CN109918530A - Method and apparatus for pushing image - Google Patents
Method and apparatus for pushing image Download PDFInfo
- Publication number
- CN109918530A CN109918530A CN201910159569.5A CN201910159569A CN109918530A CN 109918530 A CN109918530 A CN 109918530A CN 201910159569 A CN201910159569 A CN 201910159569A CN 109918530 A CN109918530 A CN 109918530A
- Authority
- CN
- China
- Prior art keywords
- information
- user
- image
- relevant
- vector
- 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
Abstract
Embodiment of the disclosure discloses the method and apparatus for pushing image.One specific embodiment of this method includes: the information for obtaining user and utilizing terminal input；Determine relevant information relevant to the information that user inputs；The information and relevant information that user is inputted, the image for including with preset image library are matched, and the information inputted with user and the matching image to match with relevant information are obtained；Matching image is sent to terminal, so that matching image is displayed on the terminals.The embodiment can fully utilize limited display screen resource, accurately show image to user, need to show more image occupied display resource and storage resource for selection by the user in the prior art so as to reduce.In addition, during embodiment of the disclosure can also reduce user's selection image, the number of the terminal to server request image of user, it is possible thereby to reduce the occupancy to Internet resources.
Description
Technical field
Embodiment of the disclosure is related to field of computer technology, and in particular to the method and apparatus for pushing image.
Background technique
In general, when user's using terminal equipment is chatted, is commented on, need using some images (such as facial expression image,
Dynamic picture etc.).Currently, when user inputs text, the text that terminal device can be inputted according to user is examined from picture library
Rope goes out the picture of the characters matching inputted with user, to quickly present and input to user when user chats, comments on
Text relevant images for user selection.User can select image from the image of presentation, and selected image is sent,
Or the image of selection is saved.
Summary of the invention
Embodiment of the disclosure proposes the method and apparatus for pushing image.
In a first aspect, embodiment of the disclosure provides a kind of method for pushing image, used this method comprises: obtaining
Family utilizes the information of terminal input；Determine relevant information relevant to the information that user inputs；The information and phase that user is inputted
Close information, the image for including with preset image library matched, obtain the information inputted with user and with relevant information phase
The matching image matched；Matching image is sent to terminal, so that matching image is displayed on the terminals.
In some embodiments, relevant information relevant to the information that user inputs is determined, comprising: in response to determining user
The information of input is the comment information commented on the information currently browsed of user, determine information that user currently browses be with
The relevant relevant information of information of user's input.
In some embodiments, relevant information relevant to the information that user inputs is determined, comprising: in response to determining user
The information of input is to be sent to the information of the terminal of target user, from the historical information that user browsed, determining and user
The relevant relevant information of the information of input.
In some embodiments, from the historical information that user browsed, phase relevant to the information that user inputs is determined
Close information, comprising: from the historical information that user browsed, determine the information of target user's publication as the letter inputted with user
Cease relevant information.
In some embodiments, for the image in image library, which corresponds to preset feature vector；And it will use
The information and relevant information of family input, the image for including with preset image library are matched, and the information inputted with user is obtained
With the matching image to match with relevant information, comprising: determine user input information feature vector as fisrt feature to
Amount, and determine the feature vector of relevant information as second feature vector；Based on first eigenvector and second feature vector,
Generate vector to be matched；For the image in image library, determine between vector to be matched feature vector corresponding with the image
Similarity；By in identified similarity, the corresponding image of similarity more than or equal to preset similarity threshold is determined as
With image.
In some embodiments, the type for the information that the information of first eigenvector characterization and second feature vector characterize is not
Together；And it is based on first eigenvector and second feature vector, generate vector to be matched, comprising: by first eigenvector and the
Two feature vectors input vector transformation model trained in advance, obtains vector to be matched, wherein vector transformation model is used for will be defeated
Two vectors entered are converted to a vector.
Second aspect, embodiment of the disclosure provide it is a kind of for pushing the device of image, the device include: obtain it is single
Member is configured to obtain the information that user utilizes terminal input；Determination unit is configured to determine the information phase with user's input
The relevant information of pass；Matching unit is configured to the information and relevant information for inputting user, includes with preset image library
Image is matched, and the information inputted with user and the matching image to match with relevant information are obtained；Transmission unit is configured
It is sent to terminal at by matching image, so that matching image is displayed on the terminals.
In some embodiments, determination unit is further configured to: in response to determining that the information of user's input is user
To the comment information that the information currently browsed is commented on, determine that the information that user currently browses is the information phase inputted with user
The relevant information of pass.
In some embodiments, determination unit is further configured to: in response to determining that the information of user's input is pending
It send to the information of the terminal of target user, from the historical information that user browsed, determination is relevant to the information that user inputs
Relevant information.
In some embodiments, determination unit is further configured to: from the historical information that user browsed, determining mesh
The information of user's publication is marked as information relevant to the information that user inputs.
In some embodiments, for the image in image library, which corresponds to preset feature vector；And matching
Unit includes: the first determining module, is configured to determine the feature vector of the information of user's input as first eigenvector, with
And determine the feature vector of relevant information as second feature vector；Generation module, be configured to based on first eigenvector and
Second feature vector generates vector to be matched；Second determining module, is configured to for the image in image library, determine to
With the similarity between vector feature vector corresponding with the image；Third determining module, being configured to will be identified similar
In degree, the corresponding image of similarity more than or equal to preset similarity threshold is determined as matching image.
In some embodiments, the type for the information that the information of first eigenvector characterization and second feature vector characterize is not
Together；And generation module is further configured to: by first eigenvector and the input of second feature vector vector trained in advance
Transformation model obtains vector to be matched, wherein vector transformation model is used to input two vectors being converted to a vector.
The third aspect, embodiment of the disclosure provide a kind of server, which includes: one or more processing
Device；Storage device is stored thereon with one or more programs；When one or more programs are executed by one or more processors,
So that one or more processors realize the method as described in implementation any in first aspect.
Fourth aspect, embodiment of the disclosure provide a kind of computer-readable medium, are stored thereon with computer program,
The method as described in implementation any in first aspect is realized when the computer program is executed by processor.
The method and apparatus for pushing image that embodiment of the disclosure provides are inputted by obtaining user using terminal
Information, determine relevant to the information that user inputs relevant information, it is and pre- then by the information and relevant information of user's input
If the image library image that includes matched, obtain matching image, matching image be finally sent to terminal, so that matching is schemed
As displayed on the terminals, to realize the information and relevant information for being flexibly utilized by user's input, matching image obtained, is helped
In targetedly to user's displaying image.In compared with the existing technology, the image and the letter of user's input that are shown to user
Cease related, the available richer matching image of embodiment of the disclosure to user to show.The terminal used due to user
Display screen size it is limited, by embodiment of the disclosure, can fully utilize limited display screen resource, accurately to
User shows image, needs to show that more image is occupied aobvious for selection by the user in the prior art so as to reduce
Show resource and storage resource.
Detailed description of the invention
By reading a detailed description of non-restrictive embodiments in the light of the attached drawings below, the disclosure is other
Feature, objects and advantages will become more apparent upon:
Fig. 1 is that one embodiment of the disclosure can be applied to exemplary system architecture figure therein；
Fig. 2 is according to an embodiment of the present disclosure for pushing the flow chart of one embodiment of the method for image；
Fig. 3 is according to an embodiment of the present disclosure for pushing the schematic diagram of an application scenarios of the method for image；
Fig. 4 is according to an embodiment of the present disclosure for pushing the flow chart of another embodiment of the method for image；
Fig. 5 is according to an embodiment of the present disclosure for pushing the structural schematic diagram of one embodiment of the device of image；
Fig. 6 is adapted for the structural schematic diagram for realizing the server of embodiment of the disclosure.
Specific embodiment
The disclosure is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining that correlation is open, rather than the restriction to the disclosure.It also should be noted that in order to
Convenient for description, is illustrated only in attached drawing and disclose relevant part to related.
It should be noted that in the absence of conflict, the feature in embodiment and embodiment in the disclosure can phase
Mutually combination.The disclosure is described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.
Fig. 1 is shown can the method for pushing image using embodiment of the disclosure or the dress for pushing image
The exemplary system architecture 100 set.
As shown in Figure 1, system architecture 100 may include terminal device 101,102,103, network 104 and server 105.
Network 104 between terminal device 101,102,103 and server 105 to provide the medium of communication link.Network 104 can be with
Including various connection types, such as wired, wireless communication link or fiber optic cables etc..
User can be used terminal device 101,102,103 and be interacted by network 104 with server 105, to receive or send out
Send message etc..Various telecommunication customer end applications can be installed, such as web browser is answered on terminal device 101,102,103
With, shopping class application, searching class application, instant messaging tools, social platform software etc..
Terminal device 101,102,103 can be hardware, be also possible to software.When terminal device 101,102,103 is hard
When part, it can be the various electronic equipments including display screen.For example including but be not limited to such as mobile phone, laptop,
Digit broadcasting receiver, PDA (personal digital assistant), PAD (tablet computer), PMP (portable media player), vehicle-mounted end
Hold the mobile terminal of (such as vehicle mounted guidance terminal) etc., and the fixed terminal of such as number TV, desktop computer etc..When
When terminal device 101,102,103 is software, it may be mounted in above-mentioned electronic equipment.Multiple softwares or soft may be implemented into it
Part module (such as providing the software of Distributed Services or software module), also may be implemented into single software or software mould
Block.It is not specifically limited herein.
Server 105 can be to provide the server of various services, such as to user's using terminal equipment 101,102,103
The background information processing server that the information of input is handled.Background information processing server can be to get, user
The information of input is handled, and processing result (such as matching image) is fed back to terminal device.
It should be noted that for pushing the method for image generally by server 105 provided by embodiment of the disclosure
It executes, correspondingly, the device for pushing image is generally positioned in server 105.
It should be noted that server can be hardware, it is also possible to software.When server is hardware, may be implemented
At the distributed server cluster that multiple servers form, individual server also may be implemented into.It, can when server is software
To be implemented as multiple softwares or software module (such as providing the software of Distributed Services or software module), also may be implemented
At single software or software module.It is not specifically limited herein.
It should be understood that the number of terminal device, network and server in Fig. 1 is only schematical.According to realization need
It wants, can have any number of terminal device, network and server.
With continued reference to Fig. 2, the process of one embodiment of the method for pushing image according to the disclosure is shown
200.The method for being used to push image, comprising the following steps:
Step 201, the information that user utilizes terminal input is obtained.
In the present embodiment, can lead to for pushing the executing subject (such as server shown in FIG. 1) of the method for image
It crosses the terminal (such as terminal device shown in FIG. 1) that wired connection mode or radio connection are used from user and obtains user's benefit
The information inputted with the terminal.Wherein, the information of above-mentioned input can include but is not limited to the information of following at least one type:
Text, image, video.
Step 202, relevant information relevant to the information that user inputs is determined.
In the present embodiment, above-mentioned executing subject can determine relevant information relevant to the information of user's input.Wherein,
Relevant information can include but is not limited to following at least one: text that user currently browses, image, video；User browsed
Text, image, video.The above-mentioned text currently browsed, image, video can be the target being pre-installed in terminal and answer
With (i.e. for user show matching image application), when it is preceding to user show text, image, video.Above-mentioned user is clear
Text, image, the video look at can be text, image, video that above-mentioned target application was once shown to user.
In some optional implementations of the present embodiment, above-mentioned executing subject can be determined in accordance with the following steps and be used
The relevant relevant information of information of family input:
In response to determining that the information of user's input is the comment information that user comments on the information currently browsed, determine
The information that user currently browses is relevant information relevant to the information that user inputs.Wherein, the information that user currently browses can
To include but is not limited to following at least one: text, image, video.In general, in the terminal that above-mentioned user uses, Ke Yixian
It is shown with comment information input field, user can use the comment information input field and input commenting for the information currently browsed for user
By information.Above-mentioned executing subject can be in response to detecting that user inputs comment information using comment information input field, in determination
Stating the information currently browsed is relevant information relevant to the information that user inputs.
Since the relevant information determined by this implementation is the information that user currently browses, use the correlation
When information and the information of user's input carry out images match, obtained matching image is related to the information that user currently browses,
Can image relevant to the information that user currently browses targetedly be shown to user.
In some optional implementations of the present embodiment, above-mentioned executing subject can be determined in accordance with the following steps and be used
The relevant relevant information of information of family input:
In response to determining that the information of user's input is to be sent to the information of the terminal of target user, browsed from user
In historical information, relevant information relevant to the information that user inputs is determined.Wherein, target user can be its use to be utilized
Terminal receive above-mentioned user input information user.In general, above-mentioned user can be used shown in terminal for mesh
Marking the terminal of user, to send the window of information (such as window for chatting with target user, private for sending to target user
The window etc. of letter) input information, to send information to the terminal of target user.Above-mentioned executing subject can be in response to detecting use
Family inputs information using the above-mentioned window for sending information, determines that the information of user's input is to be sent to the end of target user
The information at end.The historical information that above-mentioned user browsed can include but is not limited to following at least one: text, image, video.
In general, the browsing of the above-mentioned available above-mentioned user of executing subject records, the history for determining that user browsed is recorded according to browsing
Information.
Above-mentioned executing subject can be in various manners from the historical information that user browsed, determination and user's input
The relevant relevant information of information.For example, can be same type (such as being all science and technology) by the information currently browsed with user
Information be determined as relevant information.
In some optional implementations of the present embodiment, above-mentioned executing subject can in accordance with the following steps, from user
In the historical information browsed, information relevant to the information that user inputs is determined:
From the historical information that user browsed, determine the information of target user's publication as the information phase inputted with user
The information of pass.This implementation may be implemented to obtain matching image, matching image and target using the information that target user issues
The correlation degree of user is larger, thus more targetedly can show matching image relevant to target user to user, makes
User selects matching image and the information that matching image is inputted together with user is sent to the terminal of target user.
Step 203, the information and relevant information user inputted, the image for including with preset image library are matched,
Obtain the information inputted with user and the matching image to match with relevant information.
In the present embodiment, information and relevant information that above-mentioned executing subject can input user, with preset image
Library (image library can be set in above-mentioned executing subject, also can be set in communicate to connect with above-mentioned executing subject other
In electronic equipment) image that includes matched, it obtains and the information of user's input and the matching figure to match with relevant information
Picture.
Specifically, as an example, the image that image library includes can correspond respectively to description text, wherein description text
The content for including for characterizing image.When the information of user's input and relevant information include text, above-mentioned executing subject can
Merged with the text that the information and relevant information that input user include, using the method for existing determining text similarity, is divided
The similarity between obtained text and each description text, the phase for the preset similarity threshold that will be greater than or equal to Que Ding not merged
It is determined as matching image like corresponding image is spent.Alternatively, when one of information in the information and relevant information of user's input
When including text, the method that above-mentioned executing subject can use determining text similarity determines the text and each description respectively
The corresponding image of similarity of similarity between text, the preset similarity threshold that will be greater than or equal to is determined as matching image.
As another example, right when at least one information in the information and relevant information of user's input includes image
Each image in the image for including, the method that above-mentioned executing subject can use existing determining image similarity determine
The similarity between each image in the image and image library.By it is in image library, be more than or equal to preset similarity threshold
The corresponding image of similarity be determined as matching image.
Step 204, matching image is sent to terminal, so that matching image is displayed on the terminals.
In the present embodiment, matching image can be sent to terminal by above-mentioned executing subject, so that matching image is in terminal
Upper display.In general, user can click the matching image of display after matching image is displayed on the terminals, thus by matching image
Information as user's input.
With continued reference to the signal that Fig. 3, Fig. 3 are according to the application scenarios of the method for pushing image of the present embodiment
Figure.In the application scenarios of Fig. 3, server 301 obtains the comment information 303 that user utilizes terminal 302 to input first.Wherein,
Comment information 303 is the information that user comments on the article currently browsed, including text.Then, server 301 is by user
Relevant information 304 relevant to the comment information 303 that user inputs that the article currently browsed is determined as.Subsequently, server 301
Comment information 303 and relevant information 304 are matched with the image that preset image library 305 includes.For example, server 301
Text in text and relevant information 304 in comment information 303 is merged, the text after being merged.Preset image library
The method that the 305 each images for including utilize existing determining text similarity with corresponding description text, server 301,
Similarity between the corresponding description text of each image that text after determining above-mentioned merging and image library 306 include,
The corresponding image of similarity of similarity threshold of will be greater than or equal to is determined as matching image 306.Finally, server 301 will match
Image 306 is sent to terminal, so that matching image 306 is shown in terminal 302.User can click in terminal 302
With image 306, so as to the information for including as comment information 303 using matching image.
The method provided by the above embodiment of the disclosure utilizes the information of terminal input by obtaining user, determines and uses
The relevant relevant information of information of family input, the information and relevant information for then inputting user include with preset image library
Image matched, obtain matching image, matching image be finally sent to terminal, so that matching image is shown at the terminal
Show, to realize the information and relevant information for being flexibly utilized by user's input, obtains matching image, facilitate targetedly
Image is shown to user.In compared with the existing technology, the image and information-related, the disclosure of user's input that are shown to user
The available richer matching image of embodiment to be shown to user.The size of the display screen of the terminal used due to user
It is limited, by embodiment of the disclosure, limited display screen resource can be fully utilized, accurately shows image to user,
Need to show more image occupied display resource and storage money for selection by the user in the prior art so as to reduce
Source.
With further reference to Fig. 4, it illustrates the processes 400 of another embodiment of the method for pushing image.The use
In the process 400 of the method for push image, comprising the following steps:
Step 401, the information that user utilizes terminal input is obtained.
In the present embodiment, step 401 and the step 201 in Fig. 2 corresponding embodiment are almost the same, and which is not described herein again.
Step 402, relevant information relevant to the information that user inputs is determined.
In the present embodiment, step 402 and the step 202 in Fig. 2 corresponding embodiment are almost the same, and which is not described herein again.
Step 403, determine the feature vector of the information of user's input as first eigenvector, and determining relevant information
Feature vector as second feature vector.
It in the present embodiment, can be true for pushing the executing subject (such as server shown in FIG. 1) of the method for image
The feature vector of the information of user's input is determined as first eigenvector, and determines the feature vector of relevant information as second
Feature vector.
Specifically, first eigenvector and second feature vector may be respectively used for characterizing the letter of identical or different type
Breath.For example, first eigenvector can be Text eigenvector or image feature vector, second feature vector is also possible to text
Feature vector or image feature vector.Above-mentioned executing subject can by the text feature for the information text that include that user inputs to
Amount is determined as first eigenvector, or the image feature vector for the image that the information that user inputs includes is determined as fisrt feature
Vector.The Text eigenvector for the text that relevant information includes can be determined as second feature vector by above-mentioned executing subject, or
The image feature vector for the image that relevant information includes is determined as second feature vector.Wherein it is determined that Text eigenvector
Method can include but is not limited to following at least one: the method based on bag of words, the method based on word2vec model.Really
The method for determining image feature vector can include but is not limited to following at least one: neural network method, SIFT (Scale-
Invariant feature transform, Scale invariant features transform) method, SURF (Speeded Up Robust
Features accelerates robust feature) method.
It should be noted that above-mentioned image feature vector can be used for characterizing the feature of single image, table can be used for
The feature at least one video frame that sign video includes.
Step 404, it is based on first eigenvector and second feature vector, generates vector to be matched.
In the present embodiment, above-mentioned executing subject can be based on first eigenvector and second feature vector, generate to
With vector.
It specifically, as an example, when first eigenvector and second feature vector are Text eigenvector, or is figure
When as feature vector, first eigenvector and second feature addition of vectors (or can will be in identical bits by above-mentioned executing subject
The element multiplication set), obtain vector to be matched.
In some optional implementations of the present embodiment, the information and second feature vector of first eigenvector characterization
The type of the information of characterization is different.Such as in first eigenvector and second feature vector, one of them is Text eigenvector,
Another is image feature vector.Above-mentioned executing subject can generate in accordance with the following steps vector to be matched:
By first eigenvector and the input of second feature vector vector transformation model trained in advance, obtain it is to be matched to
Amount.Wherein, vector transformation model is used to input two vectors being converted to a vector.
As an example, above-mentioned vector transformation model can be preset formula, by the formula, can to fisrt feature to
Amount and second feature vector are calculated, to obtain vector to be matched.
As another example, above-mentioned vector transformation model can be based on preset training sample set, utilize engineering
The model that learning method obtains after being trained to initial model (such as convolutional neural networks, Recognition with Recurrent Neural Network etc.).Wherein, it instructs
Practicing sample may include vector after sample text feature vector, sample image feature vector, sample conversion.Turn for training vector
The executing subject of mold changing type can use machine learning method, the sample text for including by the training sample in training sample set
Feature vector and sample image feature vector as input, by with the sample text feature vector and sample image feature of input to
Vector is trained initial model as desired output after measuring corresponding sample conversion.For the sample of each training input
Text eigenvector and sample image feature vector, available reality output.Wherein, reality output is that initial model is actually defeated
Feature vector after conversion out.Then, back-propagation method, base can be used for the executing subject of training vector transformation model
In reality output and desired output, the parameter of initial model is adjusted, is instructed the model obtained after each adjusting parameter as next time
Experienced initial model, and terminate training in the case where meeting preset trained termination condition, so that training obtains vector conversion
Model.
It should be noted that the training termination condition here preset at can include but is not limited to it is at least one of following: training
Time is more than preset duration；Frequency of training is more than preset times；Using preset loss function (such as mean square error lose letter
Number, root-mean-square error loss function etc.) resulting penalty values are calculated less than default penalty values threshold value.
This optional implementation will be used to characterize the first of different types of information by using vector transformation model
Feature vector and second feature vector are converted to a vector to be matched, which can be with the figure in above-mentioned image library
The feature vector of picture is in same vector space, so as to by the feature vector of the image in vector sum image library to be matched into
Row matching.Since first eigenvector and second feature vector are respectively used to characterize different types of information, using originally may be used
The vector to be matched that the implementation of choosing obtains can more accurately characterize the feature of different types of information, to improve determination
The accuracy of matching image.
Step 405, for the image in preset image library, vector to be matched feature vector corresponding with the image is determined
Between similarity.
In the present embodiment, for the image in preset image library, above-mentioned executing subject can determine vector to be matched
Similarity between feature vector corresponding with the image.Wherein, the corresponding feature vector of the image can be to be directed in advance and be somebody's turn to do
Image carries out the obtained feature vector of feature extraction.
As an example, for the image in image library, the corresponding feature vector of the image can be image feature vector or
Text eigenvector.Wherein, image feature vector is for determining that the executing subject of the corresponding feature vector of image (such as executes
Main body or other electronic equipments) advancing with the method for determining image feature vector, to carry out feature extraction to the image obtained
Feature vector.Text eigenvector, which can be, determines text for determining that the executing subject of the corresponding feature vector of image advances with
The method of eigen vector description text corresponding to the image carries out the obtained feature vector of feature extraction.
Optionally, for the image in image library, for determining that the executing subject of the corresponding feature vector of image can make
With above-mentioned vector transformation model, it is corresponding that the corresponding image feature vector of the image and Text eigenvector are converted into the image
Feature vector.To which the corresponding feature vector of the image can be in identical vector space with above-mentioned vector to be matched.
In the present embodiment, the similarity between feature vector can use the distance between feature vector (such as cosine
Distance, Hamming distance etc.) characterization.
Step 406, by identified similarity, more than or equal to the corresponding image of similarity of preset similarity threshold
It is determined as matching image.
In the present embodiment, above-mentioned executing subject can be more than or equal to preset similarity in identified similarity
The corresponding image of the similarity of threshold value is determined as matching image.In general, feature vector to be matched feature corresponding with some image
Similarity between vector is bigger, and the information and relevant information and the degree of correlation of the image of expression user's input are bigger.To
The degree of correlation of identified matching image and information and relevant information that user inputs is larger.
Step 407, matching image is sent to terminal, so that matching image is displayed on the terminals.
In the present embodiment, step 407 and the step 204 in Fig. 2 corresponding embodiment are almost the same, and which is not described herein again.
Figure 4, it is seen that the method for pushing image compared with the corresponding embodiment of Fig. 2, in the present embodiment
Process 400 highlight first eigenvector, second feature vector utilized to generate vector to be matched, and utilize vector to be matched
The step of determining matching image.The vector to be matched in the scheme of the present embodiment description can more accurately characterize user as a result,
The information of input and the feature of relevant information help more have needle to user to improve the accuracy of determining matching image
Image is shown to property, further saves the display resource and storage resource of the terminal of user.
With further reference to Fig. 5, as the realization to method shown in above-mentioned each figure, present disclose provides one kind for pushing figure
One embodiment of the device of picture, the Installation practice is corresponding with embodiment of the method shown in Fig. 2, which can specifically answer
For in various electronic equipments.
As shown in figure 5, the present embodiment includes: acquiring unit 501 for pushing the device 500 of image, it is configured to obtain
Take the information that family utilizes terminal input；Determination unit 502 is configured to determine related letter relevant to the information of user's input
Breath；Matching unit 503, be configured to by user input information and relevant information, the image for including with preset image library into
Row matching, obtains the information inputted with user and the matching image to match with relevant information；Transmission unit 504, is configured to
Matching image is sent to terminal, so that matching image is displayed on the terminals.
In the present embodiment, acquiring unit 501 can be used by wired connection mode or radio connection from user
Terminal (such as terminal device shown in FIG. 1) obtain user utilize the terminal input information.Wherein, the information of above-mentioned input
It can include but is not limited to the information of following at least one type: text, image, video.
In the present embodiment, determination unit 502 can determine relevant information relevant to the information of user input.Its
In, the relevant information can include but is not limited to following at least one: text that user currently browses, image, video；User
Text, image, the video browsed.The above-mentioned text currently browsed, image, video, can be and be pre-installed in terminal
Target application (i.e. for showing the application of matching image to user), when the preceding text shown to user, image, video.It is above-mentioned
Text that user browsed, image, video, can be text, the image, view that above-mentioned target application was once shown to user
Frequently.
In the present embodiment, information and the relevant information that matching unit 503 can input the user, and it is default
Image library (image library can be set in above-mentioned apparatus 500, also can be set in communicating to connect with above-mentioned apparatus 500
In other electronic equipments) image that includes matched, obtain the information inputted with the user and with the relevant information phase
Matched matching image.
Specifically, as an example, the image that image library includes can correspond respectively to description text, wherein description text
The content for including for characterizing image.When the information of user's input and relevant information include text, above-mentioned matching unit 503
The information that user inputs and the text that relevant information include can be merged, using the method for existing determining text similarity,
The similarity merged between obtained text and each description text is determined respectively, and will be greater than or equal to preset similarity threshold
The corresponding image of similarity is determined as matching image.Alternatively, when one of letter in the information and relevant information of user's input
Breath include text when, the method that above-mentioned matching unit 503 can use determining text similarity, respectively determine the text with it is each
Similarity between text is described, the corresponding image of similarity for the preset similarity threshold that will be greater than or equal to is determined as matching figure
Picture.
As another example, right when at least one information in the information and relevant information of user's input includes image
Each image in the image for including, the method that above-mentioned matching unit 503 can use existing determining image similarity, really
The similarity between each image in the fixed image and image library.By it is in image library, be more than or equal to preset similarity threshold
The corresponding image of the similarity of value is determined as matching image.
In the present embodiment, the matching image can be sent to the terminal by transmission unit 504, so that the matching
Image is shown on the terminal.In general, user can click the matching image of display after matching image is displayed on the terminals,
Information to input matching image as user.
In some optional implementations of the present embodiment, determination unit 502 can be further configured to: in response to
The information for determining user's input is the comment information that user comments on the information currently browsed, determines what user currently browsed
Information is relevant information relevant to the information that user inputs.
In some optional implementations of the present embodiment, determination unit 502 can be further configured to: in response to
The information for determining user's input is to be sent to the information of the terminal of target user, from the historical information that user browsed, really
Fixed relevant information relevant to the information that user inputs.
In some optional implementations of the present embodiment, determination unit 502 can be further configured to: from user
In the historical information browsed, determine the information of target user's publication as information relevant to the information that user inputs.
In some optional implementations of the present embodiment, for the image in image library, which corresponds to default
Feature vector；And matching unit 503 may include: the first determining module (not shown), be configured to determine user
The feature vector of the information of input as first eigenvector, and determine the feature vector of relevant information as second feature to
Amount；Generation module (not shown), is configured to based on first eigenvector and second feature vector, generate it is to be matched to
Amount；Second determining module (not shown) is configured to determine vector to be matched and the image for the image in image library
Similarity between corresponding feature vector；Third determining module (not shown) is configured to identified similarity
In, the corresponding image of similarity more than or equal to preset similarity threshold is determined as matching image.
In some optional implementations of the present embodiment, the information and second feature vector of first eigenvector characterization
The type of the information of characterization is different；And generation module can be further configured to: by first eigenvector and second feature
Vector input vector transformation model trained in advance, obtains vector to be matched, wherein vector transformation model be used to input two
A vector is converted to a vector.
The device provided by the above embodiment of the disclosure utilizes the information of terminal input, determining and institute by obtaining user
State the relevant relevant information of information of user's input, the information for then inputting the user and the relevant information are and default
The image library image that includes matched, obtain matching image, the matching image be finally sent to the terminal so that
The matching image is shown on the terminal, to realize the information and relevant information for being flexibly utilized by user's input, is obtained
To matching image, help targetedly to show image to user.In compared with the existing technology, the image shown to user is only
With user input it is information-related, the available richer matching image of embodiment of the disclosure to user to show.Due to
The size of the display screen for the terminal that user uses is limited, by embodiment of the disclosure, can fully utilize limited display
Shield resource, accurately shows image to user, need to show more image for user in the prior art so as to reduce
Select occupied display resource and storage resource.
Below with reference to Fig. 6, it illustrates the structural representations for the server 600 for being suitable for being used to realize embodiment of the disclosure
Figure.Server shown in Fig. 6 is only an example, should not function to embodiment of the disclosure and use scope bring it is any
Limitation.
As shown in fig. 6, server 600 may include processing unit (such as central processing unit, graphics processor etc.) 601,
It can be loaded into random access storage according to the program being stored in read-only memory (ROM) 602 or from storage device 608
Program in device (RAM) 603 and execute various movements appropriate and processing.In RAM 603, it is also stored with the behaviour of server 600
Various programs and data needed for making.Processing unit 601, ROM 602 and RAM 603 are connected with each other by bus 604.It is defeated
Enter/export (I/O) interface 605 and is also connected to bus 604.
In general, following device can connect to I/O interface 605: including such as touch screen, touch tablet, keyboard, mouse, taking the photograph
As the input unit 606 of head, microphone, accelerometer, gyroscope etc.；Including such as liquid crystal display (LCD), loudspeaker, vibration
The output device 607 of dynamic device etc.；Storage device 608 including such as tape, hard disk etc.；And communication device 609.Communication device
609, which can permit server 600, is wirelessly or non-wirelessly communicated with other equipment to exchange data.Although Fig. 6, which is shown, to be had
The server 600 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.Each box shown in Fig. 6 can represent a device, can also be according to need
Represent multiple devices.
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 609, or from storage device 608
It is mounted, or is mounted from ROM 602.When the computer program is executed by processing unit 601, the implementation of the disclosure is executed
The above-mentioned function of being limited in the method for example.It should be noted that computer-readable medium described in embodiment of the disclosure can be with
It is computer-readable signal media or computer-readable medium either the two any combination.Computer-readable medium
Such as may be-but not limited to-system, device or the device of electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor, or
Any above combination.The more specific example of computer-readable medium can include but is not limited to: lead with one or more
The electrical connection of line, portable computer diskette, hard disk, random access storage device (RAM), read-only memory (ROM), erasable type can
Program read-only memory (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 embodiment of the disclosure, computer-readable medium can be any tangible medium for including or store program,
The program can be commanded execution system, device or device use or in connection.And in embodiment of the disclosure
In, computer-readable signal media may include in a base band or as the data-signal that carrier wave a part is propagated, wherein holding
Computer-readable program code is carried.The data-signal of this propagation can take various forms, including but not limited to electromagnetism
Signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be computer-readable medium with
Outer any computer-readable medium, the computer-readable signal media can be sent, propagated or transmitted for being held by instruction
Row system, device or device use or program in connection.The program code for including on computer-readable medium
It can transmit with any suitable medium, including but not limited to: electric wire, optical cable, RF (radio frequency) etc. or above-mentioned any conjunction
Suitable combination.
Above-mentioned computer-readable medium can be included in above-mentioned server；It is also possible to individualism, and without
It is incorporated in the server.Above-mentioned computer-readable medium carries one or more program, when said one or multiple journeys
When sequence is executed by the server, so that the server: obtaining the information that user utilizes terminal input；The determining letter with user's input
Cease relevant relevant information；The information and relevant information that user is inputted, the image for including with preset image library are matched,
Obtain the information inputted with user and the matching image to match with relevant information；Matching image is sent to terminal, so that
It is displayed on the terminals with image.
The behaviour for executing embodiment of the disclosure can be write with one or more programming languages or combinations thereof
The computer program code of work, described program design language include object oriented program language-such as Java,
Smalltalk, C++ further include conventional procedural programming language-such as " C " language or similar program design language
Speech.Program code can be executed fully on the user computer, partly be executed on the user computer, as an independence
Software package execute, part on the user computer part execute on the remote computer or completely in remote computer or
It is executed on server.In situations involving remote computers, remote computer can pass through the network of any kind --- packet
It includes local area network (LAN) or wide area network (WAN)-is connected to subscriber computer, or, it may be connected to outer computer (such as benefit
It is connected with ISP by internet).
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 embodiment of the disclosure can be realized by way of software, can also be passed through
The mode of hardware is realized.Described unit also can be set in the processor, for example, can be described as: a kind of processor
Including acquiring unit, determination unit, matching unit and transmission unit.Wherein, the title of these units is not under certain conditions
The restriction to the unit itself is constituted, for example, acquiring unit is also described as " obtaining the information that user utilizes terminal input
Unit ".
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 it should be appreciated that embodiment of the disclosure involved in invention scope, however it is not limited to the specific combination of above-mentioned technical characteristic and
At technical solution, while should also cover do not depart from foregoing invention design in the case where, by above-mentioned technical characteristic or its be equal
Feature carries out any combination and other technical solutions for being formed.Such as disclosed in features described above and embodiment of the disclosure (but
It is not limited to) technical characteristic with similar functions is replaced mutually and the technical solution that is formed.
Claims (14)
1. a kind of method for pushing image, comprising:
Obtain the information that user utilizes terminal input；
Determine relevant information relevant to the information that the user inputs；
The information and the relevant information that the user is inputted, the image for including with preset image library are matched, are obtained
The information inputted with the user and the matching image to match with the relevant information；
The matching image is sent to the terminal, so that the matching image is shown on the terminal.
2. according to the method described in claim 1, wherein, the determination related letter relevant to the information that the user inputs
Breath, comprising:
Information in response to determination user's input is the comment information that the user comments on the information currently browsed,
Determine that the information that user currently browses is relevant information relevant to the information that user inputs.
3. according to the method described in claim 1, wherein, the determination related letter relevant to the information that the user inputs
Breath, comprising:
Information in response to determination user's input is to be sent to the information of the terminal of target user, is browsed from the user
In the historical information crossed, relevant information relevant to the information that the user inputs is determined.
4. described from the historical information that the user browsed according to the method described in claim 3, wherein, determining and institute
State the relevant relevant information of information of user's input, comprising:
From the historical information that the user browsed, determine that the information of target user's publication is inputted as with the user
The relevant information of information.
5. method described in one of -4 according to claim 1, wherein for the image in described image library, which corresponds to
Preset feature vector；And
The information that the user is inputted and the relevant information, the image for including with preset image library are matched,
Obtain the information inputted with the user and the matching image to match with the relevant information, comprising:
Determine the feature vector of the information of user's input as first eigenvector, and the spy of the determining relevant information
Vector is levied as second feature vector；
Based on the first eigenvector and the second feature vector, vector to be matched is generated；
For the image in described image library, determine similar between the vector to be matched feature vector corresponding with the image
Degree；
By in identified similarity, the corresponding image of similarity more than or equal to preset similarity threshold is determined as matching figure
Picture.
6. according to the method described in claim 5, wherein, the information and the second feature of the first eigenvector characterization to
The type of the information of scale sign is different；And
It is described to be based on the first eigenvector and the second feature vector, generate vector to be matched, comprising:
By the first eigenvector and the second feature vector input vector transformation model trained in advance, obtain to be matched
Vector, wherein the vector transformation model is used to input two vectors being converted to a vector.
7. a kind of for pushing the device of image, comprising:
Acquiring unit is configured to obtain the information that user utilizes terminal input；
Determination unit is configured to determine relevant information relevant to the information of user input；
Matching unit, the information for being configured to input the user and the relevant information include with preset image library
Image is matched, and the information inputted with the user and the matching image to match with the relevant information are obtained；
Transmission unit is configured to the matching image being sent to the terminal, so that the matching image is in the terminal
Upper display.
8. device according to claim 7, wherein the determination unit is further configured to:
Information in response to determination user's input is the comment information that the user comments on the information currently browsed,
Determine that the information that user currently browses is relevant information relevant to the information that user inputs.
9. device according to claim 7, wherein the determination unit is further configured to:
Information in response to determination user's input is to be sent to the information of the terminal of target user, is browsed from the user
In the historical information crossed, relevant information relevant to the information that the user inputs is determined.
10. device according to claim 9, wherein the determination unit is further configured to:
From the historical information that the user browsed, determine that the information of target user's publication is inputted as with the user
The relevant information of information.
11. the device according to one of claim 7-10, wherein for the image in described image library, the image is corresponding
In preset feature vector；And
The matching unit includes:
First determining module, is configured to determine the feature vector of the information of user input as first eigenvector, with
And determine the feature vector of the relevant information as second feature vector；
Generation module is configured to generate vector to be matched based on the first eigenvector and the second feature vector；
Second determining module is configured to determine the vector to be matched and the image pair for the image in described image library
The similarity between feature vector answered；
Third determining module is configured in identified similarity, more than or equal to the similarity of preset similarity threshold
Corresponding image is determined as matching image.
12. device according to claim 11, wherein the information and the second feature of the first eigenvector characterization
The type of the information of vector characterization is different；And
Generation module is further configured to:
By the first eigenvector and the second feature vector input vector transformation model trained in advance, obtain to be matched
Vector, wherein the vector transformation model is used to input two vectors being converted to a vector.
13. a kind of server, comprising:
One or more processors；
Storage device is stored thereon with one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of processors are real
Now such as method as claimed in any one of claims 1 to 6.
14. a kind of computer-readable medium, is stored thereon with computer program, wherein the realization when program is executed by processor
Such as method as claimed in any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910159569.5A CN109918530A (en) | 2019-03-04 | 2019-03-04 | Method and apparatus for pushing image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910159569.5A CN109918530A (en) | 2019-03-04 | 2019-03-04 | Method and apparatus for pushing image |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109918530A true CN109918530A (en) | 2019-06-21 |
Family
ID=66963190
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910159569.5A Pending CN109918530A (en) | 2019-03-04 | 2019-03-04 | Method and apparatus for pushing image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109918530A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111310001A (en) * | 2020-03-16 | 2020-06-19 | 杨仙莲 | Cloud computing big data picture searching method and system based on block chain |
| CN112015936A (en) * | 2020-08-27 | 2020-12-01 | 北京字节跳动网络技术有限公司 | Method, device, electronic equipment and medium for generating article display diagram |
| CN112613293A (en) * | 2020-12-29 | 2021-04-06 | 北京中科闻歌科技股份有限公司 | Abstract generation method and device, electronic equipment and storage medium |
| CN112702260A (en) * | 2020-12-23 | 2021-04-23 | 维沃移动通信（杭州）有限公司 | Image sending method and device and electronic equipment |
| CN113837986A (en) * | 2020-12-15 | 2021-12-24 | 京东科技控股股份有限公司 | Method, apparatus, electronic device, and medium for recognizing tongue picture |
| WO2023206576A1 (en) * | 2022-04-29 | 2023-11-02 | 北京小米移动软件有限公司 | Image processing method and apparatus |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130159919A1 (en) * | 2011-12-19 | 2013-06-20 | Gabriel Leydon | Systems and Methods for Identifying and Suggesting Emoticons |
| CN104618222A (en) * | 2015-01-07 | 2015-05-13 | 腾讯科技（深圳）有限公司 | Method and device for matching expression image |
| CN106803090A (en) * | 2016-12-05 | 2017-06-06 | 中国银联股份有限公司 | A kind of image-recognizing method and device |
| CN108432190A (en) * | 2015-09-01 | 2018-08-21 | 三星电子株式会社 | Response message recommends method and its equipment |
| CN108446650A (en) * | 2018-03-27 | 2018-08-24 | 百度在线网络技术（北京）有限公司 | The method and apparatus of face for identification |
| CN109241323A (en) * | 2018-09-13 | 2019-01-18 | 掌阅科技股份有限公司 | The method for generating user's poster is commented on based on e-book and calculates equipment |
-
2019
- 2019-03-04 CN CN201910159569.5A patent/CN109918530A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20130159919A1 (en) * | 2011-12-19 | 2013-06-20 | Gabriel Leydon | Systems and Methods for Identifying and Suggesting Emoticons |
| CN104618222A (en) * | 2015-01-07 | 2015-05-13 | 腾讯科技（深圳）有限公司 | Method and device for matching expression image |
| CN108432190A (en) * | 2015-09-01 | 2018-08-21 | 三星电子株式会社 | Response message recommends method and its equipment |
| CN106803090A (en) * | 2016-12-05 | 2017-06-06 | 中国银联股份有限公司 | A kind of image-recognizing method and device |
| CN108446650A (en) * | 2018-03-27 | 2018-08-24 | 百度在线网络技术（北京）有限公司 | The method and apparatus of face for identification |
| CN109241323A (en) * | 2018-09-13 | 2019-01-18 | 掌阅科技股份有限公司 | The method for generating user's poster is commented on based on e-book and calculates equipment |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111310001A (en) * | 2020-03-16 | 2020-06-19 | 杨仙莲 | Cloud computing big data picture searching method and system based on block chain |
| CN112015936A (en) * | 2020-08-27 | 2020-12-01 | 北京字节跳动网络技术有限公司 | Method, device, electronic equipment and medium for generating article display diagram |
| CN112015936B (en) * | 2020-08-27 | 2021-10-26 | 北京字节跳动网络技术有限公司 | Method, device, electronic equipment and medium for generating article display diagram |
| CN113837986A (en) * | 2020-12-15 | 2021-12-24 | 京东科技控股股份有限公司 | Method, apparatus, electronic device, and medium for recognizing tongue picture |
| CN112702260A (en) * | 2020-12-23 | 2021-04-23 | 维沃移动通信（杭州）有限公司 | Image sending method and device and electronic equipment |
| CN112702260B (en) * | 2020-12-23 | 2022-08-05 | 维沃移动通信（杭州）有限公司 | Image sending method and device and electronic equipment |
| CN112613293A (en) * | 2020-12-29 | 2021-04-06 | 北京中科闻歌科技股份有限公司 | Abstract generation method and device, electronic equipment and storage medium |
| WO2023206576A1 (en) * | 2022-04-29 | 2023-11-02 | 北京小米移动软件有限公司 | Image processing method and apparatus |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109918530A (en) | Method and apparatus for pushing image | |
| CN110162670A (en) | Method and apparatus for generating expression packet | |
| CN108898185A (en) | Method and apparatus for generating image recognition model | |
| CN109993150A (en) | The method and apparatus at age for identification | |
| CN110288049A (en) | Method and apparatus for generating image recognition model | |
| CN110110811A (en) | Method and apparatus for training pattern, the method and apparatus for predictive information | |
| CN109981787B (en) | Method and device for displaying information | |
| CN108595628A (en) | Method and apparatus for pushed information | |
| CN109815365A (en) | Method and apparatus for handling video | |
| CN109857908A (en) | Method and apparatus for matching video | |
| CN110069191B (en) | Terminal-based image dragging deformation implementation method and device | |
| CN109919244A (en) | Method and apparatus for generating scene Recognition model | |
| CN110213614A (en) | The method and apparatus of key frame are extracted from video file | |
| CN109947989A (en) | Method and apparatus for handling video | |
| CN110059623A (en) | Method and apparatus for generating information | |
| CN109977839A (en) | Information processing method and device | |
| CN109934142A (en) | Method and apparatus for generating the feature vector of video | |
| CN109902446A (en) | Method and apparatus for generating information prediction model | |
| CN109933217A (en) | Method and apparatus for pushing sentence | |
| CN111897950A (en) | Method and apparatus for generating information | |
| CN110163170A (en) | The method and apparatus at age for identification | |
| CN110516099A (en) | Image processing method and device | |
| CN109862100A (en) | Method and apparatus for pushed information | |
| CN109117758A (en) | Method and apparatus for generating information | |
| CN109408748A (en) | Method and apparatus for handling information |
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 |