CN109523344A - Product information recommended method, device, computer equipment and storage medium - Google Patents
Product information recommended method, device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN109523344A CN109523344A CN201811205259.4A CN201811205259A CN109523344A CN 109523344 A CN109523344 A CN 109523344A CN 201811205259 A CN201811205259 A CN 201811205259A CN 109523344 A CN109523344 A CN 109523344A
- Authority
- CN
- China
- Prior art keywords
- video
- feature
- product information
- similar users
- point set
- 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
- 238000000034 method Methods 0.000 title claims abstract description 42
- 238000003860 storage Methods 0.000 title claims abstract description 14
- 239000000463 material Substances 0.000 claims abstract description 47
- 239000000284 extract Substances 0.000 claims abstract description 15
- 238000004590 computer program Methods 0.000 claims description 25
- 238000012795 verification Methods 0.000 claims description 20
- 238000000605 extraction Methods 0.000 claims description 8
- 238000013473 artificial intelligence Methods 0.000 abstract description 4
- 239000011521 glass Substances 0.000 description 15
- 230000006399 behavior Effects 0.000 description 6
- 230000003542 behavioural effect Effects 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 238000012790 confirmation Methods 0.000 description 3
- 230000001815 facial effect Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000012512 characterization method Methods 0.000 description 2
- 239000003086 colorant Substances 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 241000406668 Loxodonta cyclotis Species 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000003550 marker Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000001755 vocal effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/06—Buying, selling or leasing transactions
- G06Q30/0601—Electronic shopping [e-shopping]
- G06Q30/0631—Item recommendations
Landscapes
- Business, Economics & Management (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Development Economics (AREA)
- Economics (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
This application involves a kind of product information recommended method, device, computer equipment and storage mediums.It is related to artificial intelligence field, which comprises receive video；It identifies the who object in the video, and extracts the corresponding picture material of the who object；The feature point set of the who object is extracted from described image content, wherein the feature point set includes the corresponding various dimensions feature point set of multiple characteristic dimensions；Obtain the feature portrait of pre-stored user；The feature point set and feature portrait are subjected to fractional dimension characteristic matching, determine the corresponding similar users of the who object；The preferred product information of the similar users is searched, the product information is pushed into requesting terminal.Accurately product information can be carried out to unknown object any in crowd using this method to push.
Description
Technical field
This application involves field of artificial intelligence, more particularly to a kind of product information recommended method, device, computer
Equipment and storage medium.
Background technique
With the development of big data processing technique, based on the artificial intelligence that big data derives service to we work and
Life brings great convenience.If personalized recommendation is exactly the artificial intelligence service of one of comparative maturity.
Traditional personalized recommendation is achieved in that the user's big data for relying on applications client collection, is based on user
Big data carries out intellectual product recommendation to user.Traditional personalized recommendation is completed based on the applications client on line
's.And descended under Products Show environment online, above-mentioned personalized recommendation method then cannot achieve is not knowing that other side is any completely
Accurate Products Show is carried out in the case where information.Therefore, now there is an urgent need to seek a kind of ring that can descend Products Show online
Under border, scheme that unknown object any in crowd is precisely recommended.
Summary of the invention
Based on this, it is necessary to which in view of the above technical problems, essence can be carried out to unknown object any in crowd by providing one kind
Product information method for pushing, device, computer equipment and the storage medium of quasi- product information push.
A kind of product information recommended method, which comprises
Receive video；
It identifies the who object in the video, and extracts the corresponding picture material of the who object；
The feature point set of the who object is extracted from described image content, wherein the feature point set includes multiple
The corresponding various dimensions feature point set of characteristic dimension；
Obtain the feature portrait of pre-stored user；
The feature point set and feature portrait are subjected to fractional dimension characteristic matching, determine that the who object is corresponding
Similar users；
The preferred product information of the similar users is searched, the product information is pushed into requesting terminal.
In one embodiment, the who object in the identification video, and it is corresponding to extract the who object
Picture material, comprising:
Identify the who object in video；
The corresponding target image frame of the who object is determined from the video, wherein the target image frame includes band
There are all picture frames of the who object feature of the who object；
The who object feature is extracted from the target image frame；
The corresponding picture material of the who object is synthesized according to the who object feature.
In one embodiment, which comprises
The user information of the similar users is sent to requesting terminal, the requesting terminal shows the user information；
Voice messaging is received, recognition and verification result is obtained according to the voice messaging；
Obtain information recommendation algorithm corresponding with the recognition and verification result；
The product information that the similar users preference is determined according to the information recommendation algorithm, the product information is pushed
To requesting terminal.
In one embodiment, the who object in the identification video, and it is corresponding to extract the who object
Picture material, comprising:
Obtain the corresponding video recording mode of the video；
If the video recording mode shoots with video-corder mode for one, extraction is shot with video-corder in the image of the corresponding who object of focus
Hold；
If the video recording mode is that more people shoot with video-corder mode, extracts and multiple shoot with video-corder the corresponding multiple who objects of focus
Picture material.
In one embodiment, the method also includes:
Obtain the similarity of the corresponding similar users of the multiple who object；
Target person object is determined from the multiple who object according to the similarity, is the target person object
Search the product information of preference.
A kind of product information recommendation apparatus, described device include:
Video reception module, for receiving video；
Who object identification module, the who object in the video for identification, and it is corresponding to extract the who object
Picture material；
Various dimensions feature point set extraction module, for extracting the characteristic point of the who object from described image content
Collection, wherein the feature point set includes the corresponding various dimensions feature point set of multiple characteristic dimensions predetermined；
Feature portrait obtains module, and the feature for obtaining pre-stored user is drawn a portrait；
Similar users determining module, for the feature point set and feature portrait to be carried out fractional dimension characteristic matching,
Determine the corresponding similar users of the who object；
Product information determining module, for searching the preferred product information of the similar users, by the product information
Push to requesting terminal.
In one embodiment, the who object identification module, is also used to identify the who object in video；From described
The corresponding target image frame of the who object is determined in video, wherein the target image frame includes having the who object
Who object feature all picture frames；The who object feature is extracted from the target image frame；According to the people
Object characteristics of objects synthesizes the corresponding picture material of the who object.
In one embodiment, described device includes:
Recognition result determining module, for the user information of the similar users to be sent to requesting terminal, the request
Terminal shows the user information；Voice messaging is received, recognition and verification result is obtained according to the voice messaging；
The product information determining module is also used to obtain information recommendation algorithm corresponding with the recognition and verification result；
The product information is pushed to request eventually by the product information that the similar users preference is determined according to the information recommendation algorithm
End.
A kind of computer equipment, including memory and processor, the memory are stored with computer program, the processing
The step of device realizes method described above when executing the computer program.
A kind of computer readable storage medium, is stored thereon with computer program, and the computer program is held by processor
The step of method described above is realized when row.
The said goods information recommendation method, device, computer equipment and storage medium, when need into crowd any people into
When capable accurately Products Show, character features are collected in such a way that terminal shoots one section of video, server identifies in video
Who object, and the picture material of who object is extracted, which includes the feature point set of the various dimensions of who object, is led to
Cross and match the feature portrait of various dimensions feature point set and system user, server from searched in system user in video
Who object characteristic attribute having the same similar users, have same modality, clothing preference, behavioural characteristic identical two
Personal preference in other respects also can be more identical in very maximum probability.Relevant information based on similar system user can be real
Now to the accurate distribution for the who object shot with video-corder, and then realizes and accurately product information is carried out to unknown object any in crowd
Push.
Detailed description of the invention
Fig. 1 is the application scenario diagram of product information recommended method in one embodiment；
Fig. 2 is the flow diagram of product information recommended method in one embodiment；
Fig. 3 is flow diagram involved in the picture material of who object in one embodiment is extracted；
Fig. 4 is the flow diagram of product information recommended method in another embodiment；
Fig. 5 is the structural block diagram of product information recommendation apparatus in one embodiment；
Fig. 6 is the internal structure chart of computer equipment in one embodiment.
Specific embodiment
It is with reference to the accompanying drawings and embodiments, right in order to which the objects, technical solutions and advantages of the application are more clearly understood
The application is further elaborated.It should be appreciated that specific embodiment described herein is only used to explain the application, not
For limiting the application.
Product information recommended method provided by the present application, can be applied in application environment as shown in Figure 1.Wherein, eventually
End 102 is communicated with server 104 by network by network.Terminal to server sends product information recommendation request, asks
The middle video for carrying and shooting with video-corder is sought, server handles the information in video, matches, and determines the who object identified in video
Corresponding similar users, the preference based on similar users carry out Products Show.Wherein, terminal 102 can be, but not limited to be various
Portable wearable device or other portable terminals, server 104 can use independent server either multiple servers
The server cluster of composition is realized.
In one embodiment, as shown in Fig. 2, providing a kind of product information recommended method, it is applied to Fig. 1 in this way
In server for be illustrated, comprising the following steps:
Step 202, video is received.
Terminal opens picture pick-up device, and picture pick-up device shoots with video-corder video, and the video shot with video-corder is sent to server.Server connects
Receive the video that terminal is sent.
In one embodiment, terminal is the intelligent wearable device of built-in camera unit, such as intelligent glasses.Intelligent glasses are opened
Camera unit is opened, camera unit shoots with video-corder video, and the video shot with video-corder is sent to server by the transmission instruction of monitoring users triggering.
Above-mentioned video can be the video shot with video-corder to a who object, and the video for being also possible to shoot with video-corder a group who object (is clapped
Take the photograph crowd).That is, including at least a who object in video.
Step 204, it identifies the who object in video, and extracts the corresponding picture material of who object.
It include the set of all features in relation to who object occurred in video in the corresponding picture material of who object.
That is the picture material of who object is that be composed of all people's object characteristics of objects extracted from video one is completeer
Whole who object image.
Specifically, identifying all persons' object included in video first, all persons couple are then collected from video
As corresponding character pixel, set feature pixel obtains everyone the corresponding picture material of object object.Such as 4 frame figure preceding in video
There is who object A as in, then extracts character pixel related with who object A from preceding 4 frame image, combine all features
Pixel generates the corresponding picture material of the task object.
Step 206, the feature point set of the who object is extracted from described image content, wherein the feature point set
Including the corresponding various dimensions feature point set of multiple characteristic dimensions.
Multiple characteristic dimensions of pre-defined who object, such as face characteristic dimension, wearing characteristic dimension, behavioural characteristic dimension
Degree, and define the corresponding feature form of expression of each characteristic dimension.Each characteristic dimension corresponds to the performance of one or more feature
Form.Such as define the feature form of expression of facial image dimension are as follows: facial contour feature, five features, features of skin colors etc.；Wearing
The feature form of expression of characteristic dimension are as follows: apparel characteristic, ornaments feature, colour system feature etc.；The feature of behavioural characteristic dimension shows
Form are as follows: hand behavioural characteristic, leg behavioural characteristic etc..
Fractional dimension extracts feature point set from the picture material of who object.According to the corresponding mark sheet of each characteristic dimension
Existing form extracts the corresponding feature point set of each dimension from picture material.As extracted facial contour feature, five in picture material
The face characteristic of official's feature and features of skin colors as who object.
Step 208, the feature portrait of pre-stored user is obtained.
Server collects the characteristic information of all registration users or all users for buying this platform product in advance, and
The feature portrait of user is generated according to user's characteristic information.The characteristic information of collection is the characteristic information of various forms, such as with figure
As the feature of characterization, the feature characterized with profile, the feature with text characterization, fingerprint, vocal print.
Further, in order to make the efficiency for improving Characteristic Contrast, the characteristic information of the user of collection can be subjected to dimension and drawn
Point.
Step 210, feature point set and feature portrait are subjected to fractional dimension characteristic matching, determine that who object is corresponding similar
User.
The characteristic information for including in the feature of the feature point set of who object and the user of storage portrait is compared, is counted
Calculate the similarity of who object and each user.Similarity is greater than given threshold or one or more most like user
Similar users as who object.
When Characteristic Contrast, the corresponding feature with dimension is compared, if who object same dimension corresponding with user
Characteristic morphology it is not identical, then advanced row information Forms Transformation.It is (colour system pixel is special that character features such as are converted by pixel characteristic
Sign is converted to character features), character features, which are converted, by characteristics of image (converts behavior pixel characteristic or behavior characteristics of image
For character features).After the completion of conversion, fractional dimension carries out Characteristic Contrast, calculates who object according to various dimensions Characteristic Contrast result
Similarity between user.
Step 212, the preferred product information of similar users is obtained, product information is pushed into requesting terminal.
Historical behavior data, the historical purchase data for searching similar users, determine phase according to the historical data of similar users
It is carried out using the product preference of similar users as the product preference of who object based on product preference like the product preference of user
The Products Show of who object.
Although being also based on product preference in the present embodiment carries out Products Show, in the present embodiment, object to be recommended is simultaneously
Not instead of end command is determined based on image recognition technology.In addition, being not based on to the product of object recommendation to be recommended
The historical behavior data of object to be recommended, but the various dimensions feature of object to be recommended is collected based on feature identification technique, it is based on
The Characteristic Contrast of various dimensions determines similar users, and the product preference based on similar users is treated recommended progress product information and pushed away
It recommends.
It is realized in the present embodiment whenever and wherever possible to the accurately Products Show method of anyone progress in crowd.By right
Target person in crowd carries out video recording, and (without shooting with video-corder Face to face, no matter which kind of state target person is in, i.e.,
Make to be to cover face), the feature of the various dimensions such as form, clothing and the behavioural characteristic of target person is extracted from video；And then it is logical
The characteristic matching for crossing various dimensions determines system user similar with the target person.It is special with same modality, clothing preference, behavior
Levying the preference of identical two people in other respects also can be more identical in very maximum probability.Phase based on similar system user
Information is closed precisely to promote the target person shot with video-corder.Realize essence in the case where any information of unknown client to be promoted
Calibration information is recommended.
In one embodiment, as shown in figure 3, step 204, identifies the who object in video, and extract who object
Corresponding picture material, comprising:
Step 302, the who object in video is identified.
Identify the who object in video in each frame image, the who object that marker recognition goes out.Judge continuous image
Whether the who object in frame is same people, if so, same who object does identical label in different images frame.
Step 304, the corresponding target image frame of who object is determined, wherein target image frame includes with who object
All picture frames of who object feature.
The corresponding target image frame of each who object of identification is determined according to label.The target image frame of who object
Refer to the picture frame of the who object feature including the who object, that is, has the picture frame of personage's object tag.Such as video
In include 20 picture frames, 2 who objects are had identified from 20 picture frames, wherein in preceding 10 picture frames have first man
The who object feature of object object, then preceding 10 picture frames are the corresponding target image frame of the first who object.If rear 10 figures
Who object feature as having second who object in frame, then rear 10 picture frames are the corresponding target of the second who object
Picture frame.
Step 306, who object feature is extracted from target image frame.
It include the who object feature of corresponding who object in each target image frame.From all target image frames
The middle who object feature for extracting corresponding who object.
Upper part of the body feature as in first aim picture frame being who object, second target picture frame is who object
Lower part of the body feature, it is who object that third target image frame, which is the positive face feature of who object, the 4th target image frame,
Side face feature is all then the who object of who object from first object picture frame to the feature that the 4th target image frame extracts
Feature.
Step 308, according to the corresponding picture material of who object feature composite person object.
The who object feature of multiple angles from multiple target image frames is combined, generation can embody personage
The picture material of object all angles feature.The picture material is that who object is three-dimensional, comprehensively draws a portrait.
The picture material generation method based on video in through this embodiment produces who object solid, comprehensive
Portrait, the user of the picture material of who object and system user portrait compares base in this present embodiment, may compare accurate
The true similar user of determination the people's object object.
Further, when carrying out the synthesis of who object feature, who object feature can be added to the people constructed in advance
Object is drawn a portrait in model, and personage's portrait model is made of feature unit interconnected, and who object feature is added to phase
In the feature unit answered, who object is generated according to the connection relationship between feature unit and the feature for being filled into feature unit
Corresponding picture material.Before model of being drawn a portrait according to personage generates the picture material of who object, calculating task object first
Task object feature filling personage draw a portrait model feature unit filling completeness.If filling completeness is less than setting threshold
Value, then give up the task object.The number that feature unit is filled is more, and the data volume being received in is bigger, then fills completeness
It is higher.If filling completeness is less than given threshold, the who object of the who object is characterized in incomplete, shortage.Base
In incomplete shortage who object feature generate picture material be also incomplete.
Object to be recommended is selected by the way of giving up the fewer who object of characteristic quantity in the present embodiment.Give up feature
It measures that fewer who object is effective to reduce aspect ratio to occupied computer resource, improves the efficiency of operation, make
It is more efficiently timely to obtain entire product information recommendation process.
In one embodiment, as shown in figure 4, providing a kind of product information recommended method, this method is with requesting terminal
It for intelligent glasses, and is illustrated, specifically comprises the following steps: for applying in the server in Fig. 1
Step 402, the product information recommendation request that intelligent glasses are sent is received, carries and takes the photograph in the product information recommendation request
The video information of record.
Customer manager wears intelligent glasses, and issues product information recommendation to intelligent glasses by operation intelligent glasses and ask
It asks.Intelligent glasses can generate product information recommendation by the gesture motion of identification customer manager, the information such as voice, button operation and ask
It asks, and opens built-in photographic device and shoot with video-corder video.
Step 404, the who object in video information is obtained, and determines the corresponding similar users of who object.
Who object is identified in this step, is determined and is adopted in the mode and Fig. 2 and Fig. 3 embodiment of the similar users of who object
Mode is identical, and details are not described herein again.
Step 406, the user information of similar users is obtained, and the user information of similar users is sent to intelligent glasses,
Intelligent glasses show user information.
The user information of similar users is the associated user basic information of user identifier of similar users, buying behavior information
Deng.In one embodiment, user information includes the production that user images, user's Real Name, gender, age, user bought
Product mark.
Server is by the user information pushing of similar users to intelligent glasses, and intelligent glasses are in built-in virtual display
Show the user information of similar users.The user information can be seen by customer manager.
Step 408, voice messaging is received, recognition and verification result is obtained according to voice messaging.
Customer manager can go to confirm whether the similar users are this to who object according to the user information of similar users
People.Customer manager issues INQUIRE statement in the case where wearing intelligent glasses.As the essential information of the similar users of display is
" Zhang, newest so-and-so product bought ".Customer manager can go forth, and " you are Misses Zhang, you are flat at us recently for inquiry
Platform bought so-and-so product? " intelligent wearable device can monitor next sentence of the customer manager after INQUIRE statement.If
It is " you are good by Miss Zhang, so-and-so product you also satisfied ", then identifies that definitive result is " similar users i.e. client itself "；If
Be " it is too shy, you with I a client too as ", then identify that definitive result is that " similar users are not client's sheets
People ".Whether the similar users found according to sentence confirmation are the who object identified, i.e. recognition and verification result.
Step 410, information recommendation algorithm corresponding with recognition and verification result is obtained, is determined according to information recommendation algorithm similar
Product information is pushed to requesting terminal by the product information of user preference.
It is client itself and similar users are not client itselfs that recognition and verification result, which includes similar users,.Pre-define this two
The corresponding information recommendation algorithm of kind recognition and verification result, information recommendation algorithm includes information acquisition algorithm, information analysis algorithms.No
Same recognition and verification result uses different information acquisition algorithms, and the different information based on acquisition carry out different information point
Analysis, and then obtain different recommended products information.
If recognition result is that similar users are client itselfs, historical viewings data, the historical purchase data of client, root can be obtained
According to the product preference of historical viewings data and historical purchase data analysis client, determine that product to be recommended is believed according to product preference
Breath.If recognition result is that similar users are not client itselfs, the history purchaser record of similar users is obtained, by similar users
The product of purchase is as product to be recommended.The information recommendation algorithm defined for different recognition results can be according to practical need
It asks progress customized, does not do specific limitation herein.
In the present embodiment, confirmation operation is added, different proposed algorithms is selected according to confirmation result, the product recommended can be made
More meets the needs of who object.
In one embodiment, the video information carried in the product information recommendation request that intelligent glasses are sent is with video
Shoot with video-corder mode.Video recording mode, video may be selected when shooting a certain personage or crowd based on intelligent glasses in customer manager
The mode of shooting with video-corder includes that one shoots with video-corder mode and more people shoot with video-corder mode.
After server receives video, the video recording mode carried in video is obtained.If video recording mode is for one
Model, then only using the corresponding personage of focus as the who object of product information to be recommended.Obtain the corresponding who object of focus
Picture material.By comparing user's portrait of picture material and system user, focus pair is determined from system user
The similar users for the who object answered, may preference according to who object unknown in the existing information prediction video of similar users
Product information, the accurate information of unknown target person is recommended with realizing.
If video recording mode is two-player mode, using the corresponding personage of multiple focuses as the people of product information to be recommended
Object object obtains the picture material of the corresponding multiple who objects of multifocal.By by the user of picture material and system user
Portrait compares, and the similar users of each who object are determined from system user, can predict everyone according to similar users
The product information of the preference of object object.
In another embodiment, it can also be after the similar users for determining each who object, obtain each personage
Similarity (similarity is to determine what similar users had calculated) between object and similar users, will be maximum similar
Corresponding who object is spent as target person object, only to its product preference of target person object prediction.
It should be understood that although each step in the flow chart of Fig. 2-4 is successively shown according to the instruction of arrow,
These steps are not that the inevitable sequence according to arrow instruction successively executes.Unless expressly stating otherwise herein, these steps
Execution there is no stringent sequences to limit, these steps can execute in other order.Moreover, at least one in Fig. 2-4
Part steps may include that perhaps these sub-steps of multiple stages or stage are not necessarily in synchronization to multiple sub-steps
Completion is executed, but can be executed at different times, the execution sequence in these sub-steps or stage is also not necessarily successively
It carries out, but can be at least part of the sub-step or stage of other steps or other steps in turn or alternately
It executes.
In one embodiment, as shown in figure 5, providing a kind of product information recommendation apparatus, comprising:
Video reception module 502, for receiving video.
Who object identification module 504, the who object in the video for identification, and extract the who object pair
The picture material answered.
Various dimensions feature point set extraction module 506, for extracting the feature of the who object from described image content
Point set, wherein the feature point set includes the corresponding various dimensions feature point set of multiple characteristic dimensions predetermined.
Feature portrait obtains module 508, and the feature for obtaining pre-stored user is drawn a portrait.
Similar users determining module 510, for the feature point set and feature portrait to be carried out fractional dimension feature
Match, determines the corresponding similar users of the who object.
Product information determining module 512 believes the product for searching the preferred product information of the similar users
Breath pushes to requesting terminal.
In one embodiment, the who object identification module 504, is also used to identify the who object in video；From
The corresponding target image frame of the who object is determined in the video, wherein the target image frame includes having the personage
All picture frames of the who object feature of object；The who object feature is extracted from the target image frame；According to institute
It states who object feature and synthesizes the corresponding picture material of the who object.
In one embodiment, product information recommendation apparatus further include: recognition result determining module, being used for will be described similar
The user information of user is sent to requesting terminal, and the requesting terminal shows the user information；Voice messaging is received, according to institute
It states voice messaging and obtains recognition and verification result.
The product information determining module 512 is also used to obtain information recommendation corresponding with the recognition and verification result and calculates
Method；The product information that the similar users preference is determined according to the information recommendation algorithm, the product information is pushed to and is asked
Seek terminal.
In one embodiment, the who object identification module 504 is also used to obtain the corresponding video of the video and takes the photograph
Record mode；If the video recording mode shoots with video-corder mode for one, extraction is shot with video-corder in the image of the corresponding who object of focus
Hold；If the video recording mode is that more people shoot with video-corder mode, multiple figures for shooting with video-corder the corresponding multiple who objects of focus are extracted
As content.
In one embodiment, product information recommendation apparatus further include: target person object determining module, for obtaining
State the similarity of the corresponding similar users of multiple who objects；According to the similarity from the multiple who object
Middle determining target person object searches the product information of preference for the target person object.
Specific about product information recommendation apparatus limits the limit that may refer to above for product information recommended method
Fixed, details are not described herein.Modules in the said goods information recommending apparatus can fully or partially through software, hardware and its
Combination is to realize.Above-mentioned each module can be embedded in the form of hardware or independently of in the processor in computer equipment, can also be with
It is stored in the memory in computer equipment in a software form, in order to which processor calls the above modules of execution corresponding
Operation.
In one embodiment, a kind of computer equipment is provided, which can be server, internal junction
Composition can be as shown in Figure 6.The computer equipment include by system bus connect processor, memory, network interface and
Database.Wherein, the processor of the computer equipment is for providing calculating and control ability.The memory packet of the computer equipment
Include non-volatile memory medium, built-in storage.The non-volatile memory medium is stored with operating system, computer program and data
Library.The built-in storage provides environment for the operation of operating system and computer program in non-volatile memory medium.The calculating
The database of machine equipment is used to store the feature portrait of user.The network interface of the computer equipment is used to wear with external intelligence
It wears equipment and passes through network connection communication.To realize a kind of product information recommended method when the computer program is executed by processor.
It will be understood by those skilled in the art that structure shown in Fig. 6, only part relevant to application scheme is tied
The block diagram of structure does not constitute the restriction for the computer equipment being applied thereon to application scheme, specific computer equipment
It may include perhaps combining certain components or with different component layouts than more or fewer components as shown in the figure.
In one embodiment, a kind of computer equipment, including memory and processor are provided, which is stored with
Computer program, the processor perform the steps of reception video when executing computer program；Identify the personage in the video
Object, and extract the corresponding picture material of the who object；The feature of the who object is extracted from described image content
Point set, wherein the feature point set includes the corresponding various dimensions feature point set of multiple characteristic dimensions predetermined；It obtains preparatory
The feature of the user of storage is drawn a portrait；The feature point set and the feature are drawn a portrait and carry out fractional dimension characteristic matching, described in determination
The corresponding similar users of who object；The preferred product information of the similar users is searched, the product information is pushed to
Requesting terminal.
In one embodiment, the personage in identification video is also performed the steps of when processor executes computer program
Object；The corresponding target image frame of the who object is determined from the video, wherein the target image frame includes having
All picture frames of the who object feature of the who object；It is special that the who object is extracted from the target image frame
Sign；The corresponding picture material of the who object is synthesized according to the who object feature.
In one embodiment, it also performs the steps of when processor executes computer program by the similar users
User information is sent to requesting terminal, and the requesting terminal shows the user information；Voice messaging is received, according to the voice
Information obtains recognition and verification result；Obtain information recommendation algorithm corresponding with the recognition and verification result；It is pushed away according to the information
The product information that algorithm determines the similar users preference is recommended, the product information is pushed into requesting terminal.
In one embodiment, it is corresponding that the acquisition video is also performed the steps of when processor executes computer program
Video recording mode；If the video recording mode shoots with video-corder mode for one, the corresponding who object of focus is shot with video-corder in extraction
Picture material；If the video recording mode is that more people shoot with video-corder mode, extracts and multiple shoot with video-corder the corresponding multiple personages of focus
The picture material of object.
In one embodiment, it is also performed the steps of when processor executes computer program and obtains the multiple personage
The similarity of the corresponding similar users of object；Target is determined from the multiple who object according to the similarity
Who object searches the product information of preference for the target person object.
In one embodiment, a kind of computer readable storage medium is provided, computer program is stored thereon with, is calculated
Machine program performs the steps of reception video when being executed by processor；Identify the who object in the video, and described in extraction
The corresponding picture material of who object；The feature point set of the who object is extracted from described image content, wherein the spy
Levying point set includes the corresponding various dimensions feature point set of multiple characteristic dimensions predetermined；Obtain the feature of pre-stored user
Portrait；The feature point set and feature portrait are subjected to fractional dimension characteristic matching, determine the corresponding phase of the who object
Like user；The preferred product information of the similar users is searched, the product information is pushed into requesting terminal.
In one embodiment, the people in identification video is also performed the steps of when computer program is executed by processor
Object object；The corresponding target image frame of the who object is determined from the video, wherein the target image frame includes band
There are all picture frames of the who object feature of the who object；It is special that the who object is extracted from the target image frame
Sign；The corresponding picture material of the who object is synthesized according to the who object feature.
In one embodiment, it is also performed the steps of when computer program is executed by processor by the similar users
User information be sent to requesting terminal, the requesting terminal shows the user information；Voice messaging is received, according to institute's predicate
Message ceases to obtain recognition and verification result；Obtain information recommendation algorithm corresponding with the recognition and verification result；According to the information
Proposed algorithm determines the product information of the similar users preference, and the product information is pushed to requesting terminal.
In one embodiment, it is also performed the steps of when computer program is executed by processor and obtains the video pair
The video recording mode answered；If the video recording mode shoots with video-corder mode for one, the corresponding personage couple of focus is shot with video-corder in extraction
The picture material of elephant；If the video recording mode is that more people shoot with video-corder mode, extracts and multiple shoot with video-corder the corresponding multiple people of focus
The picture material of object object.
In one embodiment, it is also performed the steps of when computer program is executed by processor and obtains the multiple people
The similarity of the corresponding similar users of object object；Mesh is determined from the multiple who object according to the similarity
Who object is marked, the product information of preference is searched for the target person object.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Relevant hardware is instructed to complete by computer program, the computer program can be stored in a non-volatile computer
In read/write memory medium, the computer program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein,
To any reference of memory, storage, database or other media used in each embodiment provided herein,
Including non-volatile and/or volatile memory.Nonvolatile memory may include read-only memory (ROM), programming ROM
(PROM), electrically programmable ROM (EPROM), electrically erasable ROM (EEPROM) or flash memory.Volatile memory may include
Random access memory (RAM) or external cache.By way of illustration and not limitation, RAM is available in many forms,
Such as static state RAM (SRAM), dynamic ram (DRAM), synchronous dram (SDRAM), double data rate sdram (DDRSDRAM), enhancing
Type SDRAM (ESDRAM), synchronization link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM
(RDRAM), direct memory bus dynamic ram (DRDRAM) and memory bus dynamic ram (RDRAM) etc..
Each technical characteristic of above embodiments can be combined arbitrarily, for simplicity of description, not to above-described embodiment
In each technical characteristic it is all possible combination be all described, as long as however, the combination of these technical characteristics be not present lance
Shield all should be considered as described in this specification.
The several embodiments of the application above described embodiment only expresses, the description thereof is more specific and detailed, but simultaneously
It cannot therefore be construed as limiting the scope of the patent.It should be pointed out that coming for those of ordinary skill in the art
It says, without departing from the concept of this application, various modifications and improvements can be made, these belong to the protection of the application
Range.Therefore, the scope of protection shall be subject to the appended claims for the application patent.
Claims (10)
1. a kind of product information recommended method, which comprises
Receive video；
It identifies the who object in the video, and extracts the corresponding picture material of the who object；
The feature point set of the who object is extracted from described image content, wherein the feature point set includes multiple features
The corresponding various dimensions feature point set of dimension；
Obtain the feature portrait of pre-stored user；
The feature point set and feature portrait are subjected to fractional dimension characteristic matching, determine that the who object is corresponding similar
User；
The preferred product information of the similar users is searched, the product information is pushed into requesting terminal.
2. the method according to claim 1, wherein the who object in the identification video, and extracting
The corresponding picture material of the who object, comprising:
Identify the who object in the video；
The corresponding target image frame of the who object is determined from the video, wherein the target image frame include band
State all picture frames of the who object feature of who object；
The who object feature is extracted from the target image frame；
The corresponding picture material of the who object is synthesized according to the who object feature.
3. method according to claim 1 or 2, which is characterized in that the method also includes:
The user information of the similar users is sent to requesting terminal, the requesting terminal shows the user information；
Voice messaging is received, recognition and verification result is obtained according to the voice messaging；
Obtain information recommendation algorithm corresponding with the recognition and verification result；
The product information that the similar users preference is determined according to the information recommendation algorithm, the product information is pushed to and is asked
Seek terminal.
4. method according to claim 1 or 2, which is characterized in that the who object in the identification video, and mention
Take the corresponding picture material of the who object, comprising:
Obtain the corresponding video recording mode of the video；
If the video recording mode shoots with video-corder mode for one, the picture material for shooting with video-corder the corresponding who object of focus is extracted；
If the video recording mode is that more people shoot with video-corder mode, multiple figures for shooting with video-corder the corresponding multiple who objects of focus are extracted
As content.
5. according to the method described in claim 4, it is characterized in that, the method also includes:
Obtain the similarity of the corresponding similar users of the multiple who object；
Target person object is determined from the multiple who object according to the similarity, is searched for the target person object
The product information of preference.
6. a kind of product information recommendation apparatus, which is characterized in that described device includes:
Video reception module, for receiving video；
Who object identification module, the who object in the video for identification, and extract the corresponding figure of the who object
As content；
Various dimensions feature point set extraction module, for extracting the feature point set of the who object from described image content,
In, the feature point set includes the corresponding various dimensions feature point set of multiple characteristic dimensions；
Feature portrait obtains module, and the feature for obtaining pre-stored user is drawn a portrait；
Similar users determining module is determined for the feature point set and feature portrait to be carried out fractional dimension characteristic matching
The corresponding similar users of the who object；
Product information determining module pushes the product information for searching the preferred product information of the similar users
To requesting terminal.
7. device according to claim 6, which is characterized in that the who object identification module is also used to identify described
Who object in video；The corresponding target image frame of the who object is determined from the video, wherein the target figure
As frame includes all picture frames of the who object feature with the who object；Described in being extracted from the target image frame
Who object feature；The corresponding picture material of the who object is synthesized according to the who object feature.
8. method according to claim 6 or 7, which is characterized in that described device further include:
Recognition result determining module, for the user information of the similar users to be sent to requesting terminal, the requesting terminal
Show the user information；Voice messaging is received, recognition and verification result is obtained according to the voice messaging；
The product information determining module is also used to obtain information recommendation algorithm corresponding with the recognition and verification result；According to
The information recommendation algorithm determines the product information of the similar users preference, and the product information is pushed to requesting terminal.
9. a kind of computer equipment, including memory and processor, the memory are stored with computer program, feature exists
In the step of processor realizes any one of claims 1 to 5 the method when executing the computer program.
10. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the computer program
The step of method described in any one of claims 1 to 5 is realized when being executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811205259.4A CN109523344A (en) | 2018-10-16 | 2018-10-16 | Product information recommended method, device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811205259.4A CN109523344A (en) | 2018-10-16 | 2018-10-16 | Product information recommended method, device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109523344A true CN109523344A (en) | 2019-03-26 |
Family
ID=65770915
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811205259.4A Pending CN109523344A (en) | 2018-10-16 | 2018-10-16 | Product information recommended method, device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109523344A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110046965A (en) * | 2019-04-18 | 2019-07-23 | 北京百度网讯科技有限公司 | Information recommendation method, device, equipment and medium |
| CN110111136A (en) * | 2019-04-12 | 2019-08-09 | 平安科技（深圳）有限公司 | Video data handling procedure, device, computer equipment and storage medium |
| CN110263258A (en) * | 2019-06-25 | 2019-09-20 | 北京字节跳动网络技术有限公司 | Method, apparatus, electronic equipment and the computer-readable medium of commending contents |
| CN111859122A (en) * | 2020-06-30 | 2020-10-30 | 北京百度网讯科技有限公司 | Method and device for recommending medical and cosmetic products, electronic equipment and readable storage medium |
| CN112100524A (en) * | 2020-09-17 | 2020-12-18 | 北京百度网讯科技有限公司 | Information recommendation method, device, equipment and storage medium |
| WO2021004137A1 (en) * | 2019-07-05 | 2021-01-14 | 深圳壹账通智能科技有限公司 | Information pushing method and apparatus based on face recognition and computer device |
| CN112783984A (en) * | 2021-02-10 | 2021-05-11 | 中国人民银行数字货币研究所 | Information display method and device |
| CN112804567A (en) * | 2021-01-04 | 2021-05-14 | 青岛聚看云科技有限公司 | Display device, server and video recommendation method |
| CN113947428A (en) * | 2021-10-14 | 2022-01-18 | 上海清赟医药科技有限公司 | Object visit recommendation method based on user portrait |
| CN114513681A (en) * | 2022-01-25 | 2022-05-17 | 武汉工程大学 | Video processing system, method, device, electronic equipment and storage medium |
| CN115114508A (en) * | 2021-03-17 | 2022-09-27 | 博泰车联网科技(上海)股份有限公司 | User preference query method and device for cloud platform, storage medium and terminal |
| US11599598B2 (en) * | 2019-06-19 | 2023-03-07 | Canon Kabushiki Kaisha | System and method for recommending challenges |
| CN118229356A (en) * | 2024-04-12 | 2024-06-21 | 广州致奥科技有限公司 | Advertisement putting and user behavior interaction system |
Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120221418A1 (en) * | 2000-08-24 | 2012-08-30 | Linda Smith | Targeted Marketing System and Method |
| CN102930457A (en) * | 2012-10-24 | 2013-02-13 | 深圳市万凯达科技有限公司 | Method and system for implementing application recommendation based on facial image characteristics |
| KR20130027801A (en) * | 2011-09-08 | 2013-03-18 | 에버클라우드 주식회사 | User terminal for style matching, style matching system using the user terminal and method thereof |
| CN104778237A (en) * | 2015-04-03 | 2015-07-15 | 电子科技大学 | Individual recommending method and system based on key users |
| CN105869015A (en) * | 2016-03-28 | 2016-08-17 | 联想(北京)有限公司 | Information processing method and system |
| CN106096009A (en) * | 2016-06-23 | 2016-11-09 | 北京小米移动软件有限公司 | Method for generating message and device |
| CN106651533A (en) * | 2016-12-29 | 2017-05-10 | 合肥华凌股份有限公司 | User behavior-based personalized product recommendation method and apparatus |
| CN106776619A (en) * | 2015-11-20 | 2017-05-31 | 百度在线网络技术（北京）有限公司 | Method and apparatus for determining the attribute information of destination object |
| CN106815309A (en) * | 2016-12-20 | 2017-06-09 | 北京奇虎科技有限公司 | A kind of image method for pushing, device and mobile terminal |
| CN107291732A (en) * | 2016-03-31 | 2017-10-24 | 苏宁云商集团股份有限公司 | A kind of information-pushing method and device |
| CN107644213A (en) * | 2017-09-26 | 2018-01-30 | 司马大大（北京）智能系统有限公司 | Video person extraction method and device |
| CN107688984A (en) * | 2017-07-27 | 2018-02-13 | 上海壹账通金融科技有限公司 | Product information method for pushing, device, storage medium and computer equipment |
| CN108228914A (en) * | 2018-03-06 | 2018-06-29 | 中经汇通电子商务有限公司 | A kind of information transmission system |
| CN108446390A (en) * | 2018-03-22 | 2018-08-24 | 百度在线网络技术（北京）有限公司 | Method and apparatus for pushed information |
| CN108446385A (en) * | 2018-03-21 | 2018-08-24 | 百度在线网络技术（北京）有限公司 | Method and apparatus for generating information |
| CN108495143A (en) * | 2018-03-30 | 2018-09-04 | 百度在线网络技术（北京）有限公司 | The method and apparatus of video recommendations |
| CN108551462A (en) * | 2018-02-28 | 2018-09-18 | 阿里巴巴集团控股有限公司 | A kind of information-pushing method, device and equipment |
-
2018
- 2018-10-16 CN CN201811205259.4A patent/CN109523344A/en active Pending
Patent Citations (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120221418A1 (en) * | 2000-08-24 | 2012-08-30 | Linda Smith | Targeted Marketing System and Method |
| KR20130027801A (en) * | 2011-09-08 | 2013-03-18 | 에버클라우드 주식회사 | User terminal for style matching, style matching system using the user terminal and method thereof |
| CN102930457A (en) * | 2012-10-24 | 2013-02-13 | 深圳市万凯达科技有限公司 | Method and system for implementing application recommendation based on facial image characteristics |
| CN104778237A (en) * | 2015-04-03 | 2015-07-15 | 电子科技大学 | Individual recommending method and system based on key users |
| CN106776619A (en) * | 2015-11-20 | 2017-05-31 | 百度在线网络技术（北京）有限公司 | Method and apparatus for determining the attribute information of destination object |
| CN105869015A (en) * | 2016-03-28 | 2016-08-17 | 联想(北京)有限公司 | Information processing method and system |
| CN107291732A (en) * | 2016-03-31 | 2017-10-24 | 苏宁云商集团股份有限公司 | A kind of information-pushing method and device |
| CN106096009A (en) * | 2016-06-23 | 2016-11-09 | 北京小米移动软件有限公司 | Method for generating message and device |
| CN106815309A (en) * | 2016-12-20 | 2017-06-09 | 北京奇虎科技有限公司 | A kind of image method for pushing, device and mobile terminal |
| CN106651533A (en) * | 2016-12-29 | 2017-05-10 | 合肥华凌股份有限公司 | User behavior-based personalized product recommendation method and apparatus |
| CN107688984A (en) * | 2017-07-27 | 2018-02-13 | 上海壹账通金融科技有限公司 | Product information method for pushing, device, storage medium and computer equipment |
| CN107644213A (en) * | 2017-09-26 | 2018-01-30 | 司马大大（北京）智能系统有限公司 | Video person extraction method and device |
| CN108551462A (en) * | 2018-02-28 | 2018-09-18 | 阿里巴巴集团控股有限公司 | A kind of information-pushing method, device and equipment |
| CN108228914A (en) * | 2018-03-06 | 2018-06-29 | 中经汇通电子商务有限公司 | A kind of information transmission system |
| CN108446385A (en) * | 2018-03-21 | 2018-08-24 | 百度在线网络技术（北京）有限公司 | Method and apparatus for generating information |
| CN108446390A (en) * | 2018-03-22 | 2018-08-24 | 百度在线网络技术（北京）有限公司 | Method and apparatus for pushed information |
| CN108495143A (en) * | 2018-03-30 | 2018-09-04 | 百度在线网络技术（北京）有限公司 | The method and apparatus of video recommendations |
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110111136A (en) * | 2019-04-12 | 2019-08-09 | 平安科技（深圳）有限公司 | Video data handling procedure, device, computer equipment and storage medium |
| CN110111136B (en) * | 2019-04-12 | 2023-04-18 | 平安科技（深圳）有限公司 | Video data processing method, video data processing device, computer equipment and storage medium |
| CN110046965A (en) * | 2019-04-18 | 2019-07-23 | 北京百度网讯科技有限公司 | Information recommendation method, device, equipment and medium |
| US11599598B2 (en) * | 2019-06-19 | 2023-03-07 | Canon Kabushiki Kaisha | System and method for recommending challenges |
| CN110263258A (en) * | 2019-06-25 | 2019-09-20 | 北京字节跳动网络技术有限公司 | Method, apparatus, electronic equipment and the computer-readable medium of commending contents |
| WO2021004137A1 (en) * | 2019-07-05 | 2021-01-14 | 深圳壹账通智能科技有限公司 | Information pushing method and apparatus based on face recognition and computer device |
| CN111859122A (en) * | 2020-06-30 | 2020-10-30 | 北京百度网讯科技有限公司 | Method and device for recommending medical and cosmetic products, electronic equipment and readable storage medium |
| CN111859122B (en) * | 2020-06-30 | 2024-06-11 | 北京百度网讯科技有限公司 | Method, device, electronic equipment and readable storage medium for recommending medical and aesthetic products |
| CN112100524A (en) * | 2020-09-17 | 2020-12-18 | 北京百度网讯科技有限公司 | Information recommendation method, device, equipment and storage medium |
| CN112804567A (en) * | 2021-01-04 | 2021-05-14 | 青岛聚看云科技有限公司 | Display device, server and video recommendation method |
| CN112783984A (en) * | 2021-02-10 | 2021-05-11 | 中国人民银行数字货币研究所 | Information display method and device |
| CN115114508A (en) * | 2021-03-17 | 2022-09-27 | 博泰车联网科技(上海)股份有限公司 | User preference query method and device for cloud platform, storage medium and terminal |
| CN113947428A (en) * | 2021-10-14 | 2022-01-18 | 上海清赟医药科技有限公司 | Object visit recommendation method based on user portrait |
| CN114513681A (en) * | 2022-01-25 | 2022-05-17 | 武汉工程大学 | Video processing system, method, device, electronic equipment and storage medium |
| CN118229356A (en) * | 2024-04-12 | 2024-06-21 | 广州致奥科技有限公司 | Advertisement putting and user behavior interaction system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109523344A (en) | Product information recommended method, device, computer equipment and storage medium | |
| US11282256B2 (en) | Crowdshaping realistic 3D avatars with words | |
| EP3899788A1 (en) | Methods and systems for automatic generation of massive training data sets from 3d models for training deep learning networks | |
| EP2953090A1 (en) | Makeup assistance device, makeup assistance system, makeup assistance method, and makeup assistance program | |
| Kucer et al. | Leveraging expert feature knowledge for predicting image aesthetics | |
| US20180268458A1 (en) | Automated recommendation and virtualization systems and methods for e-commerce | |
| US20120299912A1 (en) | Avatar-based virtual dressing room | |
| CN109767261A (en) | Products Show method, apparatus, computer equipment and storage medium | |
| CN110135257A (en) | Business recommended data generation, device, computer equipment and storage medium | |
| US20130097194A1 (en) | Apparatus, method, and computer-accessible medium for displaying visual information | |
| CN105005777A (en) | Face-based audio and video recommendation method and face-based audio and video recommendation system | |
| JP2023531264A (en) | Systems and methods for improved facial attribute classification and its use | |
| CN109766491A (en) | Product search method, device, computer equipment and storage medium | |
| US20230044146A1 (en) | Video processing method, video searching method, terminal device, and computer-readable storage medium | |
| US11507781B2 (en) | Methods and systems for automatic generation of massive training data sets from 3D models for training deep learning networks | |
| CN109934092A (en) | Identify color method, apparatus, computer equipment and storage medium | |
| CN118673210A (en) | Systems and methods for providing personalized product recommendations using deep learning | |
| CN108875517A (en) | Method for processing video frequency, device and system and storage medium | |
| WO2024213025A1 (en) | Hand modeling method, hand model processing method, device, and medium | |
| CN110598097B (en) | Hair style recommendation system, method, equipment and storage medium based on CNN | |
| KR102465454B1 (en) | An apparatus and a method for generating virtual makeup beauty score by using AI recognition of facial features | |
| Ma et al. | Landmark‐Based Facial Feature Construction and Action Unit Intensity Prediction | |
| Kayaoglu et al. | Affect recognition using key frame selection based on minimum sparse reconstruction | |
| US11151598B2 (en) | Scoring image engagement in digital media | |
| CN109859011A (en) | Based on the information push method in store, system and its storage medium in jewellery wire |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190326 |