CN109495784A - Information-pushing method, device, electronic equipment and computer readable storage medium - Google Patents
Information-pushing method, device, electronic equipment and computer readable storage medium Download PDFInfo
- Publication number
- CN109495784A CN109495784A CN201811447987.6A CN201811447987A CN109495784A CN 109495784 A CN109495784 A CN 109495784A CN 201811447987 A CN201811447987 A CN 201811447987A CN 109495784 A CN109495784 A CN 109495784A
- Authority
- CN
- China
- Prior art keywords
- video frame
- frame images
- target object
- relevant information
- information
- 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.)
- Pending
Links
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]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/44—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream, rendering scenes according to MPEG-4 scene graphs
- H04N21/44008—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream, rendering scenes according to MPEG-4 scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics in the video stream
-
- 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
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/442—Monitoring of processes or resources, e.g. detecting the failure of a recording device, monitoring the downstream bandwidth, the number of times a movie has been viewed, the storage space available from the internal hard disk
- H04N21/44204—Monitoring of content usage, e.g. the number of times a movie has been viewed, copied or the amount which has been watched
-
- 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]
- H04N21/80—Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
- H04N21/81—Monomedia components thereof
- H04N21/812—Monomedia components thereof involving advertisement data
Abstract
Present disclose provides a kind of information-pushing methods, device, electronic equipment and computer readable storage medium, applied to image identification technical field, wherein this method comprises: being identified based on image-recognizing method to the video frame images in video to be processed, to determine at least one target object for including in video frame images, then the corresponding push relevant information of at least one target object is determined, then the corresponding target object of relevant information will be pushed to push in association, to realize automatically configuring for push relevant information relevant to target object in video frame images, furthermore, it is identified respectively based on different video frame images and determines corresponding push relevant information, it realizes and pushes corresponding push relevant information respectively in the different periods that user watches video, thus improve the push relevant information of configuration It is rich, and then the user experience is improved.
Description
Technical field
This disclosure relates to technical field of image processing, specifically, this disclosure relates to a kind of information-pushing method, device,
Electronic equipment and computer readable storage medium.
Background technique
With the development of video technique, due to the videos such as short-sighted frequency, live video provide content information abundant and by
To liking for users, therefore corresponding video platform has also gathered a large amount of user, and how in user's viewing video
Commdity advertisement relevant to the video content that user watches is pushed to user simultaneously, becomes the customer flow for realizing video platform
Commercial value key.
Currently, the Commdity advertisement relevant to the video content of user's viewing pushed to user is added by manual type
, i.e., video content to be pushed is watched by the operation personnel of corresponding video platform in advance, and according to the video content people of viewing
Work configures relevant Commdity advertisement, such as recommended information, the purchase link of the commodity or dependent merchandise occurred in video.However,
According to the method for existing human configuration Commdity advertisement relevant to the video content that user watches, complete related to corresponding video
Commdity advertisement configuration, need to expend longer time and a large amount of manpowers, there is a problem of that allocative efficiency is low, at high cost, this
Outside, in general, operation personnel only configure to the relevant Commdity advertisement of pushing video content, it is therefore, existing
Human configuration and user watch the method for the relevant Commdity advertisement of video content and ask there is also the Commdity advertisement of configuration is more single
Topic.
Summary of the invention
Present disclose provides a kind of information-pushing method, device, electronic equipment and computer readable storage mediums, for real
The relevant Commdity advertisement of network video content now watched to user automatically configures, and promotes the more of the Commdity advertisement configured
Sample, the disclosure the technical solution adopted is as follows:
In a first aspect, provide a kind of information-pushing method based on image recognition, this method includes,
The video frame images in video to be processed are identified based on image-recognizing method, to determine in video frame images
Including at least one target object；
Determine the corresponding push relevant information of at least one target object；
The corresponding target object of relevant information will be pushed to push in association.
Second aspect provides a kind of information push-delivery apparatus based on image recognition, which includes,
First determining module, for being identified based on image-recognizing method to the video frame images in video to be processed,
To determine at least one target object for including in video frame images；
Second determining module, the corresponding push of at least one target object determined for determining the first determining module
Relevant information；
Pushing module, for the corresponding target object of the determining push relevant information of the second determining module to be associated
Ground is pushed.
The third aspect provides a kind of electronic equipment, which includes:
One or more processors；
Memory；
One or more application program, wherein one or more application programs be stored in memory and be configured as by
One or more processors execute, and one or more programs are configured to: executing shown in first aspect based on image recognition
Information-pushing method.
Fourth aspect, provides a kind of computer readable storage medium, and computer storage medium refers to for storing computer
It enables, when run on a computer, the information that computer is executed shown in first aspect based on image recognition pushes away
Delivery method.
It is and existing present disclose provides a kind of information-pushing method, device, electronic equipment and computer readable storage medium
Technology is configured compared with the relevant Commdity advertisement of the video content that user watches by manual type, and the disclosure is by being based on image
Recognition methods identifies the video frame images in video to be processed, to determine at least one mesh for including in video frame images
Object is marked, then determines the corresponding push relevant information of at least one target object, will then push relevant information and its
Corresponding target object is pushed in association, i.e., at least one mesh in video frame images is determined by image-recognizing method
Mark object, and will push relevant information corresponding with each target object with corresponding target object is associated pushes,
To realize automatically configuring for push relevant information relevant to target object in video frame images, in addition, based on different
Video frame images identify respectively determines corresponding push relevant information, realizes in the different periods that user watches video respectively
The corresponding push relevant information of push, to improve the rich of the push relevant information of configuration, and then improves user's body
It tests.
The additional aspect of the disclosure and advantage will be set forth in part in the description, these will become from the following description
It obtains obviously, or recognized by the practice of the disclosure.
Detailed description of the invention
The disclosure is above-mentioned and/or additional aspect and advantage will become from the following description of the accompanying drawings of embodiments
Obviously and it is readily appreciated that, in which:
Fig. 1 is a kind of flow diagram of information-pushing method based on image recognition of the embodiment of the present disclosure；
Fig. 2 is a kind of structural schematic diagram of information push-delivery apparatus based on image recognition of the embodiment of the present disclosure；
Fig. 3 is the structural schematic diagram of another information push-delivery apparatus based on image recognition of the embodiment of the present disclosure；
Fig. 4 is the structural schematic diagram of a kind of electronic equipment of the embodiment of the present disclosure.
Specific embodiment
Embodiment of the disclosure is described below in detail, the example of each embodiment is shown in the accompanying drawings, wherein phase from beginning to end
Same or similar label indicates same or similar element or element with the same or similar functions.Below with reference to attached drawing
The embodiment of description is exemplary, and is only used for explaining the disclosure, and cannot be construed to the limitation to the disclosure.
Those skilled in the art of the present technique are appreciated that unless expressly stated, singular " one " used herein, "one"
It may also comprise plural form with "the".It is to be further understood that wording " comprising " used in the specification of the disclosure is
Refer to existing characteristics, integer, step, operation, element and/or component, but it is not excluded that in the presence of or addition it is one or more other
Feature, integer, step, operation, element, component and/or their group.Wording "and/or" used herein is including one or more
Multiple associated wholes for listing item or any cell and all combination.
To keep the purposes, technical schemes and advantages of the disclosure clearer, below in conjunction with attached drawing to disclosure embodiment party
Formula is described in further detail.
How the technical solution of the disclosure and the technical solution of the disclosure are solved with specifically embodiment below above-mentioned
Technical problem is described in detail.These specific embodiments can be combined with each other below, for the same or similar concept
Or process may repeat no more in certain embodiments.Below in conjunction with attached drawing, embodiment of the disclosure is described.
The executing subject server of the embodiment of the present disclosure carries out the detailed description of embodiment below by way of the angle of server.
A kind of information-pushing method based on image recognition is provided in one embodiment of the disclosure, as shown in Figure 1, should
Method may comprise steps of:
Step S101 identifies the video frame images in video to be processed based on image-recognizing method, to determine view
At least one target object for including in frequency frame image；
For the present embodiment, video to be processed can be order video, and (i.e. client receives user's click etc. and plays
Playing request is sent to server after request, corresponding video is sent to client corresponding to the playing request by server,
Short-sighted frequency, the entertainment video that such as other users upload), it is also possible to live video etc., herein without limitation.If to
The video of processing is order video, can obtain the video of video to be processed by the video frame extraction technique based on self-adaption cluster
Frame image executes the magnanimity candidate video frame image decoded from video to be processed adaptive that is, using the thought of cluster
Then the cluster answered chooses the representative of video frame images closest with cluster centre in each cluster as such, thus
Realization determines video frame images to be identified from the candidate video frame image of magnanimity.If video to be processed is live streaming view
Frequently, the video frame of identification can be the present frame of the live video of acquisition, wherein can also be based on period regular hour to pumping
The video frame images of the live video taken are identified.
For the present embodiment, identifying processing is carried out to determining video frame images based on corresponding image-recognizing method, with
Determine at least one target object included in video frame images, wherein the target object can be clothes, mobile phone, book, food
The different classes of article such as product is also possible to a certain specific style or model, brand of clothes or other articles etc., does not do herein
It limits.
Step S102 determines the corresponding push relevant information of at least one target object；
For the present embodiment, by searching for accordingly or querying method, can be determined at least based on determining target object
The corresponding push relevant information of one target object.
Step S103 will push the corresponding target object of relevant information and push in association.
It specifically, can be by establishing push relevant information and at least one video frame figure comprising corresponding target object
Incidence relation as between enables push relevant information can be synchronous aobvious with the video frame images comprising specific objective object
Show, is pushed in association so that the corresponding target object of relevant information will be pushed.Wherein, push relevant information is in visitor
Family end can show the periphery display area of target object in video in embedded form or be overlapped in the display of target object
Region is also possible to existing for plug-in subtitle or other forms, herein without limitation.
An embodiment of the present disclosure provides a kind of information-pushing method based on image recognition, passes through people with the prior art
Work mode configures compared with the relevant Commdity advertisement of video content of user's viewing, and the present embodiment is by being based on image-recognizing method
Video frame images in video to be processed are identified, to determine at least one target object for including in video frame images,
Then it determines the corresponding push relevant information of at least one target object, will then push the corresponding mesh of relevant information
Mark object is pushed in association, i.e., determines at least one target object in video frame images by image-recognizing method,
And will push relevant information corresponding with each target object with corresponding target object is associated pushes, to realize
Push relevant information relevant to target object in video frame images automatically configures, in addition, based on different video frame figures
As identifying determining corresponding push relevant information respectively, realizes and pushed respectively accordingly in the different periods that user watches video
Push relevant information, to improve the rich of the push relevant information of configuration, and then the user experience is improved.
Another embodiment of the disclosure provides a kind of possible implementation, and specifically, step S101 includes:
Step S1011 (not shown) determines at least one candidate region in video frame images；
For the present embodiment, may there is multiple the quantity of target object in video frame images, pass through corresponding image recognition
Method determines at least one candidate region in video frame images, wherein it may include mesh in the region that the candidate region, which characterizes,
Mark object.
Step S1012 (not shown), determines the characteristic value of each candidate region, and it is right respectively to obtain each candidate region
The feature vector answered；
Specifically, CNN (Convolutional Neural can such as be passed through by corresponding feature extracting method
Networks, convolutional neural networks), the characteristic value for determining each candidate region is extracted respectively, then obtains each candidate region
Corresponding feature vector；Be also possible in advance extract video frame images characteristic value, be then based on each candidate region with
The mapping relations of the characteristic value of the video image of extraction, determine the characteristic value of each candidate region, to obtain each candidate regions
The corresponding feature vector in domain is not specifically limited the method for determination of each candidate region characteristic value herein.
Obtained each feature vector is input to the first nerves network of pre-training by step S1023 (not shown)
Model, to determine at least one target object for including in video frame images.
For the present embodiment, obtained each feature vector is input to the first nerves network model of pre-training, such as may be used
To be trained SVM (Support Vector Machine, support vector machines) or Softmax classifier, for each time
The target object that favored area includes carries out identification classification, to determine at least one target object for including in video frame images.
For the present embodiment, by determining the feature vector of at least one candidate region in video frame images, and it is based on
The neural network model of pre-training identifies the feature vector of input, determines at least one mesh for including in video frame images
Object is marked, the automatic identification of target object is realized, improves the efficiency of recongnition of objects.
The another embodiment of the disclosure provides a kind of possible implementation, and specifically, step S1011 includes:
Step S10111 (not shown) carries out image segmentation to video frame images and obtains multiple regions set；
Wherein, the main purpose of image segmentation (Image Segmentation) be image (image) is divided into it is several
A specific, with unique properties region (region), then therefrom extracts interested target (object), image district
Boundary definition between domain is the key that image segmentation algorithm.
It, can be by indicating the image segmentation algorithm of (graph-based) to video frame figure based on figure for the present embodiment
As carrying out image segmentation to obtain the set of multiple regions, wherein the algorithm is able to maintain low variation (low-variability)
The details in region (region), while can ignore that the details in the region High variation (high-variability) (region), i.e.,
There is a polymerization (grouping) well in High variation region, it can be visually consistent region distribution in the same area
Domain, to realize preferable image segmentation segmentation effect.
Step S10112 (not shown) merges place to obtained multiple regions set based on selection searching algorithm
Reason obtains at least one candidate region.
Wherein, the thought of selection search (Selective search) method is first to obtain one using the method for image segmentation
Then a little initial segmentation regions are closed these initial segmentation regions using the strategy of level grouping (being similar to hierarchical clustering)
And the candidate region that these obtained regions are positioned as target, relative to the brute-force search to candidate region, selection search energy
Search space is enough greatly lowered, improves algorithm speed.
For the present embodiment, processing is merged to obtained multiple regions set and obtains at least one candidate region.Its
In, whether adjacent area can be closed by calculating the similarity between adjacent area, and according to the determination of similarity calculation result
And the similarity for then calculating new region and adjacent area after merging determines whether merging treatment, is executed by continuous iteration,
Region after determining final merge is candidate region.Wherein the judging result of similarity can be based on color, texture, size,
One or more combinations is judged in space.
For the present embodiment, by carrying out image segmentation to video frame images, and to the multiple regions obtained after segmentation
Set carries out corresponding merging treatment and obtains at least one candidate region, to solve the candidate regions that possibility includes target object
The determination problem in domain, further to lay a good foundation to the identification for the target object for including in video frame images.
Disclosure another embodiment provides a kind of possible implementation, and specifically, step S1011 includes:
Step S10113 (not shown) extracts view by the feature extraction layer of the nervus opticus network model of pre-training
The feature of frequency frame image obtains characteristics of image figure；For the present embodiment, pass through the feature of the nervus opticus network model of pre-training
The feature that extract layer extracts video frame images obtains characteristics of image figure, wherein the nervus opticus network model of the pre-training can be with
It is training based on Faster-RCNN (Faster Regions with CNN features, the target inspection based on candidate region
Survey) neural network model, can pass through the convolutional layer of the model extract input video frame images feature.
Step S10114 (not shown) passes through the nervus opticus network mould of pre-training based on obtained characteristics of image figure
The candidate region generation layer of type determines at least one candidate region.
For the present embodiment, the characteristics of image figure of obtained video frame images is input to the nervus opticus network of pre-training
The candidate region generation layer of model is so that it is determined that at least one candidate region, wherein the candidate region generation layer, which can be, to be based on
Neural network model RPN (network is suggested in egion Proposal Network, candidate region) layer of Faster-RCNN, RPN layers
For exporting region proposals, it is possible to the candidate region comprising target object.
For the present embodiment, the candidate region of the video frame images of input is determined by the neural network model of pre-training,
To improve the efficiency that candidate region determines.
Disclosure another embodiment provides a kind of possible implementation, wherein step S102 includes:
Step S1021 (not shown) inquires either objective object by preset information recommendation repository,
With the determining push relevant information with multiple candidates of either objective object matching；
For the present embodiment, based on determining either objective Object Query information recommendation repository, the wherein information recommendation
The repository that repository can be we is also possible to third-party repository, so that it is determined that going out to match with either objective object
Multiple candidates push relevant information；Wherein, candidate push relevant information can also pass through phase based on either objective object
What the search engine answered was searched for；Wherein, the push relevant information to match with either objective object may include and target
The push relevant information of the associated object of object can also recommend related to table tennis bat etc. if target object is table tennis
Push relevant information.
Step S1022 (not shown) calculates separately the push relevant information and the phase of user to be recommended of multiple candidates
Guan Xing；
It specifically, can be based on the relevant information of user to be recommended, such as age, gender, hobby, user's history behavior
Record etc. calculates the correlation for determining multiple Candidate Recommendation relevant informations and user to be recommended.
Step S1023 (not shown) is based on correlativity calculation result, from the push relevant information of multiple candidates really
Surely relevant information is pushed.
Specifically, can be based on the correlativity calculation result of obtained each Candidate Recommendation information and user to be recommended, it will
The higher candidate push relevant information of correlativity calculation result is determined as pushing relevant information in multiple candidate push relevant informations.
For the present embodiment, determining candidate push relevant information is screened, to improve push relevant information
With the correlation of user to be recommended, and then promoted user experience.
Disclosure another embodiment provides a kind of possible implementation, wherein push relevant information include with down toward
One item missing:
Commercial product recommending link；The recommended information of target object and/or affiliated partner.
For the present embodiment, pushing relevant information includes but is not limited to commercial product recommending link, and/or, target object and/or
The recommended information (such as price, brand, the place of production) of affiliated partner, wherein above-mentioned recommended links or recommended information can be generated
Graphic code, so that user, which can choose, saves the graphic code, thus realize in the case where not influencing video-see, so that with
Family can choose understanding, the purchase of suitable time progress commodity.
For the present embodiment, pushing relevant information includes but is not limited to commercial product recommending link, and/or, target object and/or
The recommended information of affiliated partner, to realize that user will appreciate that the more information of the target object in video, if user is to certain
One target object is interested, can be bought by the recommended links of commodity.
Fig. 2 is a kind of information push-delivery apparatus based on image recognition for providing in one embodiment of the disclosure, the device
20 include: the first determining module 201, the second determining module 202, pushing module 203, wherein
First determining module 201, for being known based on image-recognizing method to the video frame images in video to be processed
Not, to determine at least one target object for including in video frame images；
Second determining module 202, at least one target object for determining that the first determining module 201 determines respectively correspond
Push relevant information；
Pushing module 203, for the target object that the determining push relevant information of the second determining module 202 is corresponding
It is pushed in association.
One embodiment of the disclosure provides a kind of information push-delivery apparatus based on image recognition, passes through with the prior art
Manual type configures compared with the relevant Commdity advertisement of the video content that user watches, and the present embodiment is by being based on image recognition side
Method identifies the video frame images in video to be processed, to determine at least one target pair for including in video frame images
As then determining the corresponding push relevant information of at least one target object, it is corresponding will then to push relevant information
Target object pushed in association, i.e., at least one target pair in video frame images is determined by image-recognizing method
As, and will push relevant information corresponding with each target object with corresponding target object is associated pushes, thus
Automatically configuring for push relevant information relevant to target object in video frame images is realized, in addition, based on different videos
Frame image identifies respectively determines corresponding push relevant information, realizes and pushes respectively in the different periods that user watches video
Corresponding push relevant information, to improve the rich of the push relevant information of configuration, and then the user experience is improved.
The device of the present embodiment can be performed a kind of information based on image recognition provided in disclosure above-described embodiment and push away
Delivery method, realization principle is similar, and details are not described herein again.
An embodiment of the present disclosure provides another information push-delivery apparatus based on image recognition, as shown in figure 3, this reality
The device 30 for applying example may include: the first determining module 301, the second determining module 302, pushing module 303；
First determining module 301, for being known based on image-recognizing method to the video frame images in video to be processed
Not, to determine at least one target object for including in video frame images；
Wherein, the first determining module 301 in Fig. 3 is identical as the function of the first determining module 201 in Fig. 2 or phase
Seemingly.
Second determining module 302, at least one target object for determining that the first determining module 301 determines respectively correspond
Push relevant information；
Wherein, the second determining module 302 in Fig. 3 is identical as the function of the second determining module 202 in Fig. 2 or phase
Seemingly.
Pushing module 303, for the target object that the determining push relevant information of the second determining module 302 is corresponding
It is pushed in association.
Wherein, the pushing module 303 in Fig. 3 is same or similar with the function of pushing module 203 in Fig. 2.
The another embodiment of the disclosure provides a kind of possible implementation, and specifically, the first determining module 301 includes the
One determination unit 3011, the second determination unit 3012 and third determination unit 3013；
First determination unit 3011, for determining at least one candidate region in video frame images；
Second determination unit 3012, for determining the characteristic value of the determining each candidate region of the first determination unit 3011,
Obtain the corresponding feature vector in each candidate region；
Third determination unit 3013, each feature vector for obtaining the second determination unit 3012 are input to pre-training
First nerves network model, to determine at least one target object for including in video frame images.
For the present embodiment, by determining the feature vector of at least one candidate region in video frame images, and it is based on
The neural network model of pre-training identifies the feature vector of input, determines at least one mesh for including in video frame images
Object is marked, the automatic identification of target object is realized, improves the efficiency of recongnition of objects.
The another embodiment of the disclosure provides a kind of possible implementation, and specifically, the first determination unit 3011 is also used
Multiple regions set is obtained in carrying out image segmentation to video frame images, and for more to what is obtained based on selection searching algorithm
A regional ensemble merges processing and obtains at least one candidate region.
For the present embodiment, by carrying out image segmentation to video frame images, and to the multiple regions obtained after segmentation
Set carries out corresponding merging treatment and obtains at least one candidate region, to solve the candidate regions that possibility includes target object
The determination problem in domain, further to lay a good foundation to the identification for the target object for including in video frame images.
The another embodiment of the disclosure provides a kind of possible implementation, and specifically, the first determination unit 3011 is also used
Characteristics of image figure is obtained in the feature that the feature extraction layer of the nervus opticus network model by pre-training extracts video frame images,
And for being determined based on obtained characteristics of image figure by the candidate region generation layer of the nervus opticus network model of pre-training
At least one candidate region.
For the present embodiment, the candidate region of the video frame images of input is determined by the neural network model of pre-training,
To improve the efficiency that candidate region determines.
The another embodiment of the disclosure provides a kind of possible implementation, and specifically, the second determining module 302 includes the
Four determination units 3021, computing unit 3022 and the 5th determination unit 3023；
4th determination unit 3021, for either objective object to be inquired by preset information recommendation repository,
With the determining push relevant information with multiple candidates of either objective object matching；
Computing unit 3022, for calculating separately the push relevant information for multiple candidates that the 4th determination unit 3021 determines
With the correlation of user to be recommended；
5th determination unit 3023, for calculating determining correlativity calculation result based on computing unit 3022, from multiple
Push relevant information is determined in candidate push relevant information.
For the present embodiment, determining candidate push relevant information is screened, to improve push relevant information
With the correlation of user to be recommended, and then promoted user experience.
The another embodiment of the disclosure provides a kind of possible implementation, wherein push relevant information include with down toward
One item missing:
Commercial product recommending link；The recommended information of target object and/or affiliated partner.
For the present embodiment, pushing relevant information includes but is not limited to commercial product recommending link, and/or, target object and/or
The recommended information of affiliated partner, to realize that user will appreciate that the more information of the target object in video, if user is to certain
One target object is interested, can be bought by the recommended links of commodity.
An embodiment of the present disclosure provides a kind of information push-delivery apparatus based on image recognition, passes through people with the prior art
Work mode configures compared with the relevant Commdity advertisement of video content of user's viewing, and the present embodiment is by being based on image-recognizing method
Video frame images in video to be processed are identified, to determine at least one target object for including in video frame images,
Then it determines the corresponding push relevant information of at least one target object, will then push the corresponding mesh of relevant information
Mark object is pushed in association, i.e., determines at least one target object in video frame images by image-recognizing method,
And will push relevant information corresponding with each target object with corresponding target object is associated pushes, to realize
Push relevant information relevant to target object in video frame images automatically configures, in addition, based on different video frame figures
As identifying determining corresponding push relevant information respectively, realizes and pushed respectively accordingly in the different periods that user watches video
Push relevant information, to improve the rich of the push relevant information of configuration, and then the user experience is improved.
The embodiment of the present disclosure provides a kind of information recommending apparatus based on image recognition, is suitable for shown in above-described embodiment
Method, details are not described herein.
In an alternative embodiment, a kind of electronic equipment is provided, as shown in figure 4, real it illustrates being suitable for being used to
The structural schematic diagram of the electronic equipment (such as terminal device or server) 40 of the existing embodiment of the present disclosure.In the embodiment of the present disclosure
Terminal device can include but is not limited to such as mobile phone, laptop, digit broadcasting receiver, PDA, and (individual digital helps
Reason), the shifting of PAD (tablet computer), PMP (portable media player), car-mounted terminal (such as vehicle mounted guidance terminal) etc.
The fixed terminal of dynamic terminal and such as number TV, desktop computer etc..Electronic equipment shown in Fig. 4 is only one and shows
Example, should not function to the embodiment of the present disclosure and use scope bring any restrictions.
As shown in figure 4, electronic equipment 40 may include processing unit (such as central processing unit, graphics processor etc.) 401,
It can be loaded into random access storage according to the program being stored in read-only memory (ROM) 402 or from storage device 408
Program in device (RAM) 403 and execute various movements appropriate and processing.In RAM 403, it is also stored with the behaviour of electronic equipment 40
Various programs and data needed for making.Processing unit 401, ROM 402 and RAM 403 are connected with each other by bus 404.It is defeated
Enter/export (I/O) interface 405 and is also connected to bus 404.
In general, following device can connect to I/O interface 405: including such as touch screen, touch tablet, keyboard, mouse, taking the photograph
As the input unit 406 of head, microphone, accelerometer, gyroscope etc.；Including such as liquid crystal display (LCD), loudspeaker, vibration
The output device 407 of dynamic device etc.；Storage device 408 including such as tape, hard disk etc.；And communication device 409.Communication device
409, which can permit electronic equipment 40, is wirelessly or non-wirelessly communicated with other equipment to exchange data.Although Fig. 4, which is shown, to be had
The electronic equipment 40 of various devices, it should be understood that being not required for implementing or having all devices shown.It can substitute
Implement or have more or fewer devices in ground.
An embodiment of the present disclosure provides a kind of electronic equipment, is seen by manual type configuration with user with the prior art
The relevant Commdity advertisement of the video content seen is compared, the present embodiment by based on image-recognizing method to the view in video to be processed
Frequency frame image is identified, to determine at least one target object for including in video frame images, then determines at least one mesh
The corresponding push relevant information of object is marked, the corresponding target object of relevant information will be then pushed and carries out in association
Push, i.e., determine at least one target object in video frame images by image-recognizing method, and will be with each target object
Corresponding push relevant information with corresponding target object is associated is pushed, to realize and mesh in video frame images
The relevant push relevant information of mark object automatically configures, and determines accordingly in addition, being identified respectively based on different video frame images
Push relevant information, realize in the different periods that user watches video the corresponding push relevant information of push respectively, from
And improve configuration push relevant information it is rich, and then the user experience is improved.
A kind of electronic equipment is present embodiments provided suitable for above method embodiment, details are not described herein.
Particularly, in accordance with an embodiment of the present disclosure, it may be implemented as computer above with reference to the process of flow chart description
Software program.For example, embodiment of the disclosure includes a kind of computer program product comprising be carried on computer-readable medium
On computer program, which includes the program code for method shown in execution flow chart.In such reality
It applies in example, which can be downloaded and installed from network by communication device 409, or from storage device 408
It is mounted, or is mounted from ROM 402.When the computer program is executed by processing unit 401, the embodiment of the present disclosure is executed
Method in the above-mentioned function that limits.
It should be noted that the above-mentioned computer-readable medium of the disclosure can be computer-readable signal media or meter
Calculation machine readable storage medium storing program for executing either the two any combination.Computer readable storage medium for example can be --- but not
Be limited to --- electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor system, device or device, or any above combination.Meter
The more specific example of calculation machine readable storage medium storing program for executing can include but is not limited to: have the electrical connection, just of one or more conducting wires
Taking formula computer disk, hard disk, random access storage device (RAM), read-only memory (ROM), erasable type may be programmed read-only storage
Device (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), light storage device, magnetic memory device,
Or above-mentioned any appropriate combination.In the disclosure, computer readable storage medium can be it is any include or storage journey
The tangible medium of sequence, the program can be commanded execution system, device or device use or in connection.And at this
In open, computer-readable signal media may include in a base band or as the data-signal that carrier wave a part is propagated,
In carry computer-readable program code.The data-signal of this propagation can take various forms, including but not limited to
Electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be computer-readable and deposit
Any computer-readable medium other than storage media, the computer-readable signal media can send, propagate or transmit and be used for
By the use of instruction execution system, device or device or program in connection.Include on computer-readable medium
Program code can transmit with any suitable medium, including but not limited to: electric wire, optical cable, RF (radio frequency) etc. are above-mentioned
Any appropriate combination.
Above-mentioned computer-readable medium can be included in above-mentioned electronic equipment；It is also possible to individualism, and not
It is fitted into the electronic equipment.
Above-mentioned computer-readable medium carries one or more program, when said one or multiple programs are by the electricity
When sub- equipment executes, so that the electronic equipment: obtaining at least two internet protocol addresses；Send to Node evaluation equipment includes extremely
The Node evaluation request of few two internet protocol addresses, wherein Node evaluation equipment is from least two internet protocol addresses, choosing
It takes internet protocol address and returns；The internet protocol address that receiving node valuator device returns；Wherein, acquired Internet protocol
Address indicates the fringe node in content distributing network.
Alternatively, above-mentioned computer-readable medium carries one or more program, when said one or multiple programs
When being executed by the electronic equipment, so that the electronic equipment: receiving the Node evaluation including at least two internet protocol addresses and request；
From at least two internet protocol addresses, internet protocol address is chosen；Return to the internet protocol address selected；Wherein, it receives
To internet protocol address instruction content distributing network in fringe node.
The calculating of the operation for executing the disclosure can be write with one or more programming languages or combinations thereof
Machine program code, above procedure design language include object oriented program language-such as Java, Smalltalk, C+
+, it further include conventional procedural programming language-such as " C " language or similar programming language.Program code can
Fully to execute, partly execute on the user computer on the user computer, be executed as an independent software package,
Part executes on the remote computer or executes on a remote computer or server completely on the user computer for part.
In situations involving remote computers, remote computer can pass through the network of any kind --- including local area network (LAN)
Or wide area network (WAN)-is connected to subscriber computer, or, it may be connected to outer computer (such as utilize Internet service
Provider is connected by internet).
An embodiment of the present disclosure provides a kind of computer readable storage medium, is matched with the prior art by manual type
It sets compared with the relevant Commdity advertisement of video content of user's viewing, the present embodiment is by being based on image-recognizing method to be processed
Video frame images in video are identified, to determine at least one target object for including in video frame images, are then determined
The corresponding push relevant information of at least one target object will then push the corresponding target object phase of relevant information
Associatedly pushed, i.e., at least one target object in video frame images determined by image-recognizing method, and will with it is each
The corresponding push relevant information of a target object with corresponding target object is associated is pushed, to realize and video
Target object is relevant in frame image pushes automatically configuring for relevant information, in addition, being known respectively based on different video frame images
Relevant information Que Ding not be pushed accordingly, realized and pushed corresponding push phase respectively in the different periods that user watches video
Information is closed, to improve the rich of the push relevant information of configuration, and then the user experience is improved.
It present embodiments provides a kind of computer readable storage medium and is suitable for above method embodiment, it is no longer superfluous herein
It states.
Flow chart and block diagram in attached drawing are illustrated according to the system of the various embodiments of the disclosure, method and computer journey
The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation
A part of one module, program segment or code of table, a part of the module, program segment or code include one or more use
The executable instruction of the logic function as defined in realizing.It should also be noted that in some implementations as replacements, being marked in box
The function of note can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are actually
It can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it to infuse
Meaning, the combination of each box in block diagram and or flow chart and the box in block diagram and or flow chart can be with holding
The dedicated hardware based system of functions or operations as defined in row is realized, or can use specialized hardware and computer instruction
Combination realize.
Being described in unit involved in the embodiment of the present disclosure can be realized by way of software, can also be by hard
The mode of part is realized.Wherein, the title of unit does not constitute the restriction to the unit itself under certain conditions.
Above description is only the preferred embodiment of the disclosure and the explanation to institute's application technology principle.Those skilled in the art
Member is it should be appreciated that the open scope involved in the disclosure, however it is not limited to technology made of the specific combination of above-mentioned technical characteristic
Scheme, while should also cover in the case where not departing from design disclosed above, it is carried out by above-mentioned technical characteristic or its equivalent feature
Any combination and the other technical solutions formed.Such as features described above has similar function with (but being not limited to) disclosed in the disclosure
Can technical characteristic replaced mutually and the technical solution that is formed.
Claims (10)
1. a kind of information-pushing method based on image recognition characterized by comprising
The video frame images in video to be processed are identified based on image-recognizing method, in the determination video frame images
Including at least one target object；
Determine the corresponding push relevant information of at least one described target object；
The corresponding target object of the push relevant information is pushed in association.
2. the method according to claim 1, wherein it is described based on image-recognizing method in video to be processed
Video frame images are identified, at least one target object for including in the determination video frame images, comprising:
Determine at least one candidate region in the video frame images；
The characteristic value for determining each candidate region obtains the corresponding feature vector in each candidate region；
Obtained each feature vector is input to the first nerves network model of pre-training, in the determination video frame images
Including at least one target object.
3. according to the method described in claim 2, it is characterized in that, at least one of described video frame images of the determination are waited
Favored area, comprising:
Image segmentation is carried out to the video frame images and obtains multiple regions set；
Processing is merged to obtained multiple regions set based on selection searching algorithm and obtains at least one described candidate region.
4. according to the method described in claim 2, it is characterized in that, at least one of described video frame images of the determination are waited
Favored area, comprising:
Image is obtained by the feature that the feature extraction layer of the nervus opticus network model of pre-training extracts the video frame images
Characteristic pattern；
It is determined based on obtained described image characteristic pattern by the candidate region generation layer of the nervus opticus network model of pre-training
At least one described candidate region.
5. being pushed away the method according to claim 1, wherein at least one target object is corresponding described in determining
Send relevant information, comprising:
Either objective object is inquired by preset information recommendation repository, with the determining and either objective object
The push relevant information of the multiple candidates matched；
Calculate separately the push relevant information and the correlation of user to be recommended of multiple candidates；
Based on correlativity calculation result, push relevant information is determined from the push relevant information of multiple candidates.
6. the method according to claim 1, wherein the push relevant information includes at least one of the following:
Commercial product recommending link；The recommended information of target object and/or affiliated partner.
7. a kind of information push-delivery apparatus based on image recognition characterized by comprising
First determining module, for being identified based on image-recognizing method to the video frame images in video to be processed, with true
At least one target object for including in the fixed video frame images；
Second determining module, for determining that at least one target object described in the first determining module determination is corresponding
Push relevant information；
Pushing module, for the target object phase that the determining push relevant information of second determining module is corresponding
Associatedly pushed.
8. device according to claim 7, which is characterized in that first determining module includes the first determination unit, the
Two determination units and third determination unit；
First determination unit, for determining at least one candidate region in the video frame images；
Second determination unit is obtained for determining the characteristic value of the determining each candidate region of first determination unit
The corresponding feature vector in each candidate region；
The third determination unit, each feature vector for obtaining second determination unit are input to the of pre-training
One neural network model, at least one target object for including in the determination video frame images.
9. a kind of electronic equipment characterized by comprising
One or more processors；
Memory；
One or more application program, wherein one or more of application programs are stored in the memory and are configured
To be executed by one or more of processors, one or more of programs are configured to: being executed according to claim 1 to 6
Described in any item information-pushing methods based on image recognition.
10. a kind of computer readable storage medium, which is characterized in that the computer storage medium refers to for storing computer
Enable, when run on a computer, allow computer execute described in any one of the claims 1 to 6 based on
The information-pushing method of image recognition.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811447987.6A CN109495784A (en) | 2018-11-29 | 2018-11-29 | Information-pushing method, device, electronic equipment and computer readable storage medium |
| PCT/CN2018/125389 WO2020107624A1 (en) | 2018-11-29 | 2018-12-29 | Information pushing method and apparatus, electronic device and computer-readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811447987.6A CN109495784A (en) | 2018-11-29 | 2018-11-29 | Information-pushing method, device, electronic equipment and computer readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109495784A true CN109495784A (en) | 2019-03-19 |
Family
ID=65698736
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811447987.6A Pending CN109495784A (en) | 2018-11-29 | 2018-11-29 | Information-pushing method, device, electronic equipment and computer readable storage medium |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN109495784A (en) |
| WO (1) | WO2020107624A1 (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110189172A (en) * | 2019-05-28 | 2019-08-30 | 广州华多网络科技有限公司 | Shopping guide method and system between a kind of network direct broadcasting |
| CN110198473A (en) * | 2019-06-10 | 2019-09-03 | 北京字节跳动网络技术有限公司 | Method for processing video frequency, device, electronic equipment and computer readable storage medium |
| CN110751086A (en) * | 2019-10-17 | 2020-02-04 | 北京字节跳动网络技术有限公司 | Target searching method, device, equipment and storage medium based on video |
| CN110781980A (en) * | 2019-11-08 | 2020-02-11 | 北京金山云网络技术有限公司 | Training method of target detection model, target detection method and device |
| CN111447260A (en) * | 2020-03-20 | 2020-07-24 | 北京字节跳动网络技术有限公司 | Information pushing and information publishing method and device |
| CN111753136A (en) * | 2019-11-14 | 2020-10-09 | 北京沃东天骏信息技术有限公司 | Article information processing method, article information processing device, medium, and electronic device |
| CN111815340A (en) * | 2019-04-12 | 2020-10-23 | 百度在线网络技术（北京）有限公司 | Promotion information determination method, device, equipment and readable storage medium |
| CN112016986A (en) * | 2019-05-31 | 2020-12-01 | 阿里巴巴集团控股有限公司 | Commodity information display method and device and electronic equipment |
| CN113438532A (en) * | 2021-05-31 | 2021-09-24 | 北京达佳互联信息技术有限公司 | Video processing method, video playing method, video processing device, video playing device, electronic equipment and storage medium |
| CN113674007A (en) * | 2020-05-14 | 2021-11-19 | 北京达佳互联信息技术有限公司 | Advertisement recommendation method and device |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102930553A (en) * | 2011-08-10 | 2013-02-13 | 中国移动通信集团上海有限公司 | Method and device for identifying objectionable video content |
| CN104050682A (en) * | 2014-07-09 | 2014-09-17 | 武汉科技大学 | Image segmentation method fusing color and depth information |
| CN104077329A (en) * | 2013-03-29 | 2014-10-01 | 西门子公司 | Information recommending method and information recommending system |
| CN104573706A (en) * | 2013-10-25 | 2015-04-29 | Tcl集团股份有限公司 | Object identification method and system thereof |
| CN104754413A (en) * | 2013-12-30 | 2015-07-01 | 北京三星通信技术研究有限公司 | Image search based television signal identification and information recommendation method and device |
| CN106528586A (en) * | 2016-05-13 | 2017-03-22 | 上海理工大学 | Human behavior video identification method |
| CN106792004A (en) * | 2016-12-30 | 2017-05-31 | 北京小米移动软件有限公司 | Content item method for pushing, apparatus and system |
| CN107679495A (en) * | 2017-10-09 | 2018-02-09 | 济南大学 | A kind of detection method of transmission line of electricity periphery activity engineering truck |
| CN107844794A (en) * | 2016-09-21 | 2018-03-27 | 北京旷视科技有限公司 | Image-recognizing method and device |
| US20180091844A1 (en) * | 2011-01-06 | 2018-03-29 | Veveo, Inc. | Methods of and systems for content search based on environment sampling |
| CN108121984A (en) * | 2016-11-30 | 2018-06-05 | 杭州海康威视数字技术股份有限公司 | A kind of character identifying method and device |
| CN108205538A (en) * | 2016-12-16 | 2018-06-26 | 北京酷我科技有限公司 | The extracting method and system of a kind of key frame of video |
| CN108491419A (en) * | 2018-02-06 | 2018-09-04 | 北京奇虎科技有限公司 | It is a kind of to realize the method and apparatus recommended based on video |
Family Cites Families (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108629224B (en) * | 2017-03-15 | 2019-11-05 | 北京京东尚科信息技术有限公司 | Information demonstrating method and device |
| CN108509466A (en) * | 2017-04-14 | 2018-09-07 | 腾讯科技（深圳）有限公司 | A kind of information recommendation method and device |
| CN108446390B (en) * | 2018-03-22 | 2022-01-04 | 百度在线网络技术（北京）有限公司 | Method and device for pushing information |
| CN108734516A (en) * | 2018-05-18 | 2018-11-02 | 上海极歌企业管理咨询中心（有限合伙） | Advertisement placement method and device |
-
2018
- 2018-11-29 CN CN201811447987.6A patent/CN109495784A/en active Pending
- 2018-12-29 WO PCT/CN2018/125389 patent/WO2020107624A1/en active Application Filing
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20180091844A1 (en) * | 2011-01-06 | 2018-03-29 | Veveo, Inc. | Methods of and systems for content search based on environment sampling |
| CN102930553A (en) * | 2011-08-10 | 2013-02-13 | 中国移动通信集团上海有限公司 | Method and device for identifying objectionable video content |
| CN104077329A (en) * | 2013-03-29 | 2014-10-01 | 西门子公司 | Information recommending method and information recommending system |
| CN104573706A (en) * | 2013-10-25 | 2015-04-29 | Tcl集团股份有限公司 | Object identification method and system thereof |
| CN104754413A (en) * | 2013-12-30 | 2015-07-01 | 北京三星通信技术研究有限公司 | Image search based television signal identification and information recommendation method and device |
| CN104050682A (en) * | 2014-07-09 | 2014-09-17 | 武汉科技大学 | Image segmentation method fusing color and depth information |
| CN106528586A (en) * | 2016-05-13 | 2017-03-22 | 上海理工大学 | Human behavior video identification method |
| CN107844794A (en) * | 2016-09-21 | 2018-03-27 | 北京旷视科技有限公司 | Image-recognizing method and device |
| CN108121984A (en) * | 2016-11-30 | 2018-06-05 | 杭州海康威视数字技术股份有限公司 | A kind of character identifying method and device |
| CN108205538A (en) * | 2016-12-16 | 2018-06-26 | 北京酷我科技有限公司 | The extracting method and system of a kind of key frame of video |
| CN106792004A (en) * | 2016-12-30 | 2017-05-31 | 北京小米移动软件有限公司 | Content item method for pushing, apparatus and system |
| CN107679495A (en) * | 2017-10-09 | 2018-02-09 | 济南大学 | A kind of detection method of transmission line of electricity periphery activity engineering truck |
| CN108491419A (en) * | 2018-02-06 | 2018-09-04 | 北京奇虎科技有限公司 | It is a kind of to realize the method and apparatus recommended based on video |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111815340A (en) * | 2019-04-12 | 2020-10-23 | 百度在线网络技术（北京）有限公司 | Promotion information determination method, device, equipment and readable storage medium |
| CN111815340B (en) * | 2019-04-12 | 2023-09-01 | 百度在线网络技术（北京）有限公司 | Popularization information determination method, device, equipment and readable storage medium |
| CN110189172A (en) * | 2019-05-28 | 2019-08-30 | 广州华多网络科技有限公司 | Shopping guide method and system between a kind of network direct broadcasting |
| CN112016986A (en) * | 2019-05-31 | 2020-12-01 | 阿里巴巴集团控股有限公司 | Commodity information display method and device and electronic equipment |
| CN110198473A (en) * | 2019-06-10 | 2019-09-03 | 北京字节跳动网络技术有限公司 | Method for processing video frequency, device, electronic equipment and computer readable storage medium |
| CN110751086A (en) * | 2019-10-17 | 2020-02-04 | 北京字节跳动网络技术有限公司 | Target searching method, device, equipment and storage medium based on video |
| CN110781980A (en) * | 2019-11-08 | 2020-02-11 | 北京金山云网络技术有限公司 | Training method of target detection model, target detection method and device |
| CN110781980B (en) * | 2019-11-08 | 2022-04-12 | 北京金山云网络技术有限公司 | Training method of target detection model, target detection method and device |
| CN111753136A (en) * | 2019-11-14 | 2020-10-09 | 北京沃东天骏信息技术有限公司 | Article information processing method, article information processing device, medium, and electronic device |
| CN111447260A (en) * | 2020-03-20 | 2020-07-24 | 北京字节跳动网络技术有限公司 | Information pushing and information publishing method and device |
| CN113674007A (en) * | 2020-05-14 | 2021-11-19 | 北京达佳互联信息技术有限公司 | Advertisement recommendation method and device |
| CN113438532A (en) * | 2021-05-31 | 2021-09-24 | 北京达佳互联信息技术有限公司 | Video processing method, video playing method, video processing device, video playing device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2020107624A1 (en) | 2020-06-04 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109495784A (en) | Information-pushing method, device, electronic equipment and computer readable storage medium | |
| US11610387B2 (en) | Systems and methods for automated object recognition | |
| CN108307240B (en) | Video recommendation method and device | |
| US20190392330A1 (en) | System and method for generating aspect-enhanced explainable description-based recommendations | |
| CN107888950A (en) | A kind of method and system for recommending video | |
| CN102884538B (en) | Online Video is enriched by content detection, search and information aggregation | |
| CN108540826A (en) | Barrage method for pushing, device, electronic equipment and storage medium | |
| CN110519621A (en) | Video recommendation method, device, electronic equipment and computer-readable medium | |
| US9965478B1 (en) | Automatic generation of online media stations customized to individual users | |
| CN110020094A (en) | A kind of methods of exhibiting and relevant apparatus of search result | |
| CN110321958A (en) | Training method, the video similarity of neural network model determine method | |
| CN108140041A (en) | It is clustered for the viewing time of video search | |
| CN108040294A (en) | Automatic recommendation | |
| CN108960316A (en) | Method and apparatus for generating model | |
| CN110267097A (en) | Video pushing method, device and electronic equipment based on characteristic of division | |
| CN106874460A (en) | Information recommendation method, apparatus and system | |
| CN113760158A (en) | Target object display method, object association method, device, medium and equipment | |
| CN110363604A (en) | Page generation method and device | |
| CN109376268A (en) | Video classification methods, device, electronic equipment and computer readable storage medium | |
| CN110019949A (en) | Video recommendation method, device, terminal, server and readable medium | |
| CN109165344A (en) | Method and apparatus for pushed information | |
| CN108960912A (en) | Method and apparatus for determining target position | |
| CN110278447A (en) | Video pushing method, device and electronic equipment based on continuous feature | |
| CN110209658A (en) | Data cleaning method and device | |
| CN110276006A (en) | Methods of exhibiting, device, electronic equipment and the computer readable storage medium of painted eggshell |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190319 |