publisher is given by the information pushing of the first personage, and by the letter of the second personage
Breath is pushed to the first publisher so that the first publisher and the second publisher can obtain the personage of " accidental " appearance in the picture
Information.
In an embodiment of the disclosure, in order to improve security, the information of the first personage and the second personage is by closing
What reason approach was obtained, for example, the user corresponding to the first personage (or second personage) is provided by registering, filling in related form etc.
's.
By distinguishing the first personage, the second personage, third party's thing and the area ratio of the 4th personage, it may be determined that the property of personage
Matter, can obtain the information for appearing in the stranger in oneself image, there is provided more social activity chances, can be used for push etc. of dating.
It is the information pushing effect diagram obtained according to an exemplary embodiment referring to Figure 10 a- Figure 10 d.Figure 10 a are hair
Cloth person a is published to image P1, Figure 10 b of a social networking application for publisher b issues a society by user equipment by user equipment
Hand over the image P2 of application.Wherein, image P1 is that publisher a shoots what is obtained in sight spot J, and the A in image P1 is publisher a,
And the B in image P1 be publisher a shooting image P1 when, positioned at the personage that it does not know each other after one's death.The image P1 persons of being published a
Social networking application is published to, in image P1, publisher a is leading role/focus, and is by the b that " imprudence " is taken into image P1
Supporting role/background.
In the image P2 of publisher b issues, publisher b as leading role/focus, and a to be " imprudence " be taken into image
Supporting role/background in P2.
Due to publisher a and publisher b, Same Scene is appeared in the almost identical time, shown that two people may have
Common hobby (for example, liking tourism), therefore, Figure 10 c and Figure 10 d are participated in, according to image P1 and the image similarity of image P2
Acquisition, be capable of achieving give publisher a by the information pushing of publisher b, by the information pushing of publisher a to publisher.So, it is right
For publisher a and publisher b, can produce " be linked by ties of fate " sensation, increased add as a friend and further be familiar with, understand,
The possibility of contacts.
Referring to Figure 10 c and Figure 10 d, when information pushing is carried out, the similar photo of other side is showed as reason is pushed
Publisher, further increases social possibility.
Figure 11 is a kind of block diagram of the information push-delivery apparatus according to an exemplary embodiment.Reference picture 11, the device
1100 include：
Target image acquisition module 1101, is configured as obtaining first object image, and the first object image includes the
The image of one person of being taken；
The person's of being taken determining module 1102, is configured to carry out image recognition to the first object image, it is determined that
Second person of being taken in addition to first person of being taken；
Communication information acquisition module 1103, is configured as obtaining the corresponding destinations traffic information of second person of being taken；
Recommending module 1104, is configured as according to the destinations traffic information, recommends to first person of being taken described
Second person of being taken.
In one embodiment, the person's of being taken determining module 1102 includes：
First acquisition submodule, is configured as obtaining first character features of first person of being taken；
Second acquisition submodule, is configured as obtaining second character features of second person of being taken；
Similarity determination sub-module, be configured as determining in second character features with first character features it
Between similarity higher than predetermined similarity candidate's character features；
Determination sub-module, is configured as the corresponding person of being taken of candidate's character features being defined as described second being clapped
The person of taking the photograph.
In one embodiment, communication information acquisition module 1103 includes：
3rd acquisition submodule, is configured as obtaining third party's thing feature of second person of being taken；
Communications database acquisition submodule, is configured as obtaining communications database, and the communications database includes that personage is special
Levy the corresponding relation between the communication information；
Destinations traffic acquisition of information submodule, is configured as obtaining special with third party's thing according to the communications database
Reference ceases corresponding destinations traffic information；
Wherein, third party's thing feature indicate it is following at least one：Facial characteristics, resemblance.
In one embodiment, device 1100 also includes：
Second target image acquisition module 1105, is configured as the 4th character features according to second person of being taken,
Acquisition includes second target image of second person of being taken；
Image similarity acquisition module 1106, is configured as to the first object image and second target image,
Carry out image similarity calculating；
Info push module 1107, is configured as according to image similarity, to first person of being taken and described second
The person of being taken carries out information pushing, wherein, first person of being taken is the publisher of the first object image, described second
The person of being taken is the publisher of second target image.
In one embodiment, image similarity acquisition module 1106 includes：
Background separation submodule, is configured to carry on the back the first object image and second target image
Scape is separated, to obtain the first background area and the second background area；
Target area acquisition submodule, is configured as the image of first background area and second background area
When similarity meets pre-conditioned, in the first object region and second target image in the acquisition first object image
The second target area, the first object region and second target area include first person of being taken and described
Second person of being taken；
Image similarity determination sub-module, is configured as according to the first object region and second target area
Image similarity, determines the image similarity of the first object image and second target image.
On the device in above-described embodiment, wherein modules perform the concrete mode of operation in relevant the method
Embodiment in be described in detail, explanation will be not set forth in detail herein.
Figure 12 is a kind of block diagram of the device 1200 for information-pushing method according to an exemplary embodiment.Example
Such as, device 1200 may be provided in a server.Reference picture 12, device 1200 includes processing assembly 1201, its further bag
One or more processors, and the memory resource as representated by memory 1202 are included, can be by processing assembly for storing
The instruction of 1201 execution, such as application program.In memory 1202 store application program can include one or one with
On each correspond to the module of one group of instruction.Additionally, processing assembly 1201 is configured as execute instruction, to perform above-mentioned business
Product method for pushing
Device 1200 can also include that a power supply module 1203 is configured as the power management of performs device 1200, one
Wired or wireless network interface 1204 is configured as device 1200 being connected to network, and input and output (I/O) interface
1205.Device 1200 can operate the operating system in memory 1202, such as Windows ServerTM, Mac based on storage
OS XTM, UnixTM, LinuxTM, FreeBSDTM or similar.
In the exemplary embodiment, a kind of non-transitorycomputer readable storage medium including instructing, example are additionally provided
Such as include the memory 1202 of instruction, above-mentioned instruction can be performed by the processing assembly 1201 of device 1200 and pushed away to complete above- mentioned information
Delivery method.For example, the non-transitorycomputer readable storage medium can be ROM, random access memory (RAM), CD-
ROM, tape, floppy disk and optical data storage devices etc..
In the exemplary embodiment, device 1200 can be by one or more application specific integrated circuits (ASIC), numeral
Signal processor (DSP), digital signal processing appts (DSPD), PLD (PLD), field programmable gate array
(FPGA), controller, microcontroller, microprocessor or other electronic components realization, for performing above- mentioned information method for pushing.
In the exemplary embodiment, a kind of non-transitorycomputer readable storage medium including instructing, example are additionally provided
Such as include the memory 1202 of instruction, above-mentioned instruction can be performed by the processing assembly 1202 of device 1200 and pushed away to complete above- mentioned information
Delivery method.For example, the non-transitorycomputer readable storage medium can be ROM, random access memory (RAM), CD-
ROM, tape, floppy disk and optical data storage devices etc..
Those skilled in the art will readily occur to other embodiment party of the disclosure after considering specification and putting into practice the disclosure
Case.The application is intended to any modification, purposes or the adaptations of the disclosure, these modifications, purposes or adaptability
Change follows the general principle of the disclosure and including the undocumented common knowledge or usual skill in the art of the disclosure
Art means.Description and embodiments are considered only as exemplary, and the true scope of the disclosure and spirit are by following claim
Point out.
It should be appreciated that the disclosure is not limited to the precision architecture for being described above and being shown in the drawings, and
And can without departing from the scope carry out various modifications and changes.The scope of the present disclosure is only limited by appended claim.
Claims (13)
1. a kind of information-pushing method, it is characterised in that including：
First object image is obtained, the first object image includes the image of first person of being taken；
Image recognition is carried out by the first object image, it is determined that in addition to first person of being taken second is taken
Person；
Obtain the corresponding destinations traffic information of second person of being taken；
According to the destinations traffic information, second person of being taken is recommended to first person of being taken.
2. method according to claim 1, it is characterised in that described to carry out image knowledge by the first object image
Not, it is determined that second person of being taken in addition to first person of being taken, including：
Obtain first character features of first person of being taken；
Obtain second character features of second person of being taken；
Candidate of the similarity higher than predetermined similarity and first character features between is determined in second character features
Character features；
The corresponding person of being taken of candidate's character features is defined as second person of being taken.
3. method according to claim 2, it is characterised in that first character features or second character features refer to
Show it is following at least one：Age, sex, interest.
4. method according to claim 1, it is characterised in that the corresponding target of acquisition second person of being taken is led to
Letter information, including：
Obtain third party's thing feature of second person of being taken；
Communications database is obtained, the communications database includes the corresponding relation between character features and the communication information；
Destinations traffic information corresponding with third party's thing characteristic information is obtained according to the communications database；
Wherein, third party's thing feature indicate it is following at least one：Facial characteristics, resemblance.
5. method according to claim 1, it is characterised in that methods described also includes：
According to the 4th character features of second person of being taken, acquisition includes the second target figure of second person of being taken
Picture；
To the first object image and second target image, image similarity calculating is carried out；
According to image similarity, information pushing is carried out to first person of being taken and second person of being taken, wherein, it is described
First person of being taken is the publisher of the first object image, and second person of being taken is the hair of second target image
Cloth person.
6. method according to claim 5, it is characterised in that to the first object image and the second target figure
Picture, carrying out the step of image similarity is calculated includes：
Background separation is carried out to the first object image and second target image respectively, with obtain the first background area and
Second background area；
When the image similarity of first background area and second background area meets pre-conditioned, described the is obtained
The second target area in first object region and second target image in one target image, the first object region
Include first person of being taken and second person of being taken with second target area；
According to the first object region and the image similarity of second target area, determine the first object image and
The image similarity of second target image.
7. method according to claim 1, it is characterised in that described according to the destinations traffic information, to described first
The person of being taken recommends second person of being taken, including：
Good friend's addition request is sent to the corresponding terminal of second person of being taken according to the destinations traffic information.
8. a kind of information push-delivery apparatus, it is characterised in that including：
Target image acquisition module, is configured as obtaining first object image, and the first object image is taken including first
The image of person；
The person's of being taken determining module, is configured to carry out image recognition to the first object image, it is determined that except described the
Second person of being taken outside one person of being taken；
Communication information acquisition module, is configured as obtaining the corresponding destinations traffic information of second person of being taken；
Recommending module, is configured as according to the destinations traffic information, recommends described second to be clapped to first person of being taken
The person of taking the photograph.
9. device according to claim 8, it is characterised in that the person's of being taken determining module includes：
First acquisition submodule, is configured as obtaining first character features of first person of being taken；
Second acquisition submodule, is configured as obtaining second character features of second person of being taken；
Similarity determination sub-module, is configured as determining the phase between first character features in second character features
Candidate's character features like degree higher than predetermined similarity；
Determination sub-module, is configured as the corresponding person of being taken of candidate's character features being defined as described second being taken
Person.
10. device according to claim 9, it is characterised in that the communication information acquisition module includes：
3rd acquisition submodule, is configured as obtaining third party's thing feature of second person of being taken；
Communications database acquisition submodule, be configured as obtain communications database, the communications database include character features with
Corresponding relation between the communication information；
Destinations traffic acquisition of information submodule, is configured as being obtained according to the communications database and believes with third party's thing feature
Cease corresponding destinations traffic information；
Wherein, third party's thing feature indicate it is following at least one：Facial characteristics, resemblance.
11. devices according to claim 8, it is characterised in that described device also includes：
Second target image acquisition module, is configured as the 4th character features according to second person of being taken, and acquisition includes
Second target image of second person of being taken；
Image similarity acquisition module, is configured as, to the first object image and second target image, carrying out image
Similarity Measure；
Info push module, is configured as according to image similarity, to first person of being taken and second person of being taken
Information pushing is carried out, wherein, first person of being taken is the publisher of the first object image, second person of being taken
It is the publisher of second target image.
12. devices according to claim 11, it is characterised in that image similarity acquisition module includes：
Background separation submodule, is configured to carry out the first object image and second target image background point
From to obtain the first background area and the second background area；
Target area acquisition submodule, is configured as first background area similar with the image of second background area
Degree is when meeting pre-conditioned, obtains the in the first object region and second target image in the first object image
Two target areas, the first object region and second target area include first person of being taken and described second
The person of being taken；
Image similarity determination sub-module, is configured as the image according to the first object region and second target area
Similarity, determines the image similarity of the first object image and second target image.
A kind of 13. information push-delivery apparatus, it is characterised in that including：
Processor；
Memory for storing processor-executable instruction；
Wherein, the processor is configured as：
First object image is obtained, the first object image includes the image of first person of being taken；By to first mesh
Logo image carries out image recognition, it is determined that second person of being taken in addition to first person of being taken；Described second is obtained to be taken
The corresponding destinations traffic information of person；According to the destinations traffic information, described second is recommended to be clapped to first person of being taken
The person of taking the photograph.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611124599.5A CN106777030B (en) | 2016-12-08 | 2016-12-08 | Information pushing method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611124599.5A CN106777030B (en) | 2016-12-08 | 2016-12-08 | Information pushing method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106777030A true CN106777030A (en) | 2017-05-31 |
| CN106777030B CN106777030B (en) | 2020-12-25 |
Family
ID=58877467
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611124599.5A Active CN106777030B (en) | 2016-12-08 | 2016-12-08 | Information pushing method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106777030B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108073724A (en) * | 2018-01-04 | 2018-05-25 | 百度在线网络技术（北京）有限公司 | For generating the method and apparatus of information |
| CN110061904A (en) * | 2019-04-10 | 2019-07-26 | 北京小米智能科技有限公司 | Obtain method, apparatus, system, electronic equipment and the storage medium of social information |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1685371A (en) * | 2002-09-24 | 2005-10-19 | 皇家飞利浦电子股份有限公司 | Image recognition |
| CN102355534A (en) * | 2011-11-01 | 2012-02-15 | 宇龙计算机通信科技(深圳)有限公司 | Mobile terminal and contact information recommendation method |
| CN102637183A (en) * | 2011-02-12 | 2012-08-15 | 北京千橡网景科技发展有限公司 | Method and device for recommending friends to user in social network |
| US20130215116A1 (en) * | 2008-03-21 | 2013-08-22 | Dressbot, Inc. | System and Method for Collaborative Shopping, Business and Entertainment |
| CN103324636A (en) * | 2012-03-22 | 2013-09-25 | 三星电子（中国）研发中心 | System and method for recommending friend in social network |
| CN103577537A (en) * | 2013-09-24 | 2014-02-12 | 上海交通大学 | Image sharing website picture-oriented multi-pairing similarity determining method |
| CN104254859A (en) * | 2012-04-26 | 2014-12-31 | 谷歌公司 | Creating social network groups |
| CN105303150A (en) * | 2014-06-26 | 2016-02-03 | 腾讯科技（深圳）有限公司 | Method and system for implementing image processing |
-
2016
- 2016-12-08 CN CN201611124599.5A patent/CN106777030B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1685371A (en) * | 2002-09-24 | 2005-10-19 | 皇家飞利浦电子股份有限公司 | Image recognition |
| US20130215116A1 (en) * | 2008-03-21 | 2013-08-22 | Dressbot, Inc. | System and Method for Collaborative Shopping, Business and Entertainment |
| CN102637183A (en) * | 2011-02-12 | 2012-08-15 | 北京千橡网景科技发展有限公司 | Method and device for recommending friends to user in social network |
| CN102355534A (en) * | 2011-11-01 | 2012-02-15 | 宇龙计算机通信科技(深圳)有限公司 | Mobile terminal and contact information recommendation method |
| CN103324636A (en) * | 2012-03-22 | 2013-09-25 | 三星电子（中国）研发中心 | System and method for recommending friend in social network |
| CN104254859A (en) * | 2012-04-26 | 2014-12-31 | 谷歌公司 | Creating social network groups |
| CN103577537A (en) * | 2013-09-24 | 2014-02-12 | 上海交通大学 | Image sharing website picture-oriented multi-pairing similarity determining method |
| CN105303150A (en) * | 2014-06-26 | 2016-02-03 | 腾讯科技（深圳）有限公司 | Method and system for implementing image processing |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108073724A (en) * | 2018-01-04 | 2018-05-25 | 百度在线网络技术（北京）有限公司 | For generating the method and apparatus of information |
| CN110061904A (en) * | 2019-04-10 | 2019-07-26 | 北京小米智能科技有限公司 | Obtain method, apparatus, system, electronic equipment and the storage medium of social information |
| CN110061904B (en) * | 2019-04-10 | 2022-03-01 | 北京小米智能科技有限公司 | Method, device and system for acquiring social information, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106777030B (en) | 2020-12-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10900772B2 (en) | Apparatus and methods for facial recognition and video analytics to identify individuals in contextual video streams | |
| US11286310B2 (en) | Methods and apparatus for false positive minimization in facial recognition applications | |
| CN102930457B (en) | Method and system for implementing application recommendation based on facial image characteristics | |
| US8650242B2 (en) | Data processing apparatus and data processing method | |
| JP2019530041A (en) | Combining the face of the source image with the target image based on the search query | |
| CN109829418B (en) | Card punching method, device and system based on shadow features | |
| US11501564B2 (en) | Mediating apparatus and method, and computer-readable recording medium thereof | |
| US8843518B2 (en) | Method and apparatus for establishing a connection with known individuals | |
| CN110866236A (en) | Private picture display method, device, terminal and storage medium | |
| CN109753605A (en) | User's recommended method, device, computer equipment and computer readable storage medium | |
| KR20200025291A (en) | A shopping service procedure and shopping service system using personal community device | |
| CN106777030A (en) | Information-pushing method and device | |
| CN109886239B (en) | Portrait clustering method, device and system | |
| KR101715708B1 (en) | Automated System for Providing Relation Related Tag Using Image Analysis and Method Using Same | |
| CN110011901B (en) | Method, apparatus and storage medium for searching friend-making object | |
| CN111429210A (en) | Method, device and equipment for recommending clothes | |
| TWM467133U (en) | Real-time social system combining real-time communication and mobile augmented reality | |
| CN111026975A (en) | Information retrieval method and device based on visual information identification | |
| CN111126179A (en) | Information acquisition method and device, storage medium and electronic device | |
| CN107609195A (en) | One kind searches topic method and device | |
| CN110569758A (en) | method, device and storage medium for determining user category | |
| JP2022511058A (en) | Participant identification in the image | |
| CN113946743A (en) | Method, device and storage medium for determining picture of fan | |
| CN116629954A (en) | Intelligent clothing collocation method, device and storage medium | |
| CN111882404A (en) | Association push processing method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |