CN109068173A - A kind of method for processing video frequency and video process apparatus - Google Patents
A kind of method for processing video frequency and video process apparatus Download PDFInfo
- Publication number
- CN109068173A CN109068173A CN201811142602.5A CN201811142602A CN109068173A CN 109068173 A CN109068173 A CN 109068173A CN 201811142602 A CN201811142602 A CN 201811142602A CN 109068173 A CN109068173 A CN 109068173A
- Authority
- CN
- China
- Prior art keywords
- commodity
- several
- video
- picture
- pictures
- 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 45
- 238000012545 processing Methods 0.000 title claims abstract description 23
- 230000008569 process Effects 0.000 title claims abstract description 13
- 238000013527 convolutional neural network Methods 0.000 claims description 13
- 230000000007 visual effect Effects 0.000 claims description 10
- 238000001514 detection method Methods 0.000 claims description 6
- 238000013135 deep learning Methods 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 4
- 238000004590 computer program Methods 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 4
- 238000013461 design Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000011897 real-time detection Methods 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000003032 molecular docking Methods 0.000 description 2
- 238000006116 polymerization reaction Methods 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 238000013481 data capture Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000000423 heterosexual effect Effects 0.000 description 1
- 238000002513 implantation Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000013011 mating Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/44—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs
- H04N21/44008—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream or rendering scenes according to encoded video stream scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics in the video stream
-
- 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
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
-
- 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
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0241—Advertisements
- G06Q30/0251—Targeted advertisements
- G06Q30/0253—During e-commerce, i.e. online transactions
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/431—Generation of visual interfaces for content selection or interaction; Content or additional data rendering
- H04N21/4312—Generation of visual interfaces for content selection or interaction; Content or additional data rendering involving specific graphical features, e.g. screen layout, special fonts or colors, blinking icons, highlights or animations
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/47—End-user applications
- H04N21/478—Supplemental services, e.g. displaying phone caller identification, shopping application
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/47—End-user applications
- H04N21/478—Supplemental services, e.g. displaying phone caller identification, shopping application
- H04N21/47815—Electronic shopping
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/80—Generation or processing of content or additional data by content creator independently of the distribution process; Content per se
- H04N21/81—Monomedia components thereof
- H04N21/812—Monomedia components thereof involving advertisement data
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- Game Theory and Decision Science (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The embodiment of the present invention provides a kind of method for processing video frequency and video process apparatus, comprising: the first video flowing is split as a series of video blocks, and in order to every section of video block marking serial numbers；According to the video block, several first commodity pictures are determined；Several second commodity pictures are determined according to user operation records；The operation note includes: the access path for recording user and accessing a certain commodity；Judge whether several first commodity pictures match with several second commodity pictures determined according to user operation records；When several first commodity pictures are matched with several second commodity pictures, Video coding is carried out based on video block serial number group of second commodity picture to acquisition and generates the second video flowing, solves the problems, such as that ad content and user demand are unmatched when user watches video.
Description
Technical field
The present invention relates to field of artificial intelligence, and in particular to a kind of method for processing video frequency and video process apparatus.
Background technique
For video viewers, the existing form in teaser or tail addition advertisement seriously affects user and watches view
The experience of frequency.And required for added ad content not necessarily user, i.e., ad content and user demand be not
Match.
For advertiser, now traditional advertisement implantation is stiff, and without bright spot, and ad data cannot be real
When monitor, advertising results can not be estimated well.In addition the height of advertisement required cost, another people are left speechless.
For video platform, the advertisement form of conventional conduit may reduce user to the satisfaction of platform, even
Certain advertisement is caused to be lost.When in addition docking advertiser, none convenient comprehensive channel takes so that whole process is time-consuming
Power.
Summary of the invention
The embodiment of the present invention is designed to provide a kind of method for processing video frequency and video process apparatus, uses to solve
Ad content and user demand unmatched problem when video is watched at family.
In a first aspect, the embodiment of the present invention provides a kind of method for processing video frequency, comprising:
First video flowing is split as a series of video blocks, and in order to every section of video block marking serial numbers；
According to the video block, several first commodity pictures are determined；
Several second commodity pictures are determined according to user operation records；The operation note includes: that record user accesses certain
The access path of one commodity；
Several second commodity pictures for judging several first commodity pictures and being determined according to user operation records
Whether match；
Several second commodity pictures for judging several first commodity pictures and being determined according to user operation records
Whether match；
When several first commodity pictures are matched with several second commodity pictures, it is based on the second commodity figure
Piece carries out Video coding to the video block serial number group of acquisition and generates the second video flowing.
After determining several second commodity pictures according to user operation records, and several first commodity pictures described in interpretation
Before whether being matched with several second commodity pictures determined according to user operation records further include:
Convolutional neural networks CNN is preferably based on to the video of several first commodity pictures and the second commodity picture
Block detection framework directly generates candidate frame using bottom visual signature, on the candidate frame generating mode using parallel computation and
Anchor point position in a parallel mode varies one's tactics.
Preferably, after the generation candidate frame, and several first commodity pictures described in interpretation with according to user's operation
Before whether determining several second commodity pictures of record match further include: extract picture feature；The extraction picture is special
Sign includes: that the low layer view of selective polymerisation is carried out according to the local feature of the first commodity picture and the second commodity picture Scale invariant
Feel character representation.
Preferably, described that selectivity is carried out according to the local feature of the first commodity picture and the second commodity picture Scale invariant
After the low-level visual feature expression of polymerization, and several first commodity pictures described in interpretation are determined with according to user operation records
Several second commodity pictures whether match before further include: picture feature label；The picture feature label includes: pair
First commodity picture and the second commodity picture are expressed based on the high-level semantics features of deep learning.
Preferably, classified using classifier to the first commodity picture and the second commodity picture.
The operation note includes: the attribute data of user；
First commodity picture and the second commodity picture include: item property data.
Preferably, the second video flowing that the Video coding generates includes: two dimensional code, is swept for user using cell phone application
Code obtains the item property data of several second commodity pictures.
Second aspect, the embodiment of the present invention provide a kind of video process apparatus, comprising:
Video flowing splits module, for the first video flowing to be split as a series of video blocks, and in order to every section of video
Block marking serial numbers；
First determining module, for determining several first commodity pictures according to the video block；
Second determining module, for determining several second commodity pictures according to user operation records；The operation note packet
Include: record user accesses the access path of a certain commodity；
Judgment module, for judge several first commodity pictures with according to user operation records determine it is described several
Whether the second commodity picture matches；
Video flowing generation module when matching for several first commodity pictures with several second commodity pictures, is based on second
Commodity picture carries out Video coding to the video block serial number group of acquisition and generates the second video flowing.
The third aspect, the embodiment of the present invention provide a kind of computer-readable medium, are stored thereon with computer program, described
Method described in any one possible design of first aspect or first aspect is realized when program is executed by processor.
Fourth aspect, the embodiment of the present invention provide a kind of video process apparatus, comprising:
Several processors；
Storage device, for storing several programs, when several programs are executed by several processors, so that institute
It states several processors and realizes method described in any one possible design of first aspect or first aspect.
The embodiment of the present invention provide a kind of method for processing video frequency by judge several first commodity pictures and according to
Whether several second commodity pictures that family operation note determines match；Based on the second commodity picture to the video block sequence of acquisition
Number group carries out Video coding and generates the second video flowing, and it is unmatched to solve ad content and user demand when user watches video
Problem.
Detailed description of the invention
Fig. 1 shows the flow chart of method for processing video frequency according to an embodiment of the invention；
Fig. 2 shows the structural schematic diagrams of the video process apparatus provided according to inventive embodiments；
Fig. 3 shows another structural schematic diagram of video process apparatus provided in an embodiment of the present invention.
Specific embodiment
Embodiments of the present invention are illustrated by particular specific embodiment below, those skilled in the art can be by this explanation
Content disclosed by book is understood other advantages and efficacy of the present invention easily.
In description of the embodiment of the present invention, the vocabulary such as " first ", " second " are only used for distinguishing description, and should not be understood as referring to
Show or imply relative importance, indication or suggestion sequence can not be interpreted as.
Fig. 1 shows the flow chart of method for processing video frequency according to an embodiment of the invention, as shown in Figure 1, comprising:
S101, the first video flowing is split as a series of video blocks, and in order to every section of video block marking serial numbers.
S102, according to the video block, determine several first commodity pictures.
S103, several second commodity pictures are determined according to user operation records；The operation note includes: that record user visits
Ask the access path of a certain commodity.Specifically, it is mentioned by the records such as user's browsing, scanning, purchase and third party's data platform
The user of confession other platforms usage record and information, from primary attribute (gender, age, schooling, ethnic group, languages, state
Family, national, occupation, region, industry ...), economic attribution (income, disposable income, susceptibility ... of paying), cultural feature
(intellectual level, locating cultural circle, hobby culture, individual demand ...), (friend-making demand, heterosexual contact demand, returns community attribute
Category demand, leader's demand, cooperate demand ...), hardware attributes (possessing equipment, network condition ...), software attributes (network be familiar with
Degree, software familiarity ...) etc. dimensions user is portrayed, generate detailed user portrait.
S104, judge several first commodity pictures and several second commodity determining according to user operation records
Whether picture matches.
S105, when several first commodity pictures are matched with several second commodity pictures, be based on described second
Commodity picture carries out Video coding to the video block serial number group of acquisition and generates the second video flowing.
Method for processing video frequency provided in an embodiment of the present invention in real time detects the commodity occurred in video, and from from the background
Commodity library recalls corresponding product information.The product features extracted and user characteristics are matched.Orientation pushes commodity
To may be to the interested user of the commodity, and advertisement is shown influence the position of user's viewing experience in corner screen etc.,
Such as the white spaces such as metope, desktop in screen edge, corner or video pictures show advertisement, and it is main not block video
Picture.Wherein, it includes real that method for processing video frequency provided in an embodiment of the present invention, which carries out detection to the commodity occurred in video in real time,
When detection video pictures in vertical plane and real-time detection video in the horizontal plane that occurs.When in real-time detection video pictures
Vertical plane, such as metope.The feature of commodity in advertisement base is matched with the feature of user, advertisement related pages are shown
In the plane；And when the horizontal plane occurred in real-time detection video, such as desktop.By the feature of commodity in advertisement base and user
Feature matched, advertisement correlation 3D model is shown in the plane.
After determining several second commodity pictures according to user operation records, and several first commodity pictures described in interpretation
Before whether being matched with several second commodity pictures determined according to user operation records further include:
Based on convolutional neural networks CNN to the video block detection block of several first commodity pictures and the second commodity picture
Frame directly generates candidate frame using bottom visual signature, parallel computation is utilized on the candidate frame generating mode and in parallel mould
Anchor point position under formula varies one's tactics, and reduces the generation of useless candidate frame, to avoid excessive computing cost.When specific implementation
The method for directly generating candidate frame using bottom visual signature using the target object detection framework based on CNN is based on sliding
The strategy of window.Sliding window can fix an anchor point when generating candidate frame with certain proportion resize-window, then by setting step
Long sliding.
After the generation candidate frame, and several first commodity pictures described in interpretation are determined with according to user operation records
Several second commodity pictures whether match before further include: extract picture feature；The extraction picture feature includes: root
The low-level visual feature table of selective polymerisation is carried out according to the local feature of the first commodity picture and the second commodity picture Scale invariant
Show.From the angle of the potential distinction of local feature region when specific implementation, research interest point sequencing selection algorithm eliminates part
Characteristic noise, the distinction of enhancing polymerization description.
It is described that selective polymerisation is carried out according to the local feature of the first commodity picture and the second commodity picture Scale invariant
Low-level visual feature expression after, and several first commodity pictures described in interpretation with according to user operation records determine described in
Before whether several second commodity pictures match further include: picture feature label；The picture feature label includes: to the first quotient
Product picture and the second commodity picture are expressed based on the high-level semantics features of deep learning.
Especially optimize the feature learning method of convolutional neural networks structure.Not with the feature representation based on Scale invariant
Together, the feature representation based on deep learning has stronger Semantic Similarity, more suitable for expression semanteme or the object of concept level
Similitude between body.Around the method for multi_dimension optimization depth convolutional neural networks model structure when specific implementation, so that depth
The expression that output feature is practised for scene and object has stronger distinction, filter background noise.
Classified using classifier to the first commodity picture and the second commodity picture.In the structure of CNN when specific implementation
Rear portion connects all features by full articulamentum, gives output valve to classifier, such as the SoftMax for more classification problems
Classifier.
The operation note includes: the attribute data of user；
First commodity picture and the second commodity picture include: item property data.Specifically, pass through advertiser's docking platform
The first commodity picture is collected, brand, price, material, purposes, applicable age, classification etc. form commodity portrait.And second
The item property data capture method of commodity picture include: electric business platform directly extract, the direct extraction of the text of commodity details,
Such as crawler technology and based on convolutional neural networks according to visual angle from details figure thumbnail extract.
The second video flowing that the Video coding generates includes: two dimensional code, obtains institute using cell phone application barcode scanning for user
State the item property data of several second commodity pictures.Method for processing video frequency provided in an embodiment of the present invention provides mating APP, branch
Subsequent consumption is held, if user is interested in the commodity in advertisement, APP barcode scanning is can use or is clicked directly in APP and checked
Details further check commodity, complete further consumer behavior.
Method for processing video frequency provided in an embodiment of the present invention utilizes depth learning technology, identifies from live telecast stream specific
Commodity make the use value of user's direct feel commodity in amusement and leisure, and the order that can conveniently place an order, and have overturned tradition
TV interrupts roughly the way of program interruption advertisement, keeps businessman's marketing more humane.In addition, with traditional image recognition technology phase
Than carrying out image using machine learning (Machine Learning, ML) and the convolutional neural networks CNN of deep learning is based on
Identification, can greatly improve the accuracy of image recognition.Algorithm compared to tradition based on feature spot scan, using based on depth
The artificial neural network of study will greatly improve arithmetic speed, and so as to realize second grade operation, reaching can be real for live stream
When the level that handles.
Fig. 2 shows the structural schematic diagrams of the video process apparatus 200 provided according to inventive embodiments.As shown in Fig. 2, packet
It includes:
Video flowing splits module 201, regards for the first video flowing to be split as a series of video blocks, and in order to every section
Frequency block marking serial numbers；
First determining module 202, for determining several first commodity pictures according to the video block；
Second determining module 203, for determining several second commodity pictures according to user operation records；The operation note
It include: the access path for recording user and accessing a certain commodity；
Judgment module 204, for judging several first commodity pictures and according to user operation records determination
Whether several second commodity pictures match；
Video flowing generation module 205 when matching for several first commodity pictures with several second commodity pictures, is based on the
Two commodity pictures carry out Video coding to the video block serial number group of acquisition and generate the second video flowing.
Fig. 3 is another structural schematic diagram of video process apparatus provided in an embodiment of the present invention.As shown in figure 3, the view
Frequency processing device includes: processor 301 and memory 302.
Processor 301 and memory 302 are connected with each other by bus.It is total that the bus can be divided into address bus, data
Line, control bus etc..Only to be indicated with a thick line in Fig. 3, it is not intended that an only bus or a type convenient for indicating
The bus of type.
Memory 302 may include volatile memory (English: volatile memory), such as random access memory
Device (English: random-access memory, abbreviation: RAM)；Memory also may include nonvolatile memory (English:
Non-volatile memory), for example, flash memory (English: flashmemory), hard disk (English: hard disk
Drive, abbreviation: HDD) or solid state hard disk (English: solid-state drive, abbreviation: SSD)；Memory 302 can also wrap
Include the combination of the memory of mentioned kind.
Processor 301 can be central processing unit (English: central processing unit, abbreviation: CPU), network
The combination of processor (English: network processor, abbreviation: NP) or CPU and NP.Processor 301 can also be further
Including hardware chip.Above-mentioned hardware chip can be specific integrated circuit (English: application-specific
Integrated circuit, abbreviation: ASIC), programmable logic device (English: programmable logic device,
Abbreviation: PLD) or combinations thereof.Above-mentioned PLD can be Complex Programmable Logic Devices (English: complex programmable
Logic device, abbreviation: CPLD), field programmable gate array (English: field-programmable gate
Array, abbreviation: FPGA), Universal Array Logic (English: generic array logic, abbreviation: GAL) or any combination thereof.
Optionally, memory 302 can be also used for storage program instruction, and processor 301 calls to be stored in the memory 302
Program instruction, several steps or in which optional embodiment in method shown in Fig. 1 can be executed.
In the above-described embodiments, it all emphasizes particularly on different fields to the description of each embodiment, there is no the portion being described in detail in some embodiment
Point, reference can be made to the related descriptions of other embodiments.
Professional should further appreciate that, described in conjunction with the examples disclosed in the embodiments of the present disclosure
Unit and algorithm steps, can be realized with electronic hardware, computer software, or a combination of the two, hard in order to clearly demonstrate
The interchangeability of part and software generally describes each exemplary composition and step according to function in the above description.
These functions are implemented in hardware or software actually, the specific application and design constraint depending on technical solution.
Professional technician can use different methods to achieve the described function each specific application, but this realization
It should not be considered as beyond the scope of the present invention.
The step of method described in conjunction with the examples disclosed in this document or algorithm, can be held with hardware, processor module
The combination of capable software module or the two is implemented.Software module can be placed in random access memory (RAM), memory, read-only deposit
Reservoir (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technology
In any other form of storage medium well known in field.
Although above having used general explanation and specific embodiment, the present invention is described in detail, at this
On the basis of invention, it can be made some modifications or improvements, this will be apparent to those skilled in the art.Therefore,
These modifications or improvements without departing from theon the basis of the spirit of the present invention are fallen within the scope of the claimed invention.
Claims (10)
1. a kind of method for processing video frequency characterized by comprising
First video flowing is split as a series of video blocks, and in order to every section of video block marking serial numbers；
According to the video block, several first commodity pictures are determined；
Several second commodity pictures are determined according to user operation records；The operation note includes: that record user accesses a certain quotient
The access path of product；
Judge several first commodity pictures with according to user operation records determine several second commodity pictures whether
Matching；
When several first commodity pictures are matched with several second commodity pictures, it is based on second commodity picture pair
The video block serial number group of acquisition carries out Video coding and generates the second video flowing.
2. a kind of method for processing video frequency according to claim 1, which is characterized in that determined according to user operation records several
After second commodity picture, and several first commodity pictures described in interpretation with according to user operation records determine it is described several
Before whether the second commodity picture matches further include:
Based on convolutional neural networks CNN to the video block detection framework benefit of several first commodity pictures and the second commodity picture
Candidate frame is directly generated with bottom visual signature, using parallel computation and in a parallel mode on the candidate frame generating mode
Anchor point position vary one's tactics.
3. a kind of method for processing video frequency according to claim 2, which is characterized in that after the generation candidate frame, and
Whether several first commodity pictures described in interpretation match with several second commodity pictures determined according to user operation records
Before further include: extract picture feature；The extraction picture feature includes: according to the first commodity picture and the second commodity picture ruler
Spend the low-level visual feature expression that constant local feature carries out selective polymerisation.
4. a kind of method for processing video frequency according to claim 3, which is characterized in that described according to the first commodity picture and
After the local feature of two commodity picture Scale invariants carries out the low-level visual feature expression of selective polymerisation, and described in interpretation
Whether several first commodity pictures also wrap before matching with several second commodity pictures determined according to user operation records
It includes: picture feature label；The picture feature label includes: to be based on deep learning to the first commodity picture and the second commodity picture
High-level semantics features expression.
5. a kind of method for processing video frequency according to claim 4, which is characterized in that using classifier to the first commodity picture
Classify with the second commodity picture.
6. a kind of method for processing video frequency according to claim 1, which is characterized in that
The operation note includes: the attribute data of user；
First commodity picture and the second commodity picture include: item property data.
7. a kind of method for processing video frequency according to claim 1 or 6, which is characterized in that
The second video flowing that the Video coding generates includes: two dimensional code, if described using the acquisition of cell phone application barcode scanning for user
The item property data of dry second commodity picture.
8. a kind of video process apparatus characterized by comprising
Video flowing splits module, for the first video flowing to be split as a series of video blocks, and in order to every section of video block mark
Remember serial number；
First determining module, for determining several first commodity pictures according to the video block；
Second determining module, for determining several second commodity pictures according to user operation records；The operation note includes: note
Employ the access path that family accesses a certain commodity；
Judgment module, described several second for judging several first commodity pictures and being determined according to user operation records
Whether commodity picture matches；
Video flowing generation module when matching for several first commodity pictures with several second commodity pictures, is based on the second commodity
Picture carries out Video coding to the video block serial number group of acquisition and generates the second video flowing.
9. a kind of computer-readable medium, is stored thereon with computer program, which is characterized in that described program is executed by processor
Method of the Shi Shixian as described in any one of claims 1 to 7.
10. a kind of electronic equipment characterized by comprising
Several processors；
Storage device, for storing several programs, when several programs are executed by several processors, if so that described
Dry-cure device realizes the method as described in any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811142602.5A CN109068173A (en) | 2018-09-28 | 2018-09-28 | A kind of method for processing video frequency and video process apparatus |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811142602.5A CN109068173A (en) | 2018-09-28 | 2018-09-28 | A kind of method for processing video frequency and video process apparatus |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109068173A true CN109068173A (en) | 2018-12-21 |
Family
ID=64766497
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811142602.5A Pending CN109068173A (en) | 2018-09-28 | 2018-09-28 | A kind of method for processing video frequency and video process apparatus |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109068173A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111598601A (en) * | 2019-09-16 | 2020-08-28 | 郑素娟 | Intelligent database updating selection platform |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103218734A (en) * | 2013-04-01 | 2013-07-24 | 天脉聚源(北京)传媒科技有限公司 | Advertisement information pushing method and device |
| CN105681918A (en) * | 2015-09-16 | 2016-06-15 | 乐视致新电子科技（天津）有限公司 | Method and system for presenting article relevant information in video stream |
| CN105760475A (en) * | 2016-02-14 | 2016-07-13 | 天脉聚源（北京）科技有限公司 | Method and device for recommending related articles in video |
| CN106202317A (en) * | 2016-07-01 | 2016-12-07 | 传线网络科技（上海）有限公司 | Method of Commodity Recommendation based on video and device |
| CN106534944A (en) * | 2016-11-30 | 2017-03-22 | 北京锤子数码科技有限公司 | Video display method and device |
| US20170323307A1 (en) * | 2008-06-30 | 2017-11-09 | Nokia Technologies Oy | Specifying media content placement criteria |
| CN107484030A (en) * | 2017-07-29 | 2017-12-15 | 安徽博威康信息技术有限公司 | Merchandise news obtains system in a kind of viewing video based on spectators' operation |
| CN107730343A (en) * | 2017-09-15 | 2018-02-23 | 广州唯品会研究院有限公司 | A kind of user's merchandise news method for pushing and equipment based on picture attribute extraction |
| CN107995517A (en) * | 2017-12-21 | 2018-05-04 | 知而行（上海）营销咨询有限公司 | A kind of video ads playback method and system based on real-time shopping basket information |
-
2018
- 2018-09-28 CN CN201811142602.5A patent/CN109068173A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20170323307A1 (en) * | 2008-06-30 | 2017-11-09 | Nokia Technologies Oy | Specifying media content placement criteria |
| CN103218734A (en) * | 2013-04-01 | 2013-07-24 | 天脉聚源(北京)传媒科技有限公司 | Advertisement information pushing method and device |
| CN105681918A (en) * | 2015-09-16 | 2016-06-15 | 乐视致新电子科技（天津）有限公司 | Method and system for presenting article relevant information in video stream |
| CN105760475A (en) * | 2016-02-14 | 2016-07-13 | 天脉聚源（北京）科技有限公司 | Method and device for recommending related articles in video |
| CN106202317A (en) * | 2016-07-01 | 2016-12-07 | 传线网络科技（上海）有限公司 | Method of Commodity Recommendation based on video and device |
| CN106534944A (en) * | 2016-11-30 | 2017-03-22 | 北京锤子数码科技有限公司 | Video display method and device |
| CN107484030A (en) * | 2017-07-29 | 2017-12-15 | 安徽博威康信息技术有限公司 | Merchandise news obtains system in a kind of viewing video based on spectators' operation |
| CN107730343A (en) * | 2017-09-15 | 2018-02-23 | 广州唯品会研究院有限公司 | A kind of user's merchandise news method for pushing and equipment based on picture attribute extraction |
| CN107995517A (en) * | 2017-12-21 | 2018-05-04 | 知而行（上海）营销咨询有限公司 | A kind of video ads playback method and system based on real-time shopping basket information |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111598601A (en) * | 2019-09-16 | 2020-08-28 | 郑素娟 | Intelligent database updating selection platform |
| CN111598601B (en) * | 2019-09-16 | 2020-12-15 | 上海岳蒙设计有限公司 | Intelligent database updating selection platform |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| RU2699687C1 (en) | Detecting text fields using neural networks | |
| US9430719B2 (en) | System and method for providing objectified image renderings using recognition information from images | |
| US7809722B2 (en) | System and method for enabling search and retrieval from image files based on recognized information | |
| US8649572B2 (en) | System and method for enabling the use of captured images through recognition | |
| US7809192B2 (en) | System and method for recognizing objects from images and identifying relevancy amongst images and information | |
| US20150242689A1 (en) | System and method for determining graph relationships using images | |
| CN103988202A (en) | Image attractiveness based indexing and searching | |
| CN113762309B (en) | Object matching method, device and equipment | |
| Kadish et al. | Improving object detection in art images using only style transfer | |
| US20160005175A1 (en) | Service provision device, and method | |
| WO2006122164A2 (en) | System and method for enabling the use of captured images through recognition | |
| CN112633431A (en) | Tibetan-Chinese bilingual scene character recognition method based on CRNN and CTC | |
| CN116958957A (en) | Training method of multi-mode feature extraction network and three-dimensional feature representation method | |
| WO2022247403A1 (en) | Keypoint detection method, electronic device, program, and storage medium | |
| Yousaf et al. | Patch-CNN: deep learning for logo detection and brand recognition | |
| Intasuwan et al. | Text and object detection on billboards | |
| CN109068173A (en) | A kind of method for processing video frequency and video process apparatus | |
| Ye et al. | Saliency detection via similar image retrieval | |
| US9836799B2 (en) | Service provision program | |
| WO2022241987A1 (en) | Image retrieval method and apparatus | |
| JP6840811B1 (en) | Questionnaire input device, questionnaire input method and questionnaire input program | |
| Zhao et al. | Automatic generation of informative video thumbnail | |
| US20210012478A1 (en) | System and method for assessing quality of media files | |
| Movshovitz-Attias | Dataset curation through renders and ontology matching | |
| CN116956052A (en) | Application matching method and application matching device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181221 |