CN109922379A - Advertisement video optimization method, device, equipment and computer readable storage medium - Google Patents
Advertisement video optimization method, device, equipment and computer readable storage medium Download PDFInfo
- Publication number
- CN109922379A CN109922379A CN201910140441.4A CN201910140441A CN109922379A CN 109922379 A CN109922379 A CN 109922379A CN 201910140441 A CN201910140441 A CN 201910140441A CN 109922379 A CN109922379 A CN 109922379A
- Authority
- CN
- China
- Prior art keywords
- video
- key
- frame
- advertisement
- candidate key
- 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.)
- Granted
Links
- 238000005457 optimization Methods 0.000 title claims abstract description 56
- 238000000034 method Methods 0.000 title claims abstract description 46
- 238000003860 storage Methods 0.000 title claims abstract description 17
- 238000000605 extraction Methods 0.000 claims abstract description 26
- 238000011156 evaluation Methods 0.000 claims description 7
- 238000012544 monitoring process Methods 0.000 claims description 7
- 230000011218 segmentation Effects 0.000 claims description 5
- 239000000284 extract Substances 0.000 claims description 4
- 239000012634 fragment Substances 0.000 claims description 3
- 238000013507 mapping Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 7
- 238000004891 communication Methods 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 238000001514 detection method Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000012015 optical character recognition Methods 0.000 description 2
- 230000000306 recurrent effect Effects 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 230000003796 beauty Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000005577 local transmission Effects 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000006855 networking Effects 0.000 description 1
- 230000001537 neural effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000001737 promoting effect Effects 0.000 description 1
- 238000000638 solvent extraction Methods 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
-
- 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
-
- 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
- H04N21/83—Generation or processing of protective or descriptive data associated with content; Content structuring
- H04N21/84—Generation or processing of descriptive data, e.g. content descriptors
- H04N21/8405—Generation or processing of descriptive data, e.g. content descriptors represented by keywords
-
- 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
- H04N21/83—Generation or processing of protective or descriptive data associated with content; Content structuring
- H04N21/845—Structuring of content, e.g. decomposing content into time segments
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Two-Way Televisions, Distribution Of Moving Picture Or The Like (AREA)
Abstract
The invention discloses a kind of advertisement video optimization method, device, equipment and computer readable storage mediums, this method comprises: advertisement video to be optimized is divided into several advertisement video segments, and the extraction of key video sequence frame is carried out to each advertisement video segment by preset video frame extraction algorithm, obtain several candidate key video frames；Key indices assessment algorithm is obtained, and according to the key indices assessment algorithm, assesses each candidate key video frame in several candidate key video frames, obtains the target critical index of each candidate key video frame；According to the target critical index of each candidate key video frame, the selection target key video sequence frame from several candidate key video frames；The preposition strategy of key video sequence frame is obtained, and based on the preposition strategy of the key video sequence frame and the target critical video frame, the advertisement video is optimized, targeted advertisements video is obtained.The present invention can fast and accurately optimize advertisement video.
Description
Technical field
The present invention relates to the technical field of internet more particularly to a kind of advertisement video optimization method, device, equipment and meters
Calculation machine readable storage medium storing program for executing.
Background technique
With financial technology (Fintech), the especially fast development of internet finance, more and more enterprises are gradually adopted
Brand and product is promoted with the mode on line, the advertisement that can be launched by major advertising platform for promoting brand and product regards
Frequently, more people can be allowed to watch advertisement video, understand brand and product convenient for everybody, can be used when user needs product
Family can buy product by relevant channel.During launching advertisement video, when having skipped advertisement video there are user,
Brand and product also also fails to the problem of showing, the dispensing effect of advertisement video is poor, and advertising man needs to optimize wide thus
Accuse video.
However, advertising man needs to conceive again advertisement optimization direction before optimizing advertisement video, optimize determining
Advertisement video is just started from after direction, the fabrication cycle of advertisement video is longer, need to expend more human cost and when
Between cost, simultaneously as the optimization direction of advertisement is that advertising man artificially conceives, there are uncertainties, so that after optimization
Advertisement video cannot still obtain it is preferable launch effect, can not fast and accurately optimize advertisement video.Therefore, how quickly quasi-
True optimization advertisement video is current urgent problem to be solved.
Summary of the invention
The main purpose of the present invention is to provide a kind of advertisement video optimization method, device, equipment and computer-readable deposit
Storage media, it is intended to fast and accurately optimize advertisement video.
To achieve the above object, the present invention provides a kind of advertisement video optimization method, the advertisement video optimization method packet
Include following steps:
Advertisement video to be optimized is divided into several advertisement video segments, and passes through preset video frame extraction algorithm pair
Each advertisement video segment carries out the extraction of key video sequence frame, obtains several candidate key video frames；
Key indices assessment algorithm is obtained, and according to the key indices assessment algorithm, assesses several candidate keys
Each candidate key video frame in video frame, obtains the target critical index of each candidate key video frame；
According to the target critical index of each candidate key video frame, selected from several candidate key video frames
Select target critical video frame；
The preposition strategy of key video sequence frame is obtained, and based on the preposition strategy of the key video sequence frame and the target critical video
Frame optimizes the advertisement video, obtains targeted advertisements video.
Further, it according to the key indices assessment algorithm, assesses each in several candidate key video frames
Candidate key video frame, the step of obtaining the target critical index of each candidate key video frame include:
Obtain the text information that each candidate key video frame includes in several candidate key video frames；
According to the key indices assessment algorithm and the text information, determine that the target of each candidate key video frame is closed
Key index.
Further, according to the key indices assessment algorithm and the text information, each candidate key video is determined
The step of target critical index of frame includes:
The entropy for the text information that each candidate key video frame includes is calculated, and calculates each candidate key view
The similarity of text information and default text information that frequency frame includes；
According to the key indices assessment algorithm, the entropy of the text information and the similarity, each candidate pass is determined
The target critical index of key video frame.
Further, the text information that each candidate key video frame includes in several candidate key video frames is obtained
The step of after, further includes:
Calculate the entropy for the text information that each candidate key video frame includes；
Based on default aesthetic score prediction model, each candidate key video in several candidate key video frames is predicted
The aesthetic score of frame；
According to the key indices assessment algorithm, the entropy and the aesthetic score of the text information, determine described each
The target critical index of candidate key video frame.
Further, described based on default aesthetic score prediction model, it predicts every in several candidate key video frames
After the step of aesthetic score of one candidate key video frame, further includes:
Calculate the similarity of text information and default text information that each candidate key video frame includes；
According to the key indices assessment algorithm, the entropy of the text information, the similarity and the aesthetic score, really
The target critical index of fixed each candidate key video frame.
Further, it before described the step of advertisement video to be optimized is divided into several advertisement video segments, also wraps
It includes:
When monitoring algorithm policy more new command, according to the algorithm policy more new command, it is anti-to obtain corresponding advertisement
Present data, key indices assessment algorithm to be updated and the preposition strategy of key video sequence frame；
According to the advertising feedback data, key indices assessment algorithm to be updated and the preposition strategy of key video sequence frame are held
Row updates operation.
Further, according to the advertising feedback data, to key indices assessment algorithm and key video sequence frame to be updated
Preposition strategy execution updates the step of operation and includes:
Preset algorithm policy more new model is obtained, and the advertising feedback data are input to the algorithm policy and are updated
Model；
Obtain the first parameter for being used to update the key indices assessment algorithm of the algorithm policy more new model output
Value and the second parameter value for updating the preposition strategy of the key video sequence frame；
The key indices assessment algorithm is updated according to first parameter value, and is updated according to second parameter value
The preposition strategy of key video sequence frame.
In addition, to achieve the above object, the present invention also provides a kind of advertisement videos to optimize device, the advertisement video optimization
Device includes:
Fragment segmentation module, for advertisement video to be optimized to be divided into several advertisement video segments；
Video frame extraction module, it is crucial for being carried out by preset video frame extraction algorithm to each advertisement video segment
Video frame extraction obtains several candidate key video frames；
Key indices evaluation module, for obtaining key indices assessment algorithm, and according to the key indices assessment algorithm,
Each candidate key video frame in several candidate key video frames is assessed, the target of each candidate key video frame is obtained
Key indices；
Selecting module, for the target critical index according to each candidate key video frame, from several candidates
Selection target key video sequence frame in key video sequence frame；
Advertisement video optimization module, for obtaining the preposition strategy of key video sequence frame, and it is preposition based on the key video sequence frame
The tactful and described target critical video frame, optimizes the advertisement video, obtains targeted advertisements video.
In addition, to achieve the above object, the present invention also provides a kind of advertisement videos to optimize equipment, the advertisement video optimization
Equipment includes: memory, processor and to be stored in the advertisement video that can be run on the memory and on the processor excellent
Change program, the advertisement video optimization program realizes advertisement video optimization method as described above when being executed by the processor
Step.
The present invention also provides a kind of computer readable storage medium, advertisement is stored on the computer readable storage medium
Video optimized program, the advertisement video optimization program realize advertisement video optimization method as described above when being executed by processor
The step of.
The present invention provides a kind of advertisement video optimization method, device, equipment and computer readable storage medium, and the present invention will
Advertisement video is divided into several advertisement video segments, and carries out the extraction of key video sequence frame to each advertisement video segment, if obtaining
Dry candidate key video frame, is then based on key indices assessment algorithm, assesses each candidate key video frame, obtains every
The target critical index of one candidate key video frame, and being based on, the target critical index of each candidate key video frame, from several
Selection target key video sequence frame in candidate key video frame, finally based on the preposition strategy of key video sequence frame and the target critical video
Frame optimizes the advertisement video, obtains targeted advertisements video, and whole process does not need advertising man and conceives optimization again
Direction can not need to expend the more time yet and remove video of preparing advertising, while base to avoid the uncertainty of advertising man
In the extraction of key video sequence frame and the preposition optimization advertisement video of key video sequence frame, it can fast and accurately optimize advertisement video, it can also
To improve the dispensing effect of advertisement video.
Detailed description of the invention
Fig. 1 is the device structure schematic diagram for the hardware running environment that the embodiment of the present invention is related to；
Fig. 2 is the flow diagram of advertisement video optimization method first embodiment of the present invention；
Fig. 3 is the flow diagram of advertisement video optimization method fourth embodiment of the present invention；
Fig. 4 is the functional block diagram that advertisement video of the present invention optimizes device first embodiment.
The embodiments will be further described with reference to the accompanying drawings for the realization, the function and the advantages of the object of the present invention.
Specific embodiment
It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, it is not intended to limit the present invention.
As shown in Figure 1, Fig. 1 is the device structure schematic diagram for the hardware running environment that the embodiment of the present invention is related to.
Advertisement video of embodiment of the present invention optimization equipment can be PC, be also possible to smart phone, tablet computer, portable meter
The packaged type terminal device having a display function such as calculation machine.
As shown in Figure 1, advertisement video optimization equipment may include: processor 1001, such as CPU, communication bus 1002,
User interface 1003, network interface 1004, memory 1005.Wherein, communication bus 1002 is for realizing between these components
Connection communication.User interface 1003 may include display screen (Display), input unit such as keyboard (Keyboard), optional
User interface 1003 can also include standard wireline interface and wireless interface.Network interface 1004 optionally may include mark
Wireline interface, the wireless interface (such as WI-FI interface) of standard.Memory 1005 can be high speed RAM memory, be also possible to stablize
Memory (non-volatile memory), such as magnetic disk storage.Memory 1005 optionally can also be independently of preceding
State the storage device of processor 1001.
It will be understood by those skilled in the art that the optimization device structure of advertisement video shown in Fig. 1 is not constituted to advertisement
The restriction of video optimized equipment may include perhaps combining certain components or different than illustrating more or fewer components
Component layout.
As shown in Figure 1, as may include that operating system, network are logical in a kind of memory 1005 of computer storage medium
Believe that module, Subscriber Interface Module SIM and advertisement video optimize program.
In advertisement video optimization equipment shown in Fig. 1, network interface 1004 is mainly used for connecting background server, and rear
Platform server carries out data communication；User interface 1003 is mainly used for connecting client (user terminal), carries out data with client
Communication；And processor 1001 can be used for that the advertisement video stored in memory 1005 is called to optimize program, and execute following step
It is rapid:
Advertisement video to be optimized is divided into several advertisement video segments, and passes through preset video frame extraction algorithm pair
Each advertisement video segment carries out the extraction of key video sequence frame, obtains several candidate key video frames；
Key indices assessment algorithm is obtained, and according to the key indices assessment algorithm, assesses several candidate keys
Each candidate key video frame in video frame, obtains the target critical index of each candidate key video frame；
According to the target critical index of each candidate key video frame, selected from several candidate key video frames
Select target critical video frame；
The preposition strategy of key video sequence frame is obtained, and based on the preposition strategy of the key video sequence frame and the target critical video
Frame optimizes the advertisement video, obtains targeted advertisements video.
Further, processor 1001 can be used for that the advertisement video stored in memory 1005 is called to optimize program, also
Execute following steps:
Obtain the text information that each candidate key video frame includes in several candidate key video frames；
According to the key indices assessment algorithm and the text information, determine that the target of each candidate key video frame is closed
Key index.
Further, processor 1001 can be used for that the advertisement video stored in memory 1005 is called to optimize program, also
Execute following steps:
The entropy for the text information that each candidate key video frame includes is calculated, and calculates each candidate key view
The similarity of text information and default text information that frequency frame includes；
According to the key indices assessment algorithm, the entropy of the text information and the similarity, each candidate pass is determined
The target critical index of key video frame.
Further, processor 1001 can be used for that the advertisement video stored in memory 1005 is called to optimize program, also
Execute following steps:
Calculate the entropy for the text information that each candidate key video frame includes；
Based on default aesthetic score prediction model, each candidate key video in several candidate key video frames is predicted
The aesthetic score of frame；
According to the key indices assessment algorithm, the entropy and the aesthetic score of the text information, determine described each
The target critical index of candidate key video frame.
Further, processor 1001 can be used for that the advertisement video stored in memory 1005 is called to optimize program, also
Execute following steps:
Calculate the similarity of text information and default text information that each candidate key video frame includes；
According to the key indices assessment algorithm, the entropy of the text information, the similarity and the aesthetic score, really
The target critical index of fixed each candidate key video frame.
Further, processor 1001 can be used for that the advertisement video stored in memory 1005 is called to optimize program, also
Execute following steps:
When monitoring algorithm policy more new command, according to the algorithm policy more new command, it is anti-to obtain corresponding advertisement
Present data, key indices assessment algorithm to be updated and the preposition strategy of key video sequence frame；
According to the advertising feedback data, key indices assessment algorithm to be updated and the preposition strategy of key video sequence frame are held
Row updates operation.
Further, processor 1001 can be used for that the advertisement video stored in memory 1005 is called to optimize program, also
Execute following steps:
Preset algorithm policy more new model is obtained, and the advertising feedback data are input to the algorithm policy and are updated
Model；
Obtain the first parameter for being used to update the key indices assessment algorithm of the algorithm policy more new model output
Value and the second parameter value for updating the preposition strategy of the key video sequence frame；
The key indices assessment algorithm is updated according to first parameter value, and is updated according to second parameter value
The preposition strategy of key video sequence frame.
Wherein, the specific embodiment and following advertisement video optimization methods of advertisement video optimization equipment of the present invention is each specific
Embodiment is essentially identical, and therefore not to repeat here.
The present invention provides a kind of advertisement video optimization method.
It is the flow diagram of advertisement video optimization method first embodiment of the present invention referring to Fig. 2, Fig. 2.
In the present embodiment, which includes:
Advertisement video to be optimized is divided into several advertisement video segments, and passes through preset video frame by step S101
Extraction algorithm carries out the extraction of key video sequence frame to each advertisement video segment, obtains several candidate key video frames；
In the present embodiment, which is applied to advertisement video and optimizes equipment, and advertisement video optimization is set
It is standby to be chosen as equipment shown in FIG. 1, advertisement video optimization program is installed in advertisement video optimization equipment, it is wide when needing to optimize
When accusing video, the advertisement video optimized can will be needed to be transferred in advertisement video optimization equipment, the mode of transmission includes networking
It transmits and (optimizes in equipment as advertisement video is transferred to advertisement video by data network or wireless network) and local transmission is (such as
Advertisement video advertisement video is transferred to by data connecting line, USB flash disk or hard disk etc. to optimize in equipment), user, which can choose, to be needed
The advertisement video to be optimized, after user selects the advertisement video for needing to optimize, which optimizes equipment and obtains to excellent
Advertisement video to be optimized is divided into several advertisement video segments, and passes through preset video frame extraction by the advertisement video of change
Algorithm carries out the extraction of key video sequence frame to each advertisement video segment, obtains several candidate key video frames.
Wherein, the partitioning scheme of advertisement video is Shot border detection (Shot boundary detection, SBD), is led to
Advertisement video can be divided into several independent advertisement video segments by crossing Shot border detection, and each advertisement video segment by
Multiple video frame compositions.Shot border detection algorithm includes but is not limited to color histogram based on video frame, is based on video frame
Brightness value and edge feature based on video frame.Preset video frame extraction algorithm includes but is not limited to clustering algorithm and recurrence
Neural network algorithm specifically extracts corresponding feature vector, and be based on clustering algorithm or recurrent neural from each video frame
Network algorithm and the feature vector extracted carry out the extraction of key video sequence frame to each advertisement video segment, obtain several times
Select key video sequence frame.Wherein, clustering algorithm does not need largely have the video data of mark to be trained, and it is less to be suitable for sample number
Or the not no scene of labeled data, and recurrent neural network algorithm needs largely have the video data of mark to be trained, it is suitable
Sample number is more and has the scene of labeled data, and different algorithms can be selected based on actual application scenarios.
Step S102 obtains key indices assessment algorithm, and according to the key indices assessment algorithm, assesses described several
Each candidate key video frame in candidate key video frame, obtains the target critical index of each candidate key video frame；
In the present embodiment, after obtaining several candidate key video frames, advertisement video optimization equipment obtains key and refers to
Number assessment algorithm, and according to the key indices assessment algorithm, assess each candidate key view in several candidate key video frames
Frequency frame obtains the target critical index of each candidate key video frame.Specifically, it obtains each in several candidate key video frames
The text information that candidate key video frame includes, and according to the key indices assessment algorithm and the text information, determine each time
The target critical index for selecting key video sequence frame calculates the entropy (information for the text information that each candidate key video frame includes
Amount), and the entropy of the text information prestored and the mapping table of key indices are inquired, include by each candidate key video frame
The corresponding key indices of the entropy of text information are determined as the target critical index of each candidate key video frame, need to illustrate
It is that the entropy of text information is bigger, then key indices are higher, the entropy of text information is smaller, then key indices are lower.Wherein, candidate
The specific acquisition modes for the text information that key video sequence frame includes are to pass through OCR (Optical Character
Recognition, optical character identification) algorithm, the text information that each candidate key video frame includes is obtained, i.e., to each time
It selects key video sequence frame to be denoised, the processing such as binaryzation, Slant Rectify, Character segmentation and identification, obtains candidate key video frame
In text information.
Specifically, the specific method of determination of the target critical index of candidate key video frame can also be each candidate of calculating
The entropy for the text information that key video sequence frame includes, and calculate text information and default text that each candidate key video frame includes
The similarity of information determines each candidate key then according to the key indices assessment algorithm, the entropy of text information and similarity
The target critical index of video frame, that is, inquire the entropy of the text information prestored and the mapping table of key indices, obtains each
First key indices of candidate key video frame, and the mapping table of the similarity and key indices that prestore is inquired, it obtains
Then second key indices of each candidate key video frame obtain the first weight coefficient and from key indices assessment algorithm
Two weight coefficients, and each candidate is obtained multiplied by the first weight coefficient with the first key indices of each candidate key video frame
First weight key indices of key video sequence frame, with the second key indices of each candidate key video frame multiplied by the second weight system
Number, obtains the second weight key indices of each candidate key video frame, and the first weight of each candidate key video frame is closed
Key index is the sum of corresponding with the second weight key indices of each candidate key video frame, is determined as each candidate key video
The target critical index of frame.It should be noted that the sum of the first weight coefficient and the second weight coefficient are 1, and default text letter
Breath can be configured based on actual conditions, and the present embodiment is not especially limited this.
Step S103, according to the target critical index of each candidate key video frame, from several candidate keys
Selection target key video sequence frame in video frame；
In the present embodiment, after the key indices for obtaining each candidate key video frame, which optimizes equipment
According to the target critical index of each candidate key video frame, the selection target key video sequence from several candidate key video frames
Frame, i.e., the highest candidate key video frame of selection target key indices is regarded as target critical from several candidate key video frames
Frequency frame.
Step S104 obtains the preposition strategy of key video sequence frame, and based on the preposition strategy of the key video sequence frame and the mesh
Key video sequence frame is marked, the advertisement video is optimized, targeted advertisements video is obtained.
In the present embodiment, after obtaining target critical video frame, which optimizes equipment and obtains key video sequence frame
Preposition strategy, and based on the preposition strategy of the key video sequence frame and the target critical video frame, advertisement video is optimized, is obtained
Targeted advertisements video is based on the preposition strategy of key video sequence frame, in the preposition target critical of the corresponding position of the advertisement video
Video frame, and corresponding play time is set to the target critical video frame, to optimize advertisement video.Wherein, the key video sequence
The preposition strategy of frame includes the play time of preposition length and video frame, it should be noted that the preposition strategy of key video sequence frame can
It is configured based on actual conditions, the present embodiment is not especially limited this.
In the present embodiment, advertisement video is divided into several advertisement video segments by the present invention, and to each advertisement video piece
Duan Jinhang key video sequence frame extracts, and obtains several candidate key video frames, is then based on key indices assessment algorithm, to each time
It selects key video sequence frame to be assessed, obtains the target critical index of each candidate key video frame, and be based on, each candidate key
The target critical index of video frame, the selection target key video sequence frame from several candidate key video frames, finally based on crucial view
The preposition strategy of frequency frame and the target critical video frame, optimize the advertisement video, obtain targeted advertisements video, whole process
Do not need advertising man conceive again optimization direction, can to avoid the uncertainty of advertising man, do not need yet expend compared with
More time removes video of preparing advertising, while being extracted and the preposition optimization advertisement video of key video sequence frame, energy based on key video sequence frame
It is enough fast and accurately to optimize advertisement video, it can also be improved the dispensing effect of advertisement video.
Further, it is based on above-mentioned first embodiment, proposes the second embodiment of advertisement video optimization method of the present invention,
Difference with previous embodiment is that the specific method of determination of the target critical index of each candidate key video frame can also be
After getting the text information that each candidate key video frame includes, which optimizes equipment and calculates each candidate pass
The entropy for the text information that key video frame includes, and based on default aesthetic score prediction model, predict several candidate key video frames
In each candidate key video frame aesthetic score, then according to the key indices assessment algorithm, the entropy of text information and the beauty
Credit number determines the target critical index of each candidate key video frame, that is, the entropy and key for inquiring the text information prestored refer to
Several mapping tables, obtains the first key indices of each candidate key video frame, and the aesthetic score that prestores of inquiry with
The mapping table of key indices obtains the third key indices of each candidate key video frame, then assesses from key indices
Obtain the first weight coefficient and third weight coefficient in algorithm, and with the first key indices of each candidate key video frame multiplied by
First weight coefficient obtains the first weight key indices of each candidate key video frame, with each candidate key video frame
Third key indices obtain the third weight key indices of each candidate key video frame multiplied by third weight coefficient, finally will
The third weight key indices of the first weight key indices and each candidate key video frame of each candidate key video frame
It is the sum of corresponding, it is determined as the target critical index of each candidate key video frame.It should be noted that the first weight coefficient and
The sum of three weight coefficients are 1, and the occurrence of the first weight coefficient and third weight coefficient can be arranged based on actual conditions, this reality
It applies example and this is not especially limited.
Wherein, it can train to obtain aesthetic score prediction model by the video sample data of big data quantity, the aesthetics point
Number prediction model can be arranged based on actual conditions, and the present embodiment is not especially limited this, specifically, regard to each candidate key
Frequency frame carries out feature extraction, obtains the aesthetic features of each candidate key video frame needed for aesthetic score prediction model, and will
The aesthetic features of each candidate key video frame are input to aesthetic score prediction model, available each candidate key video frame
Aesthetic score.
In the present embodiment, the present invention is based on the entropys and candidate key video frame of the text information that candidate key video frame includes
Aesthetic score, comprehensive assessment candidate key video frame obtains the target critical index of candidate key video frame, can greatly
The confidence level for improving target critical index, accurately finds out target critical video frame convenient for subsequent.
Further, it is based on above-mentioned second embodiment, proposes the 3rd embodiment of advertisement video optimization method of the present invention,
Difference with previous embodiment is, after prediction obtains the aesthetic score of each candidate key video frame, the advertisement video
Optimization equipment calculates the similarity of text information and default text information that each candidate key video frame includes, and according to the pass
Key Index Assessment algorithm, the entropy of text information, similarity and aesthetic score determine the target critical of each candidate key video frame
Index inquires the entropy of the text information prestored and the mapping table of key indices, obtain each candidate key video frame
First key indices, and the mapping table of the similarity and key indices that prestore is inquired, obtain each candidate key video
Second key indices of frame, and the mapping table of the aesthetic score and key indices that prestore is inquired, obtain each candidate pass
Then the third key indices of key video frame obtain the first weight coefficient, the second weight coefficient from key indices assessment algorithm
With third weight coefficient, and obtained each with the first key indices of each candidate key video frame multiplied by the first weight coefficient
First weight key indices of candidate key video frame, with the second key indices of each candidate key video frame multiplied by the second power
Weight coefficient, obtains the second weight key indices of each candidate key video frame, and with the of each candidate key video frame
Three key indices obtain the third weight key indices of each candidate key video frame multiplied by third weight coefficient, finally will be every
First weight key indices of one candidate key video frame, each candidate key video frame the second weight key indices with it is each
The sum of third weight key indices of candidate key video frame are determined as the target critical index of each candidate key video frame.
It should be noted that the sum of the first weight coefficient, the second weight coefficient and third weight coefficient are 1, and the first weight coefficient, the
The occurrence of two weight coefficients and third weight coefficient can be configured based on actual conditions, and the present embodiment does not limit this specifically
It is fixed.
In the present embodiment, the entropy for the text information for including the present invention is based on candidate key video frame, candidate key video frame
Aesthetic score and the candidate key video frame text information that includes and default text information similarity, comprehensive assessment candidate closes
Key video frame obtains the target critical index of candidate key video frame, can further improve the credible of target critical index
Degree, accurately finds out target critical video frame convenient for subsequent.
Further, referring to Fig. 3, it is based on above-mentioned first, second or third embodiment, it is excellent to propose advertisement video of the present invention
The fourth embodiment of change method, the difference with previous embodiment are, before step S101, further includes:
Step S105, according to the algorithm policy more new command, is obtained and is corresponded to when monitoring algorithm policy more new command
Advertising feedback data, key indices assessment algorithm to be updated and the preposition strategy of key video sequence frame；
Step S106, according to the advertising feedback data, to key indices assessment algorithm and key video sequence frame to be updated
Preposition strategy execution updates operation.
In the present embodiment, before advertisement video to be optimized is divided into several advertisement video segments, there are needs more
The situation of new key indices assessment algorithm and the preposition strategy of key video sequence frame, for this purpose, when monitoring algorithm policy more new command,
The advertisement video optimizes equipment according to the algorithm policy more new command, obtains corresponding advertising feedback data, key to be updated
Index Assessment algorithm and the preposition strategy of key video sequence frame comment key indices to be updated then according to the advertising feedback data
Estimation algorithm and the preposition strategy execution of key video sequence frame update operation.Wherein, advertising feedback data include but is not limited to clicking rate, use
Family registration rate, brand touching watch duration up to rate, product buying rate and advertisement.
Specifically, preset algorithm policy more new model is obtained, and the advertising feedback data are input to algorithm policy more
Then new model obtains the first parameter value for being used to update the key indices assessment algorithm of the algorithm policy more new model output
And the second parameter value for updating the preposition strategy of the key video sequence frame, finally key indices are updated according to first parameter value
Assessment algorithm, and the preposition strategy of key video sequence frame is updated according to the second parameter value.Wherein, the algorithm policy more new model is
It is obtained by the training of the advertising feedback data of big data quantity, the first parameter value and the second parameter value, specific training method can
It is configured based on actual conditions, the present embodiment is not especially limited this, and the first parameter value includes the first weight coefficient
Value, the value of the value of the second weight coefficient and third weight coefficient, when the second parameter value includes the value and video playing of preposition length
Between value, i.e., key indices assessment algorithm is updated, is then to the first weight coefficient in key indices assessment algorithm
The value of value, the value of the second weight coefficient and third weight coefficient is updated, and is updated to the preposition strategy of key video sequence frame, then
Be it is preposition to key video sequence frame strategy in the value of preposition length and the value of video playback time be updated.
In the present embodiment, the present invention is before optimizing advertisement video, before key indices assessment algorithm and key video sequence frame
It sets strategy to be updated, using more preferably key indices assessment algorithm and the preposition strategy of key video sequence frame, can accurately optimize
Advertisement video.
The present invention also provides a kind of advertisement videos to optimize device.
It is the functional block diagram that advertisement video of the present invention optimizes device first embodiment referring to Fig. 4, Fig. 4.
In the present embodiment, advertisement video optimization device includes:
Fragment segmentation module 101, for advertisement video to be optimized to be divided into several advertisement video segments；
Video frame extraction module 102, for being carried out by preset video frame extraction algorithm to each advertisement video segment
Key video sequence frame extracts, and obtains several candidate key video frames；
Key indices evaluation module 103 for obtaining key indices assessment algorithm, and is assessed according to the key indices and is calculated
Method assesses each candidate key video frame in several candidate key video frames, obtains each candidate key video frame
Target critical index；
Selecting module 104, for the target critical index according to each candidate key video frame, from several times
Select selection target key video sequence frame in key video sequence frame；
Advertisement video optimization module 105, for obtaining the preposition strategy of key video sequence frame, and based on before the key video sequence frame
Strategy and the target critical video frame are set, the advertisement video is optimized, targeted advertisements video is obtained.
Further, the key indices evaluation module 103 is also used to:
Obtain the text information that each candidate key video frame includes in several candidate key video frames；
According to the key indices assessment algorithm and the text information, determine that the target of each candidate key video frame is closed
Key index.
Further, the key indices evaluation module 103 is also used to:
The entropy for the text information that each candidate key video frame includes is calculated, and calculates each candidate key view
The similarity of text information and default text information that frequency frame includes；
According to the key indices assessment algorithm, the entropy of the text information and the similarity, each candidate pass is determined
The target critical index of key video frame.
Further, the key indices evaluation module 103 is also used to:
Calculate the entropy for the text information that each candidate key video frame includes；
Based on default aesthetic score prediction model, each candidate key video in several candidate key video frames is predicted
The aesthetic score of frame；
According to the key indices assessment algorithm, the entropy and the aesthetic score of the text information, determine described each
The target critical index of candidate key video frame.
Further, the key indices evaluation module 103 is also used to:
Calculate the similarity of text information and default text information that each candidate key video frame includes；
According to the key indices assessment algorithm, the entropy of the text information, the similarity and the aesthetic score, really
The target critical index of fixed each candidate key video frame.
Further, the advertisement video optimizes device further include:
Module is obtained, for according to the algorithm policy more new command, obtaining when monitoring algorithm policy more new command
Corresponding advertising feedback data, key indices assessment algorithm to be updated and the preposition strategy of key video sequence frame；
Update module, for being regarded to key indices assessment algorithm and key to be updated according to the advertising feedback data
The preposition strategy execution of frequency frame updates operation.
Further, the update module is also used to:
Preset algorithm policy more new model is obtained, and the advertising feedback data are input to the algorithm policy and are updated
Model；
Obtain the first parameter for being used to update the key indices assessment algorithm of the algorithm policy more new model output
Value and the second parameter value for updating the preposition strategy of the key video sequence frame；
The key indices assessment algorithm is updated according to first parameter value, and is updated according to second parameter value
The preposition strategy of key video sequence frame.
Wherein, the specific embodiment of advertisement video optimization device of the present invention and above-mentioned each embodiment of advertisement video optimization method
Essentially identical, therefore not to repeat here.
In addition, the embodiment of the present invention also proposes a kind of computer readable storage medium, the computer readable storage medium
On be stored with advertisement video optimization program and execute following steps when advertisement video optimization program is executed by processor:
Advertisement video to be optimized is divided into several advertisement video segments, and passes through preset video frame extraction algorithm pair
Each advertisement video segment carries out the extraction of key video sequence frame, obtains several candidate key video frames；
Key indices assessment algorithm is obtained, and according to the key indices assessment algorithm, assesses several candidate keys
Each candidate key video frame in video frame, obtains the target critical index of each candidate key video frame；
According to the target critical index of each candidate key video frame, selected from several candidate key video frames
Select target critical video frame；
The preposition strategy of key video sequence frame is obtained, and based on the preposition strategy of the key video sequence frame and the target critical video
Frame optimizes the advertisement video, obtains targeted advertisements video.
Further, when the advertisement video optimization program is executed by processor, following steps are also executed:
Obtain the text information that each candidate key video frame includes in several candidate key video frames；
According to the key indices assessment algorithm and the text information, determine that the target of each candidate key video frame is closed
Key index.
Further, when the advertisement video optimization program is executed by processor, following steps are also executed:
The entropy for the text information that each candidate key video frame includes is calculated, and calculates each candidate key view
The similarity of text information and default text information that frequency frame includes；
According to the key indices assessment algorithm, the entropy of the text information and the similarity, each candidate pass is determined
The target critical index of key video frame.
Further, when the advertisement video optimization program is executed by processor, following steps are also executed:
Calculate the entropy for the text information that each candidate key video frame includes；
Based on default aesthetic score prediction model, each candidate key video in several candidate key video frames is predicted
The aesthetic score of frame；
According to the key indices assessment algorithm, the entropy and the aesthetic score of the text information, determine described each
The target critical index of candidate key video frame.
Further, when the advertisement video optimization program is executed by processor, following steps are also executed:
Calculate the similarity of text information and default text information that each candidate key video frame includes；
According to the key indices assessment algorithm, the entropy of the text information, the similarity and the aesthetic score, really
The target critical index of fixed each candidate key video frame
Further, when the advertisement video optimization program is executed by processor, following steps are also executed:
When monitoring algorithm policy more new command, according to the algorithm policy more new command, it is anti-to obtain corresponding advertisement
Present data, key indices assessment algorithm to be updated and the preposition strategy of key video sequence frame；
According to the advertising feedback data, key indices assessment algorithm to be updated and the preposition strategy of key video sequence frame are held
Row updates operation.
Further, when the advertisement video optimization program is executed by processor, following steps are also executed:
Preset algorithm policy more new model is obtained, and the advertising feedback data are input to the algorithm policy and are updated
Model；
Obtain the first parameter for being used to update the key indices assessment algorithm of the algorithm policy more new model output
Value and the second parameter value for updating the preposition strategy of the key video sequence frame；
The key indices assessment algorithm is updated according to first parameter value, and is updated according to second parameter value
The preposition strategy of key video sequence frame.
Wherein, the specific embodiment of computer readable storage medium of the present invention is respectively implemented with above-mentioned advertisement video optimization method
Example is essentially identical, and therefore not to repeat here.
It should be noted that, in this document, the terms "include", "comprise" or its any other variant are intended to non-row
His property includes, so that the process, method, article or the system that include a series of elements not only include those elements, and
And further include other elements that are not explicitly listed, or further include for this process, method, article or system institute it is intrinsic
Element.In the absence of more restrictions, the element limited by sentence "including a ...", it is not excluded that including being somebody's turn to do
There is also other identical elements in the process, method of element, article or system.
The serial number of the above embodiments of the invention is only for description, does not represent the advantages or disadvantages of the embodiments.
Through the above description of the embodiments, those skilled in the art can be understood that above-described embodiment side
Method can be realized by means of software and necessary general hardware platform, naturally it is also possible to by hardware, but in many cases
The former is more preferably embodiment.Based on this understanding, technical solution of the present invention substantially in other words does the prior art
The part contributed out can be embodied in the form of software products, which is stored in one as described above
In storage medium (such as ROM/RAM, magnetic disk, CD), including some instructions are used so that terminal device (it can be mobile phone,
Computer, server, air conditioner or network equipment etc.) execute method described in each embodiment of the present invention.
The above is only a preferred embodiment of the present invention, is not intended to limit the scope of the invention, all to utilize this hair
Equivalent structure or equivalent flow shift made by bright specification and accompanying drawing content is applied directly or indirectly in other relevant skills
Art field, is included within the scope of the present invention.
Claims (10)
1. a kind of advertisement video optimization method, which is characterized in that the advertisement video optimization method the following steps are included:
Advertisement video to be optimized is divided into several advertisement video segments, and by preset video frame extraction algorithm to each
Advertisement video segment carries out the extraction of key video sequence frame, obtains several candidate key video frames；
Key indices assessment algorithm is obtained, and according to the key indices assessment algorithm, assesses several candidate key videos
Each candidate key video frame in frame, obtains the target critical index of each candidate key video frame；
According to the target critical index of each candidate key video frame, mesh is selected from several candidate key video frames
Mark key video sequence frame；
The preposition strategy of key video sequence frame is obtained, and is based on the preposition strategy of the key video sequence frame and the target critical video frame,
The advertisement video is optimized, targeted advertisements video is obtained.
2. advertisement video optimization method as described in claim 1, which is characterized in that according to the key indices assessment algorithm,
Each candidate key video frame in several candidate key video frames is assessed, the target of each candidate key video frame is obtained
The step of key indices includes:
Obtain the text information that each candidate key video frame includes in several candidate key video frames；
According to the key indices assessment algorithm and the text information, determine that the target critical of each candidate key video frame refers to
Number.
3. advertisement video optimization method as claimed in claim 2, which is characterized in that according to the key indices assessment algorithm and
The text information, the step of determining the target critical index of each candidate key video frame include:
The entropy for the text information that each candidate key video frame includes is calculated, and calculates each candidate key video frame
The similarity of the text information and default text information that include；
According to the key indices assessment algorithm, the entropy of the text information and the similarity, each candidate key view is determined
The target critical index of frequency frame.
4. advertisement video optimization method as claimed in claim 2, which is characterized in that obtain several candidate key video frames
In each candidate key video frame include text information the step of after, further includes:
Calculate the entropy for the text information that each candidate key video frame includes；
Based on default aesthetic score prediction model, each candidate key video frame in several candidate key video frames is predicted
Aesthetic score；
According to the key indices assessment algorithm, the entropy and the aesthetic score of the text information, each candidate is determined
The target critical index of key video sequence frame.
5. advertisement video optimization method as claimed in claim 4, which is characterized in that described to predict mould based on default aesthetic score
Type, after the step of predicting the aesthetic score of each candidate key video frame in several candidate key video frames, further includes:
Calculate the similarity of text information and default text information that each candidate key video frame includes；
According to the key indices assessment algorithm, the entropy of the text information, the similarity and the aesthetic score, institute is determined
State the target critical index of each candidate key video frame.
6. advertisement video optimization method according to any one of claims 1 to 5, which is characterized in that it is described will be to be optimized wide
Video segmentation is accused at before the step of several advertisement video segments, further includes:
When monitoring algorithm policy more new command, according to the algorithm policy more new command, corresponding advertising feedback number is obtained
According to, key indices assessment algorithm to be updated and the preposition strategy of key video sequence frame；
According to the advertising feedback data, more to key indices assessment algorithm to be updated and the preposition strategy execution of key video sequence frame
New operation.
7. advertisement video optimization method as claimed in claim 6, which is characterized in that according to the advertising feedback data, treat
The preposition strategy execution of the key indices assessment algorithm and key video sequence frame of update updates the step of operating
Preset algorithm policy more new model is obtained, and the advertising feedback data are input to the algorithm policy and update mould
Type；
Obtain the output of the algorithm policy more new model for update the first parameter value of the key indices assessment algorithm with
And the second parameter value for updating the preposition strategy of the key video sequence frame；
The key indices assessment algorithm is updated according to first parameter value, and according to described in second parameter value update
The preposition strategy of key video sequence frame.
8. a kind of advertisement video optimizes device, which is characterized in that the advertisement video optimizes device and includes:
Fragment segmentation module, for advertisement video to be optimized to be divided into several advertisement video segments；
Video frame extraction module, for carrying out key video sequence to each advertisement video segment by preset video frame extraction algorithm
Frame extracts, and obtains several candidate key video frames；
Key indices evaluation module, for obtaining key indices assessment algorithm, and according to the key indices assessment algorithm, assessment
Each candidate key video frame in several candidate key video frames, obtains the target critical of each candidate key video frame
Index；
Selecting module, for the target critical index according to each candidate key video frame, from several candidate keys
Selection target key video sequence frame in video frame；
Advertisement video optimization module for obtaining the preposition strategy of key video sequence frame, and is based on the preposition strategy of key video sequence frame
With the target critical video frame, the advertisement video is optimized, obtains targeted advertisements video.
9. a kind of advertisement video optimizes equipment, which is characterized in that the advertisement video optimization equipment includes: memory, processor
And it is stored in the advertisement video optimization program that can be run on the memory and on the processor, the advertisement video optimization
The step of advertisement video optimization method as described in any one of claims 1 to 7 is realized when program is executed by the processor.
10. a kind of computer readable storage medium, which is characterized in that be stored with advertisement view on the computer readable storage medium
Frequency optimization program, the advertisement video optimization program are realized as described in any one of claims 1 to 7 when being executed by processor
The step of advertisement video optimization method.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910140441.4A CN109922379B (en) | 2019-02-22 | 2019-02-22 | Advertisement video optimization method, device and equipment and computer readable storage medium |
| PCT/CN2019/078528 WO2020168606A1 (en) | 2019-02-22 | 2019-03-18 | Advertisement video optimising method, apparatus and device and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910140441.4A CN109922379B (en) | 2019-02-22 | 2019-02-22 | Advertisement video optimization method, device and equipment and computer readable storage medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109922379A true CN109922379A (en) | 2019-06-21 |
| CN109922379B CN109922379B (en) | 2022-01-04 |
Family
ID=66962344
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910140441.4A Active CN109922379B (en) | 2019-02-22 | 2019-02-22 | Advertisement video optimization method, device and equipment and computer readable storage medium |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN109922379B (en) |
| WO (1) | WO2020168606A1 (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111104370A (en) * | 2019-12-18 | 2020-05-05 | 北京大龙得天力广告传媒有限公司 | Advertisement video storage system and method |
| CN111294646A (en) * | 2020-02-17 | 2020-06-16 | 腾讯科技（深圳）有限公司 | Video processing method, device, equipment and storage medium |
| CN113378002A (en) * | 2021-08-11 | 2021-09-10 | 北京达佳互联信息技术有限公司 | Information delivery method and device, electronic equipment and storage medium |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113610557A (en) * | 2021-07-09 | 2021-11-05 | 苏州众言网络科技股份有限公司 | Advertisement video processing method and device |
| CN114708008A (en) * | 2021-12-30 | 2022-07-05 | 北京有竹居网络技术有限公司 | Promotion content processing method, device, equipment, medium and product |
| CN114697762B (en) * | 2022-04-07 | 2023-11-28 | 脸萌有限公司 | Processing method, processing device, terminal equipment and medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070183497A1 (en) * | 2006-02-03 | 2007-08-09 | Jiebo Luo | Extracting key frame candidates from video clip |
| CN104581380A (en) * | 2014-12-30 | 2015-04-29 | 联想(北京)有限公司 | Information processing method and mobile terminal |
| CN108632641A (en) * | 2018-05-04 | 2018-10-09 | 百度在线网络技术（北京）有限公司 | Method for processing video frequency and device |
| CN109121021A (en) * | 2018-09-28 | 2019-01-01 | 北京周同科技有限公司 | A kind of generation method of Video Roundup, device, electronic equipment and storage medium |
| CN109194978A (en) * | 2018-10-15 | 2019-01-11 | 广州虎牙信息科技有限公司 | Live video clipping method, device and electronic equipment |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090094113A1 (en) * | 2007-09-07 | 2009-04-09 | Digitalsmiths Corporation | Systems and Methods For Using Video Metadata to Associate Advertisements Therewith |
| CN107172481A (en) * | 2017-05-09 | 2017-09-15 | 深圳市炜光科技有限公司 | Video segment splices method of combination and system |
| CN107392666A (en) * | 2017-07-24 | 2017-11-24 | 北京奇艺世纪科技有限公司 | Advertisement data processing method, device and advertisement placement method and device |
-
2019
- 2019-02-22 CN CN201910140441.4A patent/CN109922379B/en active Active
- 2019-03-18 WO PCT/CN2019/078528 patent/WO2020168606A1/en active Application Filing
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20070183497A1 (en) * | 2006-02-03 | 2007-08-09 | Jiebo Luo | Extracting key frame candidates from video clip |
| CN104581380A (en) * | 2014-12-30 | 2015-04-29 | 联想(北京)有限公司 | Information processing method and mobile terminal |
| CN108632641A (en) * | 2018-05-04 | 2018-10-09 | 百度在线网络技术（北京）有限公司 | Method for processing video frequency and device |
| CN109121021A (en) * | 2018-09-28 | 2019-01-01 | 北京周同科技有限公司 | A kind of generation method of Video Roundup, device, electronic equipment and storage medium |
| CN109194978A (en) * | 2018-10-15 | 2019-01-11 | 广州虎牙信息科技有限公司 | Live video clipping method, device and electronic equipment |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111104370A (en) * | 2019-12-18 | 2020-05-05 | 北京大龙得天力广告传媒有限公司 | Advertisement video storage system and method |
| CN111294646A (en) * | 2020-02-17 | 2020-06-16 | 腾讯科技（深圳）有限公司 | Video processing method, device, equipment and storage medium |
| CN111294646B (en) * | 2020-02-17 | 2022-08-30 | 腾讯科技（深圳）有限公司 | Video processing method, device, equipment and storage medium |
| CN113378002A (en) * | 2021-08-11 | 2021-09-10 | 北京达佳互联信息技术有限公司 | Information delivery method and device, electronic equipment and storage medium |
| CN113378002B (en) * | 2021-08-11 | 2022-01-21 | 北京达佳互联信息技术有限公司 | Information delivery method and device, electronic equipment and storage medium |
| US11632586B2 (en) | 2021-08-11 | 2023-04-18 | Beijing Dajia Internet Information Technology Co., Ltd. | Method for placing delivery information, electronic device, and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109922379B (en) | 2022-01-04 |
| WO2020168606A1 (en) | 2020-08-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109922379A (en) | Advertisement video optimization method, device, equipment and computer readable storage medium | |
| US11290775B2 (en) | Computerized system and method for automatically detecting and rendering highlights from streaming videos | |
| US11651002B2 (en) | Method for providing intelligent service, intelligent service system and intelligent terminal based on artificial intelligence | |
| US11032512B2 (en) | Server and operating method thereof | |
| US9934515B1 (en) | Content recommendation system using a neural network language model | |
| US11367150B2 (en) | Demographic-based targeting of electronic media content items | |
| CN106462608A (en) | Knowledge source personalization to improve language models | |
| CN111582116B (en) | Video erasing trace detection method, device, equipment and storage medium | |
| JP7208595B2 (en) | Movie Success Index Prediction | |
| CN102640168A (en) | Method and apparatus for local binary pattern based facial feature localization | |
| CN110135497B (en) | Model training method, and method and device for estimating strength of facial action unit | |
| CN105095625B (en) | Clicking rate prediction model method for building up, device and information providing method, system | |
| CN108766585A (en) | Generation method, device and the computer readable storage medium of influenza prediction model | |
| CN111813532A (en) | Image management method and device based on multitask machine learning model | |
| CN110415009B (en) | Computerized system and method for intra-video modification | |
| EP3699840B1 (en) | Method for analyzing supply chain risk of suppliers | |
| CN114281976A (en) | Model training method and device, electronic equipment and storage medium | |
| CN103324641A (en) | Information record recommendation method and device | |
| CN115170390B (en) | File stylization method, device, equipment and storage medium | |
| CN111988407B (en) | Content pushing method and related device | |
| CN115345635A (en) | Processing method and device for recommended content, computer equipment and storage medium | |
| CN112905892A (en) | Big data processing method and big data server applied to user portrait mining | |
| CN112559877A (en) | CTR (China railway) estimation method and system based on cross-platform heterogeneous data and behavior context | |
| CN115809372B (en) | Click rate prediction model training method and device based on decoupling invariant learning | |
| US20240066355A1 (en) | Live Tournament Predictions in Tennis |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |