CN108122029A - A kind of recommendation method and device of camera special effect - Google Patents
A kind of recommendation method and device of camera special effect Download PDFInfo
- Publication number
- CN108122029A CN108122029A CN201711478934.6A CN201711478934A CN108122029A CN 108122029 A CN108122029 A CN 108122029A CN 201711478934 A CN201711478934 A CN 201711478934A CN 108122029 A CN108122029 A CN 108122029A
- Authority
- CN
- China
- Prior art keywords
- special effect
- visible layer
- camera special
- vector
- user
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/047—Probabilistic or stochastic networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/50—Network services
- H04L67/55—Push-based network services
Abstract
The invention discloses a kind of recommendation method of camera special effect, including：Receive the historical data that the user that client reports uses camera special effect；Based on the historical data, the input vector of the visible layer of preset model is constructed；The input vector of the visible layer is inputted into the preset model, obtains the predicted vector of the visible layer of the preset model；The element for meeting preset condition is selected from the predicted vector of the visible layer；The corresponding camera special effect of the element is recommended into the user.The present invention can effectively improve the recommendation effect of camera special effect, so the technical issues of solving camera App in the prior art, can not recommending camera special effect to user well, realizes the recommendation effect for improving camera special effect, improve the technique effect of product liveness.Meanwhile the invention also discloses a kind of recommendation apparatus of camera special effect.
Description
Technical field
The present invention relates to Internet technical field more particularly to a kind of recommendation method and devices of camera special effect.
Background technology
Nowadays, many camera App (Application, using) can be provided to user and shoot graceful, interesting figure
Piece and video, and by picture and video sharing to the ability of good friend.And this graceful, interesting shooting ability, concentrated reflection
On camera special effect (we are referred to as " sprouting face " special efficacy).Each camera special effect both provides a kind of specific shooting ability, than
Such as：The U.S. bat ability of the thin face of big eye grinds the U.S. bat ability of bark effect, the energy of 3D (Three Dimensional, three-dimensional) effect
Power, the ability of AR (Augmented Reality, augmented reality) effect, etc..
At present, camera App has added up to have devised and embodied more than thousands of kinds camera special effects, and still, user can not possibly use
To all camera special effects.It is to carry so how to recommend to user to best suit the camera special effect of its hobby in first time
The important means of the product vigor of high camera App.
However, different from the big product of the scale of construction, the scale of construction of camera App is generally smaller, and client end capacity is also insufficient, and causing can
It is extremely limited with the user characteristics of acquisition.So traditional recommendation method (such as：Recurrence, decision tree, taxonomic clustering, etc.),
It can not be applied well herein.
To sum up, current camera App when recommending camera special effect to user, ask by the poor technology of generally existing recommendation effect
Topic.
The content of the invention
In view of the above problems, it is proposed that the present invention overcomes the above problem in order to provide one kind or solves at least partly
State the recommendation method and device of the camera special effect of problem.
The first aspect of the invention provides a kind of recommendation method of camera special effect, including：
Receive the historical data that the user that client reports uses camera special effect；
Based on the historical data, the input vector of the visible layer of preset model is constructed；
The input vector of the visible layer is inputted into the preset model, obtains the visible layer of the preset model
Predicted vector；
Select the element for meeting preset condition from the predicted vector of the visible layer, and by the corresponding camera of the element
Special efficacy recommends the user.
Preferably, the historical data for receiving the user that client reports and using camera special effect, including：
The user that reception client reports is in the first history number of past preset time period internal trigger camera special effect
According to.
Preferably, the preset model, specially：
Limit Boltzmann machine RBM models.
Preferably, before the input vector by the visible layer is inputted into the preset model, further include：
Determine the parameter in the RBM models.
Preferably, the parameter determined in the RBM models, including：
Obtain second historical data of multiple other users in institute's preset time period internal trigger camera special effect；
Using second historical data as training sample, it is trained by artificial intelligence learning system, described in acquisition
Parameter in RBM models；Wherein, the parameter in the RBM models includes：The bias vector of hidden layer, institute in the RBM models
State the bias vector of visible layer in RBM models, the visible layer to the weight matrix of the hidden layer.
Preferably, the input vector by the visible layer is inputted into the preset model, obtains the default mould
The predicted vector of the visible layer of type, including：
The input vector of the visible layer is inputted into the RBM models, so that the RBM models can based on described in
See layer input vector calculate the RBM models hidden layer vector, and based on the hidden layer vector calculate described in can
See the predicted vector of layer.
Preferably, it is described that the element for meeting preset condition is selected from the predicted vector of the visible layer, including：
The camera that the user is deleted from the predicted vector of the visible layer in the preset time period internal trigger is special
Corresponding element is imitated, and the highest M element of value is selected from surplus element, M is positive integer.
Preferably, it is described that the corresponding camera special effect of the element is recommended into the user, including：
The startup icon of the corresponding camera special effect of the element is created, and by the startup icon shows in the client
Recommendation region in.
The second aspect of the invention based on same inventive concept, provides a kind of recommendation apparatus of camera special effect, bag
It includes：
Receiving unit, for receiving the historical data that the user that client reports uses camera special effect；
Structural unit for being based on the historical data, constructs the input vector of the visible layer of preset model；
Arithmetic element for inputting the input vector of the visible layer into the preset model, obtains described default
The predicted vector of the visible layer of model；
Recommendation unit, for selecting the element for meeting preset condition from the predicted vector of the visible layer, and by described in
The corresponding camera special effect of element recommends the user.
Preferably, the receiving unit, is specifically used for：
The user that reception client reports is in the first history number of past preset time period internal trigger camera special effect
According to.
Preferably, the preset model, specially：
Limit Boltzmann machine RBM models.
Preferably, the recommendation apparatus of the camera special effect, further includes：
Determination unit before being inputted for the input vector by the visible layer into the preset model, determines
Parameter in the RBM models.
Preferably, the determination unit, is specifically used for：
Obtain second historical data of multiple other users in institute's preset time period internal trigger camera special effect；By described second
Historical data is trained by artificial intelligence learning system as training sample, obtains the parameter in the RBM models；Its
In, the parameter in the RBM models includes：The bias vector of hidden layer in the RBM models, visible layer in the RBM models
Bias vector, the visible layer to the weight matrix of the hidden layer.
Preferably, the arithmetic element, is specifically used for：
The input vector of the visible layer is inputted into the RBM models, so that the RBM models can based on described in
See layer input vector calculate the RBM models hidden layer vector, and based on the hidden layer vector calculate described in can
See the predicted vector of layer.
Preferably, the recommendation unit, is specifically used for：
The camera that the user is deleted from the predicted vector of the visible layer in the preset time period internal trigger is special
Corresponding element is imitated, and the highest M element of value is selected from surplus element, M is positive integer.
Preferably, the recommendation unit, is specifically used for：
The startup icon of the corresponding camera special effect of the element is created, and by the startup icon shows in the client
Recommendation region in.
The third aspect of the invention based on same inventive concept, provides a kind of recommendation apparatus of camera special effect, including
Memory, processor and storage on a memory and the computer program that can run on a processor, the processor execution institute
Following steps are realized when stating program：
Receive the historical data that the user that client reports uses camera special effect；Based on the historical data, construction is default
The input vector of the visible layer of model；The input vector of the visible layer is inputted into the preset model, is obtained described pre-
If the predicted vector of the visible layer of model；The element for meeting preset condition is selected from the predicted vector of the visible layer, and will
The corresponding camera special effect of the element recommends the user.
The fourth aspect of the invention provides a kind of computer readable storage medium, the computer-readable storage medium
Computer program is stored in matter, which realizes following steps when being executed by processor：
Receive the historical data that the user that client reports uses camera special effect；Based on the historical data, construction is default
The input vector of the visible layer of model；The input vector of the visible layer is inputted into the preset model, is obtained described pre-
If the predicted vector of the visible layer of model；The element for meeting preset condition is selected from the predicted vector of the visible layer, and will
The corresponding camera special effect of the element recommends the user.
The technical solution provided in the embodiment of the present application, has at least the following technical effects or advantages：
A kind of recommendation method of camera special effect according to the present invention, including：It receives the user that client reports and uses camera
The historical data of special efficacy；Based on the historical data, the input vector of the visible layer of preset model is constructed；By the visible layer
Input vector is inputted into the preset model, obtains the predicted vector of the visible layer of the preset model；From the visible layer
Predicted vector in select the element for meeting preset condition, and the corresponding camera special effect of the element is recommended into the user.
The recommendation method of machine special efficacy in the present invention, can effectively improve the recommendation effect of camera special effect, so solve existing skill
Camera App in art, can not well to user recommend camera special effect the technical issues of, realize improve camera special effect recommendation effect
Fruit improves the technique effect of product liveness.
Above description is only the general introduction of technical solution of the present invention, in order to better understand the technological means of the present invention,
And can be practiced according to the content of specification, and in order to allow above and other objects of the present invention, feature and advantage can
It is clearer and more comprehensible, below the special specific embodiment for lifting the present invention.
Description of the drawings
By reading the detailed description of hereafter preferred embodiment, it is various other the advantages of and benefit it is common for this field
Technical staff will be apparent understanding.Attached drawing is only used for showing the purpose of preferred embodiment, and is not considered as to the present invention
Limitation.And throughout the drawings, the same reference numbers will be used to refer to the same parts.In the accompanying drawings：
Fig. 1 shows a kind of flow chart of the recommendation method of camera special effect according to an embodiment of the invention；
Fig. 2 shows RBM network structures according to an embodiment of the invention；
Fig. 3 shows a kind of structure chart of the recommendation apparatus of camera special effect according to an embodiment of the invention；
Fig. 4 shows a kind of structure chart of the recommendation apparatus of camera special effect according to an embodiment of the invention；
Fig. 5 shows a kind of structure chart of computer readable storage medium according to an embodiment of the invention.
Specific embodiment
An embodiment of the present invention provides a kind of recommendation method and device of camera special effect, to solve the camera of the prior art
App there is technical issues that recommendation effect when recommending camera special effect to user.
The exemplary embodiment of the disclosure is more fully described below with reference to accompanying drawings.Although the disclosure is shown in attached drawing
Exemplary embodiment, it being understood, however, that may be realized in various forms the disclosure without should be by embodiments set forth here
It is limited.On the contrary, these embodiments are provided to facilitate a more thoroughly understanding of the present invention, and can be by the scope of the present disclosure
Completely it is communicated to those skilled in the art.
Embodiment one
A kind of recommendation method of camera special effect is present embodiments provided, applied in server, which specifically refers to
The corresponding network servers of camera App (Application, using), for providing network service for camera App.Wherein, the phase
The client of machine App can be installed in mobile terminal (such as：Smart mobile phone or tablet computer, etc.) in, user can open
The client for moving camera App is taken pictures, meanwhile, it is provided in the client there are many camera special effect, user can use
These camera special effects, so as to obtain rich and varied effect of taking pictures.
Specifically, as shown in Figure 1, the recommendation method of the camera special effect, including：
Step S101：Receive the historical data that the user that client reports uses camera special effect.
In specific implementation process, in camera App clients provide there are many camera special effect (such as：The thin face of big eye
U.S. clap special efficacy, mill bark effect it is beautiful clap special efficacy, whitening effect it is beautiful clap special efficacy, the special efficacy of 3D effect, the special efficacy of AR effects, etc.
Deng), the startup figure of each camera special effect can be specifically shown on the UI (User Interface, user interface) of camera App
Mark, if user is wanted using a certain camera special effect, can trigger the camera special effect it is corresponding start icon (such as：Click is opened
Cardon mark), so as to start the camera special effect.
In specific implementation process, client can record camera special effect that user triggers every time (i.e.：User makes every time
Camera special effect), and by the identification information of the camera special effect (such as：The title of camera special effect or the ID of camera special effect) with
And the usage time of the camera special effect is stored in a storage region specified.It is default in the past one that client can collect user
In period (such as：In past one month or it is one week past in) triggering camera special effect historical data (i.e.：First goes through
History data), and the historical data is packaged into a data packet and is reported to server.
In specific implementation process, server opens its remote interface, for receiving one or more client remotes
The data packet of report.For the data packet that each client reports, server needs to parse it, to obtain it
In user use camera special effect historical data.As it was noted above, record has user pre- in the past one in the historical data
If in the period (such as：In past one month or it is one week past in) triggering camera special effect relevant information.
In specific implementation process, the substantial amounts of personal information of user can be stored in the terminal, such as：Gender, the age,
Constellation, blood group, home address, work unit, often activity region, etc., these personal information are enough the habit for embodying user
Used and preference, client can collect these personal information, and be packaged and be reported to server.Furthermore client can be based on using
Above-mentioned personal information is obtained in the web page browsing record at family and the short message of user's contact or mail or instant communication message.
In specific implementation process, server can receive the personal information for the user that client reports, can inclusive
Not, the age, constellation, blood group, home address, work unit, often activity region, etc., and based on these individual subscribers believe
Breath generates the portrait of the user, and the user, which draws a portrait, can reflect the preference of user, can be used as subsequent recommendation camera special effect
Reference frame.
Herein, after the historical data for receiving that client reports in server, you can perform step S102.
Step S102：Based on the historical data, the input vector of the visible layer of preset model is constructed.
As a kind of optional embodiment, the preset model is specially：RBM(Restricted Boltzmann
Machine limits Boltzmann machine) model.
In specific implementation process, server uses the historical data of camera special effect receiving the user that client reports
Afterwards, you can using RBM models, which is handled, obtains one or more camera special effects to be recommended.
Wherein, RBM models are a kind of available random neutral nets (stochastic neural network) to explain
Probability graph model (probabilistic graphical model), it is by this ambiguous base (PaulSmolensky) of Borrow Si
It was proposed in 1986 on the basis of BM (Boltzmann Machine, Boltzmann machine), so-called " random " refers to this network
In neuron Shi Sui Ji Zhong through member, output only has two states (un-activation, activation), generally with binary zero and 1 come table
Show, and the specific value of state is then determined according to probability statistics rule.With the rapid raising for counting grate machine meter grate ability and soon
The continuous development of fast grate method, RBM various correlation machines learn grate method in the sixth of the twelve Earthly Branches through becoming pratical and feasible.
As shown in Fig. 2, Fig. 2 gives a common RBM network structure, there are 2 node layers in RBM network structures, respectively
For visible layer v and hidden layer h, wherein, it is seen that layer has n visible node layers, and hidden layer has m hiding node layers, and visible
It is connected entirely between layer and hiding node layer, that is to say, that each visible node layer is only related to m hiding node layers, different
Visible node layer between independently of each other, it is seen that the state of node layer is only influenced by m hiding node layers；It is hidden for each
Node layer is also the same, is only influenced by n visible node layers.
In RBM networks, mainly there is 3 parameters, one be visible layer bias vector a, dimension is equal to visible layer
Node number, each dimension correspond to a node, a=(a1、a2、a3..., am)；One be hidden layer bias vector b,
Dimension is equal to the node number of hidden layer, and each dimension corresponds to a node, b=(b1、b2、b3..., bn)；There are one be
Visible layer is to the weight matrix W of hidden layer, and line number is visible layer number of nodes, and columns is hidden layer number of nodes；These parameters
Determine the encoding samples that a n is tieed up are tieed up sample by RBM networks into what kind of m.
In specific implementation process, can based on the user that client reports in one preset time period of past (such as：It crosses
In one month gone or it is one week past in) historical data of triggering camera special effect, generate the input vector v of visible layer.
For example, element of all camera special effects as visible layer in camera App can be chosen, user was in the past
The data vector of triggering behavior construction visible layer in one month.For example, it is assumed that a shared camera special effect (E1, E2, E3, E4,
E5, E6, E7, E8, E9, E10) it is used as visible node layer, then, it is seen that the input vector v=(0 11000000 0) of layer,
It represents that the user clicked on two camera special effects of E2 and E3 within past one month.Pay attention to：10 camera special effects herein are only
Be to illustrate for convenience, may there is thousands of a camera special effects in practical application, those skilled in the art after the present embodiment is read,
Sample size is easily expanded to any value, is not specifically limited herein for the sum of camera special effect.
Step S103：It will be seen that the input vector of layer is inputted into preset model, obtain preset model visible layer it is pre-
Direction finding amount.
As a kind of optional embodiment, before step S103, further include：
It determines the parameter in RBM models, specifically, multiple other users can be obtained and setting period internal trigger camera spy
Second historical data of effect；Using the second historical data as training sample, pass through artificial intelligence learning system (Tensorflow)
It is trained, obtains RBM Model Parameters.
In specific implementation process, first, server need to receive camera App a large number of users (such as：The whole network user)
In one preset time period of past (such as：In past one month or in past one week) trigger the historical data of camera special effect (i.e.：
Second historical data), record has which camera each user of the whole network all employs in the preset time period in the historical data
Special efficacy, this can reflect preference of the whole network user to camera special effect；Then, it is default at this based on each user in the whole network user
In period for camera spy to triggering behavior, generate a large amount of training samples；Finally, learn to instruct by Tensorflow
Practice, feature of the whole network user to the preference of camera special effect is arrived in study, so as to obtain RBM Model Parameters (i.e.：The biasing of visible layer
Vectorial a, the bias vector b of hidden layer and visible layer are to the weight matrix W of hidden layer).Wherein, TensorFlow is Google
Based on DistBelief (first generation artificial intelligence learning system of Google) research and development second generation artificial intelligence learning system, completely
It increases income, name is from the operation logic of itself.Tensor (tensor) means N-dimensional array, and Flow (stream) means to be based on
The calculating of data flow diagram, TensorFlow flow to other end calculating process for tensor from one end of flow graph.TensorFlow is
Complicated data structure is transmitted to the system that analysis and processing procedure are carried out in artificial intelligence nerve net.Herein, it is determined that RBM
Parameters in model have also determined that the operation rule of RBM models.
In specific implementation process, the number of nodes that can take hidden layer is default quantity, which can be：
30 or 40 or 50 or 60 or 70, etc..For the number of nodes of hidden layer, can flexibly be set according to actual conditions, this
Embodiment is not specifically limited.
Herein, after the input vector v of visible layer is obtained, and the parameters in RBM models are determined, you can perform
Step S103, i.e.,：It will be seen that the input vector of layer is inputted into the preset model, the prediction of the visible layer of preset model is obtained
Vector.
In specific implementation process, it will be seen that the input vector v of layer is inputted into RBM models, passes through RBM models
Computing, you can each element in the predicted vector v ' of one visible layer of acquisition, the predicted vector v ' of the visible layer represents one
Camera special effect, the value of each element can reflect preference of the user to the corresponding camera special effect of the element, and value is got over
Greatly, then it is higher to the preference of the camera special effect to represent user.
In specific implementation process, after the input vector v that will be seen that layer is inputted into RBM models, RBM models can be based on
The input vector v of visible layer calculates hidden layer vector h；Hidden layer vector h is based on again, calculates the predicted vector of visible layer
v’。
Specifically, the characteristic based on RBM networks in itself can overturn the input vector v of visible layer twice, first
Input vector v can be mapped on hidden layer by secondary overturning, obtain hidden layer vector h, and second of overturning can will hide layer vector h again
It is mapped in visible layer, obtains the predicted vector v ' of visible layer.
Specifically, equation below (1) can be based on, calculates hidden layer vector：
H=sigmoid (vW+b) --- (formula 1)
Wherein, v is the input vector of visible layer；H is hidden layer vector；B is the parameter in RBM models, for representing hidden
Hide the bias vector of layer.
Specifically, equation below can be based on, calculates the predicted vector of visible layer：
V '=sigmoid (hWT+ a) --- (formula 2)
Wherein, v ' is the predicted vector of visible layer；A is the parameter in RBM models, for representing being biased towards for visible layer
Amount；W is the parameter in RBM models, for representing visible layer to the weight matrix of hidden layer；
Wherein, sigmoid functions are the activation primitives in neutral net, are defined as：
The domain of sigmoid functions is (- ∞ ,+∞), and codomain is (0,1).
Herein, after the predicted vector v ' of visible layer is obtained, you can perform step S104.
Step S104：The element for meeting preset condition is selected from the predicted vector of visible layer, and the element is corresponded to
Camera special effect recommend user.
As a kind of optional embodiment, the element for meeting preset condition is selected in the predicted vector from visible layer,
Including：
Camera special effect corresponding element of the user in preset time period internal trigger is deleted from the predicted vector of visible layer,
And the highest M element of value is selected from surplus element, M is positive integer.
In specific implementation process, pass through the computing of RBM models, you can obtain the predicted vector v ' of a visible layer, this can
The each element seen in the predicted vector v ' of layer represents a camera special effect, and the value of each element can reflect user couple
The preference of the corresponding camera special effect of the element, value is bigger, then it is higher to the preference of the camera special effect to represent user.
Herein, can be removed in all elements in the predicted vector v ' of visible layer user in preset time period (such as：Past one
In a month or in the past in one week) the corresponding element of used camera special effect, for remaining element, according still further to value from greatly to
Small order sorts successively, and select exclude forward M element (such as：Can select highest preceding 5 elements of value or
Preceding 10 elements or preceding 50 elements or preceding 100 element or preceding 150 elements, etc.), the M element is M corresponding
Camera special effect is to need to recommend the camera special effect of user.Herein, for the value of M, can freely be set according to actual demand
It puts, the present embodiment is not specifically limited.
It is described that the corresponding camera special effect of the element is recommended into user as a kind of optional embodiment, including：
The startup icon of the corresponding camera special effect of the element is created, and recommended area of the icon shows in client will be started
In domain.
In specific implementation process, server, can be to client after determining to need M camera special effect recommended to the user
End sends one and recommends instruction, so as to which mobile client be controlled to recommend the corresponding icon shows that start of the M camera special effect
In region.Client then can create institute after recommendation instruction is received in the recommendation region on the UI interfaces of camera App
State the corresponding startup icon of M camera special effect.In this way, user when using camera App, easily has found to recommend these in region
Camera special effect, and use these camera special effects.
In specific implementation process, user's portrait of user, camera special effect corresponding to the M element are also based on
It is screened, further filters out the camera special effect for meeting user preference (for example, this can be filtered out according to the gender of user
The camera special effect of gender user preference alternatively, according to the age of user, filters out the special efficacy of the age bracket user preference), and will
The camera special effect finally filtered out recommends user according to the method described above.
The technical solution provided in the embodiment of the present application, has at least the following technical effects or advantages：
A kind of recommendation method of camera special effect according to the present invention, including：It receives the user that client reports and uses camera
The historical data of special efficacy；Based on the historical data, the input vector of the visible layer of preset model is constructed；By the visible layer
Input vector is inputted into the preset model, obtains the predicted vector of the visible layer of the preset model；From the visible layer
Predicted vector in select the element for meeting preset condition, and the corresponding camera special effect of the element is recommended into the user.
The recommendation method of machine special efficacy in the present invention, can effectively improve the recommendation effect of camera special effect, so solve existing skill
Camera App in art, can not well to user recommend camera special effect the technical issues of, realize improve camera special effect recommendation effect
Fruit improves the technique effect of product liveness.
Embodiment two
Based on same inventive concept, a kind of recommendation apparatus 200 of camera special effect is present embodiments provided, as shown in figure 3, wrapping
It includes：
Receiving unit 201, for receiving the historical data that the user that client reports uses camera special effect；
Structural unit 202 for being based on the historical data, constructs the input vector of the visible layer of preset model；
Arithmetic element 203 for inputting the input vector of the visible layer into the preset model, obtains described pre-
If the predicted vector of the visible layer of model；
Recommendation unit 204, for selecting the element for meeting preset condition from the predicted vector of the visible layer, and by institute
It states the corresponding camera special effect of element and recommends the user.
As a kind of optional embodiment, the receiving unit 201 is specifically used for：
The user that reception client reports is in the first history number of past preset time period internal trigger camera special effect
According to.
As a kind of optional embodiment, the preset model is specially：
Limit Boltzmann machine RBM models.
As a kind of optional embodiment, the recommendation apparatus of the camera special effect further includes：
Determination unit before being inputted for the input vector by the visible layer into the preset model, determines
Parameter in the RBM models.
As a kind of optional embodiment, the determination unit is specifically used for：
Obtain second historical data of multiple other users in institute's preset time period internal trigger camera special effect；By described second
Historical data is trained by artificial intelligence learning system as training sample, obtains the parameter in the RBM models；Its
In, the parameter in the RBM models includes：The bias vector of hidden layer in the RBM models, visible layer in the RBM models
Bias vector, the visible layer to the weight matrix of the hidden layer.
As a kind of optional embodiment, the arithmetic element is specifically used for：
The input vector of the visible layer is inputted into the RBM models, so that the RBM models can based on described in
See layer input vector calculate the RBM models hidden layer vector, and based on the hidden layer vector calculate described in can
See the predicted vector of layer.
As a kind of optional embodiment, the recommendation unit 204 is specifically used for：
The camera that the user is deleted from the predicted vector of the visible layer in the preset time period internal trigger is special
Corresponding element is imitated, and the highest M element of value is selected from surplus element, M is positive integer.
As a kind of optional embodiment, the recommendation unit 204 is specifically used for：
The startup icon of the corresponding camera special effect of the element is created, and by the startup icon shows in the client
Recommendation region in.
The recommendation apparatus for the camera special effect introduced by the present embodiment is implements camera special effect in the embodiment of the present invention
Device used by recommendation method, so the recommendation method based on the camera special effect described in the embodiment of the present invention, this field
Those of skill in the art can understand the specific embodiment and its various change of the recommendation apparatus of the camera special effect of the present embodiment
Form, so how to realize that the method in the embodiment of the present invention is no longer situated between in detail for the recommendation apparatus of the camera special effect at this
It continues.As long as device used by the recommendation method of camera special effect in those skilled in the art's implementation embodiment of the present invention, all
Belong to the scope of the invention to be protected.
Technical solution in the embodiments of the present invention, at least has the following technical effect that or advantage：
It is according to the present invention to present embodiments provide a kind of recommendation apparatus of camera special effect, including：Receiving unit, for connecing
Receive the historical data that the user that client reports uses camera special effect；Structural unit, for being based on the historical data, construction is pre-
If the input vector of the visible layer of model；Arithmetic element, for inputting the input vector of the visible layer to the default mould
In type, the predicted vector of the visible layer of the preset model is obtained；Recommendation unit, for from the predicted vector of the visible layer
The element for meeting preset condition is selected, and the corresponding camera special effect of the element is recommended into the user.Machine in the present invention
The recommendation method of special efficacy, can effectively improve the recommendation effect of camera special effect, so solve camera App in the prior art,
The technical issues of can not camera special effect being recommended to user well, the recommendation effect for improving camera special effect is realized, improve product
The technique effect of liveness.
Embodiment three
As shown in figure 4, based on same inventive concept, a kind of recommendation apparatus 300 of camera special effect is present embodiments provided, is wrapped
It includes memory 310, processor 320 and is stored in the computer program 311 that can be run on memory 310 and on the processor 320,
The processor 320 realizes following steps when performing described program 311：
Receive the historical data that the user that client reports uses camera special effect；Based on the historical data, construction is default
The input vector of the visible layer of model；The input vector of the visible layer is inputted into the preset model, is obtained described pre-
If the predicted vector of the visible layer of model；The element for meeting preset condition is selected from the predicted vector of the visible layer, and will
The corresponding camera special effect of the element recommends the user.
As a kind of optional embodiment, the historical data for receiving the user that client reports and using camera special effect,
Including：
The user that reception client reports is in the first history number of past preset time period internal trigger camera special effect
According to.
As a kind of optional embodiment, the preset model is specially：
Limit Boltzmann machine RBM models.
As a kind of optional embodiment, the input vector by the visible layer inputs into the preset model it
Before, it further includes：
Determine the parameter in the RBM models.
As a kind of optional embodiment, the parameter determined in the RBM models, including：
Obtain second historical data of multiple other users in institute's preset time period internal trigger camera special effect；
Using second historical data as training sample, it is trained by artificial intelligence learning system, described in acquisition
Parameter in RBM models；Wherein, the parameter in the RBM models includes：The bias vector of hidden layer, institute in the RBM models
State the bias vector of visible layer in RBM models, the visible layer to the weight matrix of the hidden layer.
As a kind of optional embodiment, the input vector by the visible layer is inputted into the preset model,
The predicted vector of the visible layer of the preset model is obtained, including：
The input vector of the visible layer is inputted into the RBM models, so that the RBM models can based on described in
See layer input vector calculate the RBM models hidden layer vector, and based on the hidden layer vector calculate described in can
See the predicted vector of layer.
It is described that the member for meeting preset condition is selected from the predicted vector of the visible layer as a kind of optional embodiment
Element, including：
The camera that the user is deleted from the predicted vector of the visible layer in the preset time period internal trigger is special
Corresponding element is imitated, and the highest M element of value is selected from surplus element, M is positive integer.
It is described that the corresponding camera special effect of the element is recommended into the user as a kind of optional embodiment, including：
The startup icon of the corresponding camera special effect of the element is created, and by the startup icon shows in the client
Recommendation region in.
In the embodiment of the present application, the recommendation effect of camera special effect can be effectively improved using RBM models, so solve
Camera App in the prior art, realizes raising camera special effect at the technical issues of can not recommending camera special effect to user well
Recommendation effect, improve product liveness technique effect.
Example IV
As shown in figure 5, based on same inventive concept, a kind of computer readable storage medium 400 is present embodiments provided,
On be stored with computer program 411, which realizes following steps when being executed by processor：
Receive the historical data that the user that client reports uses camera special effect；Based on the historical data, construction is default
The input vector of the visible layer of model；The input vector of the visible layer is inputted into the preset model, is obtained described pre-
If the predicted vector of the visible layer of model；The element for meeting preset condition is selected from the predicted vector of the visible layer, and will
The corresponding camera special effect of the element recommends the user.
As a kind of optional embodiment, the historical data for receiving the user that client reports and using camera special effect,
Including：
The user that reception client reports is in the first history number of past preset time period internal trigger camera special effect
According to.
As a kind of optional embodiment, the preset model is specially：
Limit Boltzmann machine RBM models.
As a kind of optional embodiment, the input vector by the visible layer inputs into the preset model it
Before, it further includes：
Determine the parameter in the RBM models.
As a kind of optional embodiment, the parameter determined in the RBM models, including：
Obtain second historical data of multiple other users in institute's preset time period internal trigger camera special effect；
Using second historical data as training sample, it is trained by artificial intelligence learning system, described in acquisition
Parameter in RBM models；Wherein, the parameter in the RBM models includes：The bias vector of hidden layer, institute in the RBM models
State the bias vector of visible layer in RBM models, the visible layer to the weight matrix of the hidden layer.
As a kind of optional embodiment, the input vector by the visible layer is inputted into the preset model,
The predicted vector of the visible layer of the preset model is obtained, including：
The input vector of the visible layer is inputted into the RBM models, so that the RBM models can based on described in
See layer input vector calculate the RBM models hidden layer vector, and based on the hidden layer vector calculate described in can
See the predicted vector of layer.
It is described that the member for meeting preset condition is selected from the predicted vector of the visible layer as a kind of optional embodiment
Element, including：
The camera that the user is deleted from the predicted vector of the visible layer in the preset time period internal trigger is special
Corresponding element is imitated, and the highest M element of value is selected from surplus element, M is positive integer.
It is described that the corresponding camera special effect of the element is recommended into the user as a kind of optional embodiment, including：
The startup icon of the corresponding camera special effect of the element is created, and by the startup icon shows in the client
Recommendation region in.
In the embodiment of the present application, the recommendation effect of camera special effect can be effectively improved using RBM models, particularly with
This smaller application of user's scale of construction of camera, recommendation effect are more notable.So it solves this for camera in the prior art
The smaller application of user's scale of construction, can not well to user recommend camera special effect the technical issues of, realize raising camera special effect
Recommendation effect, improve the technique effect of user activity.
Model and display be not inherently related to any certain computer, virtual system or miscellaneous equipment provided herein.
Various general-purpose systems can also be used together with teaching based on this.As described above, required by constructing this kind of system
Structure be obvious.In addition, the present invention is not also directed to any certain programmed language.It should be understood that it can utilize various
Programming language realizes the content of invention described herein, and the description done above to language-specific is to disclose this hair
Bright preferred forms.
In the specification provided in this place, numerous specific details are set forth.It is to be appreciated, however, that the implementation of the present invention
Example can be put into practice without these specific details.In some instances, well known method, structure is not been shown in detail
And technology, so as not to obscure the understanding of this description.
Similarly, it should be understood that in order to simplify the disclosure and help to understand one or more of each inventive aspect,
Above in the description of exemplary embodiment of the present invention, each feature of the invention is grouped together into single implementation sometimes
In example, figure or descriptions thereof.However, the method for the disclosure should be construed to reflect following intention：I.e. required guarantor
Shield the present invention claims the more features of feature than being expressly recited in each claim.It is more precisely, such as following
Claims reflect as, inventive aspect is all features less than single embodiment disclosed above.Therefore,
Thus the claims for following specific embodiment are expressly incorporated in the specific embodiment, wherein each claim is in itself
Separate embodiments all as the present invention.
Those skilled in the art, which are appreciated that, to carry out adaptively the module in the equipment in embodiment
Change and they are arranged in one or more equipment different from the embodiment.It can be the module or list in embodiment
Member or component be combined into a module or unit or component and can be divided into addition multiple submodule or subelement or
Sub-component.In addition at least some in such feature and/or process or unit exclude each other, it may be employed any
Combination is disclosed to all features disclosed in this specification (including adjoint claim, summary and attached drawing) and so to appoint
Where all processes or unit of method or equipment are combined.Unless expressly stated otherwise, this specification is (including adjoint power
Profit requirement, summary and attached drawing) disclosed in each feature can be by providing the alternative features of identical, equivalent or similar purpose come generation
It replaces.
In addition, it will be appreciated by those of skill in the art that although some embodiments in this include institute in other embodiments
Including some features rather than other feature, but the combination of the feature of different embodiment means in the scope of the present invention
Within and form different embodiments.For example, in the following claims, embodiment claimed it is arbitrary it
One mode can use in any combination.
The all parts embodiment of the present invention can be with hardware realization or to be run on one or more processor
Software module realize or realized with combination thereof.It will be understood by those of skill in the art that it can use in practice
Microprocessor or digital signal processor (DSP) realize a kind of recommendation apparatus of camera special effect according to embodiments of the present invention
In some or all components some or all functions.The present invention is also implemented as performing as described herein
The some or all equipment or program of device (for example, computer program and computer program product) of method.So
Realization the present invention program can may be stored on the computer-readable medium or can have one or more signal shape
Formula.Such signal can be downloaded from internet website to be obtained either providing or with any other shape on carrier signal
Formula provides.
It should be noted that the present invention will be described rather than limits the invention for above-described embodiment, and ability
Field technique personnel can design alternative embodiment without departing from the scope of the appended claims.In the claims,
Any reference symbol between bracket should not be configured to limitations on claims.Word "comprising" does not exclude the presence of not
Element or step listed in the claims.Word "a" or "an" before element does not exclude the presence of multiple such
Element.The present invention can be by means of including the hardware of several different elements and being come by means of properly programmed computer real
It is existing.If in the unit claim for listing equipment for drying, several in these devices can be by same hardware branch
To embody.The use of word first, second, and third does not indicate that any order.These words can be explained and run after fame
Claim.
The invention discloses A1, a kind of recommendation methods of camera special effect, which is characterized in that including：
Receive the historical data that the user that client reports uses camera special effect；
Based on the historical data, the input vector of the visible layer of preset model is constructed；
The input vector of the visible layer is inputted into the preset model, obtains the visible layer of the preset model
Predicted vector；
Select the element for meeting preset condition from the predicted vector of the visible layer, and by the corresponding camera of the element
Special efficacy recommends the user.
The recommendation method of A2, camera special effect as described in A1, which is characterized in that described to receive the user that client reports and make
With the historical data of camera special effect, including：
The user that reception client reports is in the first history number of past preset time period internal trigger camera special effect
According to.
The recommendation method of A3, camera special effect as described in A1, which is characterized in that the preset model is specially：
Limit Boltzmann machine RBM models.
The recommendation method of A4, camera special effect as described in A3, which is characterized in that the input vector by the visible layer
Before input is into the preset model, further include：
Determine the parameter in the RBM models.
The recommendation method of A5, camera special effect as described in A4, which is characterized in that the ginseng determined in the RBM models
Number, including：
Obtain second historical data of multiple other users in institute's preset time period internal trigger camera special effect；
Using second historical data as training sample, it is trained by artificial intelligence learning system, described in acquisition
Parameter in RBM models；Wherein, the parameter in the RBM models includes：The bias vector of hidden layer, institute in the RBM models
State the bias vector of visible layer in RBM models, the visible layer to the weight matrix of the hidden layer.
The recommendation method of A6, camera special effect as described in A3, which is characterized in that the input vector by the visible layer
Input obtains the predicted vector of the visible layer of the preset model into the preset model, including：
The input vector of the visible layer is inputted into the RBM models, so that the RBM models can based on described in
See layer input vector calculate the RBM models hidden layer vector, and based on the hidden layer vector calculate described in can
See the predicted vector of layer.
The recommendation method of A7, camera special effect as described in A3, which is characterized in that the predicted vector from the visible layer
In select the element for meeting preset condition, including：
The camera that the user is deleted from the predicted vector of the visible layer in the preset time period internal trigger is special
Corresponding element is imitated, and the highest M element of value is selected from surplus element, M is positive integer.
The recommendation method of A8, camera special effect as described in A1~A7 is any, which is characterized in that described to correspond to the element
Camera special effect recommend the user, including：
The startup icon of the corresponding camera special effect of the element is created, and by the startup icon shows in the client
Recommendation region in.
B9, a kind of recommendation apparatus of camera special effect, which is characterized in that including：
Receiving unit, for receiving the historical data that the user that client reports uses camera special effect；
Structural unit for being based on the historical data, constructs the input vector of the visible layer of preset model；
Arithmetic element for inputting the input vector of the visible layer into the preset model, obtains described default
The predicted vector of the visible layer of model；
Recommendation unit, for selecting the element for meeting preset condition from the predicted vector of the visible layer, and by described in
The corresponding camera special effect of element recommends the user.
The recommendation apparatus of B10, camera special effect as described in B9, which is characterized in that the receiving unit is specifically used for：
The user that reception client reports is in the first history number of past preset time period internal trigger camera special effect
According to.
The recommendation apparatus of B11, camera special effect as described in B9, which is characterized in that the preset model is specially：
Limit Boltzmann machine RBM models.
The recommendation apparatus of B12, camera special effect as described in B11, which is characterized in that further include：
Determination unit before being inputted for the input vector by the visible layer into the preset model, determines
Parameter in the RBM models.
The recommendation apparatus of B13, camera special effect as described in B12, which is characterized in that the determination unit is specifically used for：
Obtain second historical data of multiple other users in institute's preset time period internal trigger camera special effect；By described second
Historical data is trained by artificial intelligence learning system as training sample, obtains the parameter in the RBM models；Its
In, the parameter in the RBM models includes：The bias vector of hidden layer in the RBM models, visible layer in the RBM models
Bias vector, the visible layer to the weight matrix of the hidden layer.
The recommendation apparatus of B14, camera special effect as described in B11, which is characterized in that the arithmetic element is specifically used for：
The input vector of the visible layer is inputted into the RBM models, so that the RBM models can based on described in
See layer input vector calculate the RBM models hidden layer vector, and based on the hidden layer vector calculate described in can
See the predicted vector of layer.
The recommendation apparatus of B15, camera special effect as described in B11, which is characterized in that the recommendation unit is specifically used for：
The camera that the user is deleted from the predicted vector of the visible layer in the preset time period internal trigger is special
Corresponding element is imitated, and the highest M element of value is selected from surplus element, M is positive integer.
The recommendation apparatus of B16, camera special effect as described in B9~B15 is any, which is characterized in that the recommendation unit, tool
Body is used for：
The startup icon of the corresponding camera special effect of the element is created, and by the startup icon shows in the client
Recommendation region in.
C17, a kind of recommendation apparatus of camera special effect, which is characterized in that including memory, processor and be stored in memory
Computer program that is upper and can running on a processor, the processor realize following steps when performing described program：
Receive the historical data that the user that client reports uses camera special effect；Based on the historical data, construction is default
The input vector of the visible layer of model；The input vector of the visible layer is inputted into the preset model, is obtained described pre-
If the predicted vector of the visible layer of model；The element for meeting preset condition is selected from the predicted vector of the visible layer, and will
The corresponding camera special effect of the element recommends the user.
D18, a kind of computer readable storage medium, which is characterized in that be stored on the computer readable storage medium
Computer program, the program realize following steps when being executed by processor：
Receive the historical data that the user that client reports uses camera special effect；Based on the historical data, construction is default
The input vector of the visible layer of model；The input vector of the visible layer is inputted into the preset model, is obtained described pre-
If the predicted vector of the visible layer of model；The element for meeting preset condition is selected from the predicted vector of the visible layer, and will
The corresponding camera special effect of the element recommends the user.
Claims (10)
1. a kind of recommendation method of camera special effect, which is characterized in that including：
Receive the historical data that the user that client reports uses camera special effect；
Based on the historical data, the input vector of the visible layer of preset model is constructed；
The input vector of the visible layer is inputted into the preset model, obtains the prediction of the visible layer of the preset model
Vector；
Select the element for meeting preset condition from the predicted vector of the visible layer, and by the corresponding camera special effect of the element
Recommend the user.
2. the recommendation method of camera special effect as described in claim 1, which is characterized in that the user for receiving client and reporting
Using the historical data of camera special effect, including：
The user that reception client reports is in the first historical data of past preset time period internal trigger camera special effect.
3. the recommendation method of camera special effect as described in claim 1, which is characterized in that the preset model, specially：
Limit Boltzmann machine RBM models.
4. the recommendation method of camera special effect as claimed in claim 3, which is characterized in that the input by the visible layer to
Before amount input is into the preset model, further include：
Determine the parameter in the RBM models.
5. the recommendation method of camera special effect as claimed in claim 4, which is characterized in that described to determine in the RBM models
Parameter, including：
Obtain second historical data of multiple other users in institute's preset time period internal trigger camera special effect；
Using second historical data as training sample, it is trained by artificial intelligence learning system, obtains the RBM moulds
Parameter in type；Wherein, the parameter in the RBM models includes：The bias vector of hidden layer, the RBM in the RBM models
The bias vector of visible layer in model, the visible layer to the weight matrix of the hidden layer.
6. the recommendation method of camera special effect as claimed in claim 3, which is characterized in that the input by the visible layer to
Amount input obtains the predicted vector of the visible layer of the preset model into the preset model, including：
The input vector of the visible layer is inputted into the RBM models, so that the RBM models are based on the visible layer
Input vector calculate the hidden layer vectors of the RBM models, and the visible layer is calculated based on the hidden layer vector
Predicted vector.
7. the recommendation method of camera special effect as claimed in claim 3, which is characterized in that the pre- direction finding from the visible layer
The element for meeting preset condition is selected in amount, including：
Camera special effect pair of the user in the preset time period internal trigger is deleted from the predicted vector of the visible layer
The element answered, and the highest M element of value is selected from surplus element, M is positive integer.
8. a kind of recommendation apparatus of camera special effect, which is characterized in that including：
Receiving unit, for receiving the historical data that the user that client reports uses camera special effect；
Structural unit for being based on the historical data, constructs the input vector of the visible layer of preset model；
Arithmetic element for inputting the input vector of the visible layer into the preset model, obtains the preset model
Visible layer predicted vector；
Recommendation unit, for selecting the element for meeting preset condition from the predicted vector of the visible layer, and by the element
Corresponding camera special effect recommends the user.
9. a kind of recommendation apparatus of camera special effect, which is characterized in that including memory, processor and storage on a memory and can
The computer program run on a processor, the processor realize following steps when performing described program：
Receive the historical data that the user that client reports uses camera special effect；Based on the historical data, preset model is constructed
Visible layer input vector；The input vector of the visible layer is inputted into the preset model, obtains the default mould
The predicted vector of the visible layer of type；The element for meeting preset condition is selected from the predicted vector of the visible layer, and by described in
The corresponding camera special effect of element recommends the user.
10. a kind of computer readable storage medium, which is characterized in that be stored with computer on the computer readable storage medium
Program, the program realize following steps when being executed by processor：
Receive the historical data that the user that client reports uses camera special effect；Based on the historical data, preset model is constructed
Visible layer input vector；The input vector of the visible layer is inputted into the preset model, obtains the default mould
The predicted vector of the visible layer of type；The element for meeting preset condition is selected from the predicted vector of the visible layer, and by described in
The corresponding camera special effect of element recommends the user.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711478934.6A CN108122029A (en) | 2017-12-29 | 2017-12-29 | A kind of recommendation method and device of camera special effect |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711478934.6A CN108122029A (en) | 2017-12-29 | 2017-12-29 | A kind of recommendation method and device of camera special effect |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108122029A true CN108122029A (en) | 2018-06-05 |
Family
ID=62232394
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711478934.6A Pending CN108122029A (en) | 2017-12-29 | 2017-12-29 | A kind of recommendation method and device of camera special effect |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108122029A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108846479A (en) * | 2018-07-13 | 2018-11-20 | 河海大学 | The training method and device of recommended method, RBM model based on RBM model |
| CN108848309A (en) * | 2018-07-13 | 2018-11-20 | 维沃移动通信有限公司 | A kind of camera programm starting method and mobile terminal |
| CN111199540A (en) * | 2019-12-27 | 2020-05-26 | Oppo广东移动通信有限公司 | Image quality evaluation method, image quality evaluation device, electronic device, and storage medium |
| EP3846439A4 (en) * | 2018-10-15 | 2021-09-15 | Huawei Technologies Co., Ltd. | Intelligent photographing method and system, and related device |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104715236A (en) * | 2015-03-06 | 2015-06-17 | 广东欧珀移动通信有限公司 | Face beautifying photographing method and device |
| CN105677780A (en) * | 2014-12-31 | 2016-06-15 | Tcl集团股份有限公司 | Scalable user intent mining method and system thereof |
| CN106231434A (en) * | 2016-07-25 | 2016-12-14 | 武汉斗鱼网络科技有限公司 | A kind of living broadcast interactive specially good effect realization method and system based on Face datection |
| CN106302124A (en) * | 2016-08-18 | 2017-01-04 | 北京奇虎科技有限公司 | A kind of method adding specially good effect and electronic equipment |
| CN106303354A (en) * | 2016-08-18 | 2017-01-04 | 北京奇虎科技有限公司 | A kind of face specially good effect recommends method and electronic equipment |
| CN106294831A (en) * | 2016-08-18 | 2017-01-04 | 北京奇虎科技有限公司 | A kind of information recommendation method and electronic equipment |
| CN106453055A (en) * | 2016-10-28 | 2017-02-22 | 努比亚技术有限公司 | Method and apparatus for pushing information through user behaviors, and terminal |
-
2017
- 2017-12-29 CN CN201711478934.6A patent/CN108122029A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105677780A (en) * | 2014-12-31 | 2016-06-15 | Tcl集团股份有限公司 | Scalable user intent mining method and system thereof |
| CN104715236A (en) * | 2015-03-06 | 2015-06-17 | 广东欧珀移动通信有限公司 | Face beautifying photographing method and device |
| CN106231434A (en) * | 2016-07-25 | 2016-12-14 | 武汉斗鱼网络科技有限公司 | A kind of living broadcast interactive specially good effect realization method and system based on Face datection |
| CN106302124A (en) * | 2016-08-18 | 2017-01-04 | 北京奇虎科技有限公司 | A kind of method adding specially good effect and electronic equipment |
| CN106303354A (en) * | 2016-08-18 | 2017-01-04 | 北京奇虎科技有限公司 | A kind of face specially good effect recommends method and electronic equipment |
| CN106294831A (en) * | 2016-08-18 | 2017-01-04 | 北京奇虎科技有限公司 | A kind of information recommendation method and electronic equipment |
| CN106453055A (en) * | 2016-10-28 | 2017-02-22 | 努比亚技术有限公司 | Method and apparatus for pushing information through user behaviors, and terminal |
Non-Patent Citations (3)
| Title |
|---|
| 崔心怡: ""美图系列"APP推广策略研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 李莹: "基于Android平台的相机特效模式的研究与实现", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 许倩婷: "美图类APP的使用对用户自我认知的影响", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108846479A (en) * | 2018-07-13 | 2018-11-20 | 河海大学 | The training method and device of recommended method, RBM model based on RBM model |
| CN108848309A (en) * | 2018-07-13 | 2018-11-20 | 维沃移动通信有限公司 | A kind of camera programm starting method and mobile terminal |
| CN108848309B (en) * | 2018-07-13 | 2020-08-07 | 维沃移动通信有限公司 | Camera program starting method and mobile terminal |
| EP3846439A4 (en) * | 2018-10-15 | 2021-09-15 | Huawei Technologies Co., Ltd. | Intelligent photographing method and system, and related device |
| US11470246B2 (en) | 2018-10-15 | 2022-10-11 | Huawei Technologies Co., Ltd. | Intelligent photographing method and system, and related apparatus |
| CN111199540A (en) * | 2019-12-27 | 2020-05-26 | Oppo广东移动通信有限公司 | Image quality evaluation method, image quality evaluation device, electronic device, and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108345937B (en) | Circulation is merged with library | |
| US20210256403A1 (en) | Recommendation method and apparatus | |
| CN108052670A (en) | A kind of recommendation method and device of camera special effect | |
| Bankes | Tools and techniques for developing policies for complex and uncertain systems | |
| CN110046304A (en) | A kind of user's recommended method and device | |
| CN105956161B (en) | A kind of information recommendation method and device | |
| CN108122029A (en) | A kind of recommendation method and device of camera special effect | |
| CN109902708A (en) | A kind of recommended models training method and relevant apparatus | |
| CN108431833A (en) | End-to-end depth collaborative filtering | |
| CN110378434A (en) | Training method, recommended method, device and the electronic equipment of clicking rate prediction model | |
| CN106250403A (en) | Customer loss Forecasting Methodology and device | |
| CN107704485A (en) | A kind of position recommends method and computing device | |
| CN109960761A (en) | Information recommendation method, device, equipment and computer readable storage medium | |
| US20230244938A1 (en) | Using Chains of Thought to Prompt Machine-Learned Models Pre-Trained on Diversified Objectives | |
| CN109155005A (en) | Learnt using the enhancing of spurious count | |
| CN102486802A (en) | Attraction-based data visualization | |
| CN112100221B (en) | Information recommendation method and device, recommendation server and storage medium | |
| CN108595493A (en) | Method for pushing and device, storage medium, the electronic device of media content | |
| CN110413888A (en) | A kind of books recommended method and device | |
| CN106445954A (en) | Business object display method and apparatus | |
| Ge et al. | Neural-sim: Learning to generate training data with nerf | |
| CN110235149A (en) | Neural plot control | |
| CN109670927A (en) | The method of adjustment and its device of credit line, equipment, storage medium | |
| CN106227881A (en) | A kind of information processing method and server | |
| CN113726545B (en) | Network traffic generation method and device for generating countermeasure network based on knowledge enhancement |
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: 20180605 |