CN110381368A - Video cover generation method, device and electronic equipment - Google Patents
Video cover generation method, device and electronic equipment Download PDFInfo
- Publication number
- CN110381368A CN110381368A CN201910622565.6A CN201910622565A CN110381368A CN 110381368 A CN110381368 A CN 110381368A CN 201910622565 A CN201910622565 A CN 201910622565A CN 110381368 A CN110381368 A CN 110381368A
- Authority
- CN
- China
- Prior art keywords
- image
- candidate
- video
- cover
- target
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- 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, rendering scenes according to MPEG-4 scene graphs
- H04N21/44008—Processing of video elementary streams, e.g. splicing a video clip retrieved from local storage with an incoming video stream, rendering scenes according to MPEG-4 scene graphs involving operations for analysing video streams, e.g. detecting features or characteristics in the video stream
-
- 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
- H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
- H04N21/454—Content or additional data filtering, e.g. blocking advertisements
- H04N21/4545—Input to filtering algorithms, e.g. filtering a region of the image
-
- 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
- H04N21/45—Management operations performed by the client for facilitating the reception of or the interaction with the content or administrating data related to the end-user or to the client device itself, e.g. learning user preferences for recommending movies, resolving scheduling conflicts
- H04N21/466—Learning process for intelligent management, e.g. learning user preferences for recommending movies
- H04N21/4662—Learning process for intelligent management, e.g. learning user preferences for recommending movies characterized by learning algorithms
- H04N21/4665—Learning process for intelligent management, e.g. learning user preferences for recommending movies characterized by learning algorithms involving classification methods, e.g. Decision trees
-
- 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
- H04N21/475—End-user interface for inputting end-user data, e.g. personal identification number [PIN], preference data
- H04N21/4756—End-user interface for inputting end-user data, e.g. personal identification number [PIN], preference data for rating content, e.g. scoring a recommended movie
Abstract
A kind of video cover generation method, device and electronic equipment are provided in the embodiment of the present disclosure, are belonged to technical field of image processing, this method comprises: parsing to target video, are obtained multiple parsing images；Clustering processing is carried out to the multiple parsing image, obtains multiple candidate cover images；Based on the critical point detection to the target object execution for including in the candidate cover image as a result, line-of-sight detection is executed and detection of opening eyes to the target object for including in the multiple candidate cover image, to filter out the candidate cover image for not meeting preset requirement；Based on image quality evaluation network to the quality score of candidate's cover image remaining after filtering, final cover image of the target image as the target video is chosen from candidate's cover image remaining after filtering.By the processing scheme of the disclosure, the video cover of high quality can be automatically generated.
Description
Technical field
This disclosure relates to which technical field of image processing more particularly to a kind of video cover generation method, device and electronics are set
It is standby.
Background technique
Image procossing (image processing) is also known as image processing, is needed for being reached with computer to image
As a result technology.Originating from the 1920s, generally Digital Image Processing.The main contents of image processing techniques include figure
As compression, enhancing restore, matching description identification 3 parts, common processing have image digitazation, image coding, image enhancement,
Image restoration, image segmentation and image analysis etc..Image procossing is to be processed image information to meet people using computer
Visual psychology or application demand behavior, be widely used, be chiefly used in mapping science, atmospheric science, astronomy, U.S. figure, make figure
As improving identification etc..
One application scenarios of image procossing are the views that a video frame is selected in one section of video as this section of video
How frequency cover selects the video frame that one representative, picture quality is high as video from numerous video frames
Cover becomes the technical issues that need to address.
Summary of the invention
In view of this, the embodiment of the present disclosure provides a kind of video cover generation method, device and electronic equipment, at least partly
Solve problems of the prior art.
In a first aspect, the embodiment of the present disclosure provides a kind of video cover generation method, comprising:
Target video is parsed, multiple parsing images are obtained；
Clustering processing is carried out to the multiple parsing image, obtains multiple candidate cover images；
Based on the critical point detection executed to the target object for including in the candidate cover image as a result, to the multiple
The target object for including in candidate cover image executes line-of-sight detection and detection of opening eyes, to filter out the time for not meeting preset requirement
Select cover image；
It is remaining after filtering based on image quality evaluation network to the quality score of candidate's cover image remaining after filtering
Candidate cover image in choose final cover image of the target image as the target video.
It is described that target video is parsed according to a kind of specific implementation of the embodiment of the present disclosure, obtain multiple solutions
Analyse image, comprising:
All video frame performance objectives detection to including in the target video；
It is based on target detection as a result, whether judging to form in the video frame of target video comprising the target object；
If so, setting the parsing image for the video frame comprising the target object.
It is described that the multiple parsing image is carried out at cluster according to a kind of specific implementation of the embodiment of the present disclosure
Reason obtains multiple candidate cover images, comprising:
The cluster calculation of k class is executed to the multiple parsing image；
Select the image for meeting preset condition as the candidate cover image in each cluster.
It is described that the multiple parsing image execution k class is gathered according to a kind of specific implementation of the embodiment of the present disclosure
Class calculates, comprising:
It is initial cluster center that k sample point is chosen in the multiple parsing image, is denoted as z1 (l), z2 (l) ... ...
Zk (l), iteration serial number l=1；
All samples are assigned in k class ω j (k) representated by each cluster centre using Nearest Neighbor Method, it is all kinds of to be wrapped
The sample number contained is Nj (l)；
All kinds of centers of gravity is calculated, the center of gravity being calculated is determined as to new cluster centre；
For iteration j, judge whether the value of zj (l+1) and zj (l) is identical, continues to change as zj (l+1) ≠ zj (l)
In generation, calculates, and as zj (l+1)=zj (l), stops iterative calculation.
It is described to including in the multiple candidate cover image according to a kind of specific implementation of the embodiment of the present disclosure
Before target object executes line-of-sight detection and detection of opening eyes, the method also includes:
Quality evaluation is executed to the multiple candidate cover image using pre-set convolutional neural networks；
Based on the quality evaluation as a result, determining that the quality of each image in the multiple candidate cover image is commented
Point.
It is described to including in the multiple candidate cover image according to a kind of specific implementation of the embodiment of the present disclosure
Target object executes line-of-sight detection and detection of opening eyes, comprising:
Obtain the critical point detection result executed to the target object for including in candidate cover image；
Based on the critical point detection as a result, determining pupil key point, eyeball key point and eye profile key point；
Based on the pupil key point, eyeball key point and eye profile key point, the eyeball of the target object is executed
Modelling operability；
Based on modelling operability as a result, being detected the target object is executed line-of-sight detection and be opened eyes.
According to a kind of specific implementation of the embodiment of the present disclosure, it is described based on modelling operability as a result, come to the mesh
It marks object and executes line-of-sight detection and detection of opening eyes, comprising:
Go out the key point of eyeball using CLNF model inspection；
Three-dimensional modeling is carried out to eyes；
It connects origin and forms a ray to pupil center, its intersection point with eyeball is calculated, by eyeball center to intersection point side
To vector as direction of visual lines.
According to a kind of specific implementation of the embodiment of the present disclosure, it is described based on modelling operability as a result, come to the mesh
It marks object and executes line-of-sight detection and detection of opening eyes, comprising:
By modelling operability as a result, calculating the ratio of width to height of external eyes profile；
Judge whether the target object is in eyes-open state by judging whether described the ratio of width to height is greater than preset threshold.
It is described to be remained based on image quality evaluation network to after filtering according to a kind of specific implementation of the embodiment of the present disclosure
The quality score of remaining candidate cover image chooses target image as the mesh from candidate's cover image remaining after filtering
Mark the final cover image of video, comprising:
The highest candidate cover image of quality score is chosen from candidate's cover image remaining after filtering；
Using the highest candidate cover image of the quality score as the final cover image of the target video.
Second aspect, the embodiment of the present disclosure provide a kind of video cover generating means, comprising:
Parsing module obtains multiple parsing images for parsing to target video；
Cluster module obtains multiple candidate cover images for carrying out clustering processing to the multiple parsing image；
Filtering module, for based on the critical point detection knot executed to the target object for including in the candidate cover image
Fruit executes line-of-sight detection and opens eyes and detect, is not inconsistent with filtering out to the target object for including in the multiple candidate cover image
Close the candidate cover image of preset requirement；
Selecting module, for being commented based on quality of the image quality evaluation network to candidate's cover image remaining after filtering
Point, final cover image of the target image as the target video is chosen from candidate's cover image remaining after filtering.
The third aspect, the embodiment of the present disclosure additionally provide a kind of electronic equipment, which includes:
At least one processor；And
The memory being connect at least one processor communication；Wherein,
The memory is stored with the instruction that can be executed by least one processor, and the instruction is by least one processor
It executes, so that at least one processor is able to carry out the view in any implementation of aforementioned first aspect or first aspect
Frequency cover generation method.
Fourth aspect, the embodiment of the present disclosure additionally provide a kind of non-transient computer readable storage medium, the non-transient meter
Calculation machine readable storage medium storing program for executing stores computer instruction, and the computer instruction is for making the computer execute aforementioned first aspect or the
Video cover generation method in any implementation of one side.
5th aspect, the embodiment of the present disclosure additionally provide a kind of computer program product, which includes
The calculation procedure being stored in non-transient computer readable storage medium, the computer program include program instruction, when the program
When instruction is computer-executed, the computer is made to execute the video in aforementioned first aspect or any implementation of first aspect
Cover generation method.
Video cover in the embodiment of the present disclosure generates scheme, including parses to target video, obtains multiple parsings
Image；Clustering processing is carried out to the multiple parsing image, obtains multiple candidate cover images；Based on to the candidate surface plot
The critical point detection that the target object for including as in executes is as a result, to the target object for including in the multiple candidate cover image
Line-of-sight detection is executed and detection of opening eyes, to filter out the candidate cover image for not meeting preset requirement；Based on image quality evaluation
Network chooses target from candidate's cover image remaining after filtering to the quality score of candidate's cover image remaining after filtering
Final cover image of the image as the target video.By the scheme of the disclosure, the allusion quotation of high quality can be automatically selected
Cover image of the type video frame as video.
Detailed description of the invention
It, below will be to needed in the embodiment attached in order to illustrate more clearly of the technical solution of the embodiment of the present disclosure
Figure is briefly described, it should be apparent that, the accompanying drawings in the following description is only some embodiments of the present disclosure, for this field
For those of ordinary skill, without creative efforts, it can also be obtained according to these attached drawings other attached drawings.
Fig. 1 is a kind of video cover product process schematic diagram that the embodiment of the present disclosure provides；
Fig. 2 is another video cover product process schematic diagram that the embodiment of the present disclosure provides；
Fig. 3 is another video cover product process schematic diagram that the embodiment of the present disclosure provides；
Fig. 4 is another video cover product process schematic diagram that the embodiment of the present disclosure provides；
Fig. 5 is a kind of video cover generating means structural schematic diagram that the embodiment of the present disclosure provides；
Fig. 6 is the electronic equipment schematic diagram that the embodiment of the present disclosure provides.
Specific embodiment
The embodiment of the present disclosure is described in detail with reference to the accompanying drawing.
Illustrate embodiment of the present disclosure below by way of specific specific example, those skilled in the art can be by this specification
Disclosed content understands other advantages and effect of the disclosure easily.Obviously, described embodiment is only the disclosure
A part of the embodiment, instead of all the embodiments.The disclosure can also be subject to reality by way of a different and different embodiment
It applies or applies, the various details in this specification can also be based on different viewpoints and application, in the spirit without departing from the disclosure
Lower carry out various modifications or alterations.It should be noted that in the absence of conflict, the feature in following embodiment and embodiment can
To be combined with each other.Based on the embodiment in the disclosure, those of ordinary skill in the art are without creative efforts
Every other embodiment obtained belongs to the range of disclosure protection.
It should be noted that the various aspects of embodiment within the scope of the appended claims are described below.Ying Xian
And be clear to, aspect described herein can be embodied in extensive diversified forms, and any specific structure described herein
And/or function is only illustrative.Based on the disclosure, it will be understood by one of ordinary skill in the art that one described herein
Aspect can be independently implemented with any other aspect, and can combine the two or both in these aspects or more in various ways.
For example, carry out facilities and equipments in terms of any number set forth herein can be used and/or practice method.In addition, can make
With other than one or more of aspect set forth herein other structures and/or it is functional implement this equipment and/or
Practice the method.
It should also be noted that, diagram provided in following embodiment only illustrates the basic structure of the disclosure in a schematic way
Think, component count, shape and the size when only display is with component related in the disclosure rather than according to actual implementation in schema are drawn
System, when actual implementation kenel, quantity and the ratio of each component can arbitrarily change for one kind, and its assembly layout kenel can also
It can be increasingly complex.
In addition, in the following description, specific details are provided for a thorough understanding of the examples.However, fields
The skilled person will understand that the aspect can be practiced without these specific details.
The embodiment of the present disclosure provides a kind of video cover generation method.Video cover generation method provided in this embodiment can
To be executed by a computing device, which can be implemented as software, or be embodied as the combination of software and hardware, the meter
It calculates device and can integrate and be arranged in server, terminal device etc..
Referring to Fig. 1, a kind of video cover generation method of embodiment of the present disclosure offer, comprising:
S101 parses target video, obtains multiple parsing images.
Target video is one section of video file for recording audio and image data, and target video can be arbitrary format
Video file is also possible to other lattice for example, target video can be the video file of mpg, mp4, rm, rmvb, wax format
The video file of formula.
It include video frame in target video, video frame is the set of all video pictures in target video, for example, for
For the video that one frame per second is 30fps, according to normal video playout speed, it can be split out in the video of 1 second length
30 video frames.Certainly actual needs are based on, can also be obtained more by way of carrying out interleave in 30 video frames
Video frame, alternatively, the selected section video frame from 30 video frames.
It, can be to the view in target video in order to improve the efficiency of parsing during being parsed to target video
Frequency frame is screened, and will meet the video frame of screening conditions as parsing image.As an example, the target can be regarded
All video frame performance objectives detection for including in frequency, passes through target detection, it can be determined that whether video frame includes target object
(for example, people) then sets the parsing for the video frame comprising the target object when in video frame including target object
Image.In this way, the specific aim of multiple parsing images can further be improved.Carrying out target detection to video frame can
To use a variety of object detection methods for image, mode of target detection is not construed as limiting at this.
S102 carries out clustering processing to the multiple parsing image, obtains multiple candidate cover images.
Contain target object in parsing image, in order to be sieved to the parsing image comprising target object is further
Choosing, can execute clustering processing to multiple parsing images can be selected by clustering to image in multiple parsing images
A part of typical image (multiple candidate's cover images) is further to be handled.
Specifically, k class can be arranged to multiple parsing images in advance, to carry out cluster calculation.Firstly, the multiple
Parsing and choosing k sample point in image is initial cluster center, is denoted as z1 (l), z2 (l) ... ... zk (l), iteration serial number l=1；
Next, all samples are assigned in k class ω j (k) representated by each cluster centre using Nearest Neighbor Method, it is all kinds of to be included
Sample number be Nj (l)；Next, the center of gravity being calculated is determined as new gather by way of calculating all kinds of centers of gravity
Class center；Finally, judge whether the value of zj (l+1) and zj (l) is identical for iteration j by way of loop iteration, when
Continue to iterate to calculate when zj (l+1) ≠ zj (l), as zj (l+1)=zj (l), stops iterative calculation.After iteration stopping,
Can by it is finally clustering as a result, in each cluster select a parsing image, ultimately form k candidate cover image.
S103, based on the critical point detection executed to the target object for including in the candidate cover image as a result, to institute
It states the target object for including in multiple candidate cover images and executes line-of-sight detection and detection of opening eyes, do not meet default want to filter out
The candidate cover image asked.
After getting multiple candidate cover images, key point inspection can be executed to the target object in candidate cover image
It surveys, by critical point detection, the key point on available target image head, the key point based on head can be to target pair
As the sight in candidate cover image and whether opening eyes is detected, to target object sight be deviated or eyes are not complete
The image filtering opened entirely falls.
Specifically, the available critical point detection executed to the target object for including in candidate cover image is as a result, base
In these critical point detections as a result, determining pupil key point, eyeball key point and eye profile key point；It is closed based on the pupil
Key point, eyeball key point and eye profile key point execute modelling operability to the eyeball of the target object, pass through modelling operability
It is detected as a result, it is possible to which the target object is executed line-of-sight detection and opened eyes.
As a kind of mode, the key point of eyeball can be gone out using CLNF model inspection by the result of modelling operability, together
When three-dimensional modeling is carried out to eyes, and after completing three-dimensional modeling to eyes, connection origin forms one to pupil center and penetrates
Line calculates its intersection point with eyeball, using the vector at eyeball center to intersection point direction as direction of visual lines.To further judge
Whether the sight of target object meets the requirements.
Alternatively mode, can be by modelling operability as a result, the ratio of width to height of calculating external eyes profile, passes through judgement
Whether described the ratio of width to height is greater than preset threshold to judge whether the target object is in eyes-open state.
It, can will be undesirable when finding that candidate cover image is undesirable by line-of-sight detection and detection of opening eyes
Candidate cover image filter out, thus remaining candidate's cover image set after being filtered.
S104, based on image quality evaluation network to the quality score of candidate's cover image remaining after filtering, from filtering
Final cover image of the target image as the target video is chosen in remaining candidate cover image afterwards.
After being filtered after remaining candidate's cover image set, a figure can be further screened from this collection
Picture, so that it is determined that the final cover image of target video.
As a kind of mode, the target object for including in the multiple candidate cover image can be executed line-of-sight detection and
It opens eyes before detection, quality score, matter is carried out to all candidate cover images by using convolutional neural networks trained in advance
During amount scoring, overall merit can be carried out from many aspects such as image quality, color, environment and expressions, thus to every
A candidate's cover image provides a specific quality score.
It can pass through after the quality score of each image in remaining candidate's cover image set after being filtered
The mode of sequence, final surface plot of the candidate cover image of the one or more for selecting quality score high as the target video
Picture.
By above step, screening can be filtered to the video frame in target video in several ways, thus
Screen the final cover image to match for target video.
Target video is parsed referring to fig. 2 according to a kind of optional implementation of the embodiment of the present disclosure, is obtained more
A parsing image, may include steps of:
S201 detects all video frame performance objectives for including in the target video.
All video frame performance objectives for including in the target video can be detected, by target detection, can be sentenced
Whether disconnected video frame includes target object (for example, people), and carrying out target detection to video frame can be using a variety of for image
Object detection method is not construed as limiting the mode of target detection at this.
S202, it is based on target detection as a result, whether judging to form in the video frame of target video comprising the target pair
As.
By detecting to video frame performance objective, multiple object detection results can be obtained, by by multiple object detections
As a result similarity is carried out with target object to compare, can further judge whether contain target pair in the video frame of target video
As.
S203, if so, setting the parsing image for the video frame comprising the target object.
By the step in step S201-203, targetedly video frame images can be selected, obtained multiple
Parse image.
Alternatively situation is carrying out clustering processing to the multiple parsing image, obtains multiple candidate surface plots
As during, the cluster calculation of k class can be executed to the multiple parsing image, selection one meets pre- in each cluster
If the image of condition is as the candidate cover image.
According to a kind of optional implementation of the embodiment of the present disclosure, referring to Fig. 3, k class is executed to the multiple parsing image
Cluster calculation, may include steps of:
S301, choosing k sample point in the multiple parsing image is initial cluster center, is denoted as z1 (l), z2
(l) ... ... zk (l), iteration serial number l=1；
All samples are assigned in k class ω j (k) representated by each cluster centre by S302 using Nearest Neighbor Method, all kinds of
The sample number for being included is Nj (l)；
S303 calculates all kinds of centers of gravity, and the center of gravity being calculated is determined as to new cluster centre；
S304 judges whether the value of zj (l+1) and zj (l) is identical, as zj (l+1) ≠ zj (l) for iteration j
Continue to iterate to calculate, as zj (l+1)=zj (l), stops iterative calculation.
It is described to including in the multiple candidate cover image according to a kind of optional implementation of the embodiment of the present disclosure
Before target object executes line-of-sight detection and detection of opening eyes, the method also includes: utilize pre-set convolutional neural networks
Quality evaluation is executed to the multiple candidate cover image；Based on the quality evaluation as a result, determining the multiple candidate envelope
The quality score of each image in the image of face.
Referring to fig. 4, according to a kind of optional implementation of the embodiment of the present disclosure, to being wrapped in the multiple candidate cover image
The target object contained executes line-of-sight detection and detection of opening eyes, comprising:
S401 obtains the critical point detection result executed to the target object for including in candidate cover image.
Specifically, critical point detection can be executed for the head zone of target object in cover image, pass through key point
Detection, can obtain the crucial point data of the multiple organs of head zone (for example, eyes).
S402, based on the critical point detection as a result, determining that pupil key point, eyeball key point and eye profile are crucial
Point.
By carrying out target identification to the key point that detects, be capable of determining that out pupil key point, eyeball key point and
Eye profile key point.
S403 is based on the pupil key point, eyeball key point and eye profile key point, to the eyeball of the target object
Execute modelling operability.
It, can be based on these crucial point datas to target after pupil key point, eyeball key point and eye profile key point
The eyeball of object executes modelling operability, and after modeling, more careful quantization can be carried out to the eyeball of target object.For mesh
The modelling operability for marking object eyeball can be carried out using various ways, be not limited thereto.
S404, based on modelling operability as a result, being detected the target object is executed line-of-sight detection and be opened eyes.
During realizing step S404, eyeball can be gone out using CLNF model inspection by the result of modelling operability
Key point carries out three-dimensional modeling to eyes by the key point of eyeball, to the model after three-dimensional modeling, connects origin to pupil
It is centrally formed a ray, calculates its intersection point with eyeball, using the vector at eyeball center to intersection point direction as direction of visual lines.From
And whether sight requirement is met based on the sight of the walking direction target object of determining sight.
It, can also be by modelling operability as a result, the ratio of width to height of calculating external eyes profile, leads to other than carrying out line-of-sight detection
It crosses and judges whether described the ratio of width to height is greater than preset threshold to judge whether the target object is in eyes-open state.To delete
The candidate cover image of eyes-open state is not met.
It is described to be remained based on image quality evaluation network to after filtering according to a kind of optional implementation of the embodiment of the present disclosure
The quality score of remaining candidate cover image chooses target image as the mesh from candidate's cover image remaining after filtering
Mark the final cover image of video, comprising: choose the highest candidate of quality score from candidate's cover image remaining after filtering
Cover image；Using the highest candidate cover image of the quality score as the final cover image of the target video.
Corresponding with above method embodiment, referring to Fig. 5, the embodiment of the present disclosure additionally provides a kind of video cover generation
Device 50, comprising:
Parsing module 501 obtains multiple parsing images for parsing to target video.
Target video is one section of video file for recording audio and image data, and target video can be arbitrary format
Video file is also possible to other lattice for example, target video can be the video file of mpg, mp4, rm, rmvb, wax format
The video file of formula.
It include video frame in target video, video frame is the set of all video pictures in target video, for example, for
For the video that one frame per second is 30fps, according to normal video playout speed, it can be split out in the video of 1 second length
30 video frames.Certainly actual needs are based on, can also be obtained more by way of carrying out interleave in 30 video frames
Video frame, alternatively, the selected section video frame from 30 video frames.
It, can be to the view in target video in order to improve the efficiency of parsing during being parsed to target video
Frequency frame is screened, and will meet the video frame of screening conditions as parsing image.As an example, the target can be regarded
All video frame performance objectives detection for including in frequency, passes through target detection, it can be determined that whether video frame includes target object
(for example, people) then sets the parsing for the video frame comprising the target object when in video frame including target object
Image.In this way, the specific aim of multiple parsing images can further be improved.Carrying out target detection to video frame can
To use a variety of object detection methods for image, mode of target detection is not construed as limiting at this.
Cluster module 502 obtains multiple candidate cover images for carrying out clustering processing to the multiple parsing image.
Contain target object in parsing image, in order to be sieved to the parsing image comprising target object is further
Choosing, can execute clustering processing to multiple parsing images can be selected by clustering to image in multiple parsing images
A part of typical image (multiple candidate's cover images) is further to be handled.
Specifically, k class can be arranged to multiple parsing images in advance, to carry out cluster calculation.Firstly, the multiple
Parsing and choosing k sample point in image is initial cluster center, is denoted as z1 (l), z2 (l) ... ... zk (l), iteration serial number l=1；
Next, all samples are assigned in k class ω j (k) representated by each cluster centre using Nearest Neighbor Method, it is all kinds of to be included
Sample number be Nj (l)；Next, the center of gravity being calculated is determined as new gather by way of calculating all kinds of centers of gravity
Class center；Finally, judge whether the value of zj (l+1) and zj (l) is identical for iteration j by way of loop iteration, when
Continue to iterate to calculate when zj (l+1) ≠ zj (l), as zj (l+1)=zj (l), stops iterative calculation.After iteration stopping,
Can by it is finally clustering as a result, in each cluster select a parsing image, ultimately form k candidate cover image.
Filtering module 503, for being examined based on the key point executed to the target object for including in the candidate cover image
It surveys and is detected as a result, executing line-of-sight detection and eye opening to the target object for including in the multiple candidate cover image, to filter out
The candidate cover image of preset requirement is not met.
After getting multiple candidate cover images, key point inspection can be executed to the target object in candidate cover image
It surveys, by critical point detection, the key point on available target image head, the key point based on head can be to target pair
As the sight in candidate cover image and whether opening eyes is detected, to target object sight be deviated or eyes are not complete
The image filtering opened entirely falls.
Specifically, the available critical point detection executed to the target object for including in candidate cover image is as a result, base
In these critical point detections as a result, determining pupil key point, eyeball key point and eye profile key point；It is closed based on the pupil
Key point, eyeball key point and eye profile key point execute modelling operability to the eyeball of the target object, pass through modelling operability
It is detected as a result, it is possible to which the target object is executed line-of-sight detection and opened eyes.
Eye can be gone out using CLNF model inspection by the result of modelling operability as a kind of mode as a kind of mode
The key point of ball, while three-dimensional modeling is carried out to eyes, and after completing three-dimensional modeling to eyes, connect origin into pupil
It is formed centrally a ray, calculates its intersection point with eyeball, using the vector at eyeball center to intersection point direction as direction of visual lines.To
Further judge whether the sight of target object meets the requirements.
Alternatively mode, can be by modelling operability as a result, the ratio of width to height of calculating external eyes profile, passes through judgement
Whether described the ratio of width to height is greater than preset threshold to judge whether the target object is in eyes-open state.
It, can will be undesirable when finding that candidate cover image is undesirable by line-of-sight detection and detection of opening eyes
Candidate cover image filter out, thus remaining candidate's cover image set after being filtered.
Selecting module 504, for the quality based on image quality evaluation network to candidate's cover image remaining after filtering
Scoring chooses final cover image of the target image as the target video from candidate's cover image remaining after filtering.
After being filtered after remaining candidate's cover image set, a figure can be further screened from this collection
Picture, so that it is determined that the final cover image of target video.
As a kind of mode, the target object for including in the multiple candidate cover image can be executed line-of-sight detection and
It opens eyes before detection, quality score, matter is carried out to all candidate cover images by using convolutional neural networks trained in advance
During amount scoring, overall merit can be carried out from many aspects such as image quality, color, environment and expressions, thus to every
A candidate's cover image provides a specific quality score.
It can pass through after the quality score of each image in remaining candidate's cover image set after being filtered
The mode of sequence, final surface plot of the candidate cover image of the one or more for selecting quality score high as the target video
Picture.
Fig. 5 shown device can it is corresponding execute above method embodiment in content, what the present embodiment was not described in detail
Part, referring to the content recorded in above method embodiment, details are not described herein.
Referring to Fig. 6, the embodiment of the present disclosure additionally provides a kind of electronic equipment 60, which includes:
At least one processor；And
The memory being connect at least one processor communication；Wherein,
The memory is stored with the instruction that can be executed by least one processor, and the instruction is by least one processor
It executes, so that at least one processor is able to carry out video cover generation method in preceding method embodiment.
The embodiment of the present disclosure additionally provides a kind of non-transient computer readable storage medium, and the non-transient computer is readable to deposit
Storage media stores computer instruction, and the computer instruction is for executing the computer in preceding method embodiment.
The embodiment of the present disclosure additionally provides a kind of computer program product, and the computer program product is non-temporary including being stored in
Calculation procedure on state computer readable storage medium, the computer program include program instruction, when the program instruction is calculated
When machine executes, the computer is made to execute the video cover generation method in preceding method embodiment.
Below with reference to Fig. 6, it illustrates the structural schematic diagrams for the electronic equipment 60 for being suitable for being used to realize the embodiment of the present disclosure.
Electronic equipment in the embodiment of the present disclosure can include but is not limited to such as mobile phone, laptop, Digital Broadcasting Receiver
Device, PDA (personal digital assistant), PAD (tablet computer), PMP (portable media player), car-mounted terminal are (such as vehicle-mounted
Navigation terminal) etc. mobile terminal and such as number TV, desktop computer etc. fixed terminal.Electronics shown in Fig. 6
Equipment is only an example, should not function to the embodiment of the present disclosure and use scope bring any restrictions.
As shown in fig. 6, electronic equipment 60 may include processing unit (such as central processing unit, graphics processor etc.) 601,
It can be loaded into random access storage according to the program being stored in read-only memory (ROM) 602 or from storage device 608
Program in device (RAM) 603 and execute various movements appropriate and processing.In RAM 603, it is also stored with the behaviour of electronic equipment 60
Various programs and data needed for making.Processing unit 601, ROM 602 and RAM 603 are connected with each other by bus 604.It is defeated
Enter/export (I/O) interface 605 and is also connected to bus 604.
In general, following device can connect to I/O interface 605: including such as touch screen, touch tablet, keyboard, mouse, figure
As the input unit 606 of sensor, microphone, accelerometer, gyroscope etc.；Including such as liquid crystal display (LCD), loudspeaking
The output device 607 of device, vibrator etc.；Storage device 608 including such as tape, hard disk etc.；And communication device 609.It is logical
T unit 609 can permit electronic equipment 60 and wirelessly or non-wirelessly be communicated with other equipment to exchange data.Although showing in figure
The electronic equipment 60 with various devices is gone out, it should be understood that being not required for implementing or having all devices shown.
It can alternatively implement or have more or fewer devices.
Particularly, in accordance with an embodiment of the present disclosure, it may be implemented as computer above with reference to the process of flow chart description
Software program.For example, embodiment of the disclosure includes a kind of computer program product comprising be carried on computer-readable medium
On computer program, which includes the program code for method shown in execution flow chart.In such reality
It applies in example, which can be downloaded and installed from network by communication device 609, or from storage device 608
It is mounted, or is mounted from ROM 602.When the computer program is executed by processing unit 601, the embodiment of the present disclosure is executed
Method in the above-mentioned function that limits.
It should be noted that the above-mentioned computer-readable medium of the disclosure can be computer-readable signal media or meter
Calculation machine readable storage medium storing program for executing either the two any combination.Computer readable storage medium for example can be --- but not
Be limited to --- electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor system, device or device, or any above combination.Meter
The more specific example of calculation machine readable storage medium storing program for executing can include but is not limited to: have the electrical connection, just of one or more conducting wires
Taking formula computer disk, hard disk, random access storage device (RAM), read-only memory (ROM), erasable type may be programmed read-only storage
Device (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), light storage device, magnetic memory device,
Or above-mentioned any appropriate combination.In the disclosure, computer readable storage medium can be it is any include or storage journey
The tangible medium of sequence, the program can be commanded execution system, device or device use or in connection.And at this
In open, computer-readable signal media may include in a base band or as the data-signal that carrier wave a part is propagated,
In carry computer-readable program code.The data-signal of this propagation can take various forms, including but not limited to
Electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be computer-readable and deposit
Any computer-readable medium other than storage media, the computer-readable signal media can send, propagate or transmit and be used for
By the use of instruction execution system, device or device or program in connection.Include on computer-readable medium
Program code can transmit with any suitable medium, including but not limited to: electric wire, optical cable, RF (radio frequency) etc. are above-mentioned
Any appropriate combination.
Above-mentioned computer-readable medium can be included in above-mentioned electronic equipment；It is also possible to individualism, and not
It is fitted into the electronic equipment.
Above-mentioned computer-readable medium carries one or more program, when said one or multiple programs are by the electricity
When sub- equipment executes, so that the electronic equipment: obtaining at least two internet protocol addresses；Send to Node evaluation equipment includes institute
State the Node evaluation request of at least two internet protocol addresses, wherein the Node evaluation equipment is internet from described at least two
In protocol address, chooses internet protocol address and return；Receive the internet protocol address that the Node evaluation equipment returns；Its
In, the fringe node in acquired internet protocol address instruction content distributing network.
Alternatively, above-mentioned computer-readable medium carries one or more program, when said one or multiple programs
When being executed by the electronic equipment, so that the electronic equipment: receiving the Node evaluation including at least two internet protocol addresses and request；
From at least two internet protocol address, internet protocol address is chosen；Return to the internet protocol address selected；Wherein,
The fringe node in internet protocol address instruction content distributing network received.
The calculating of the operation for executing the disclosure can be write with one or more programming languages or combinations thereof
Machine program code, above procedure design language include object oriented program language-such as Java, Smalltalk, C+
+, it further include conventional procedural programming language-such as " C " language or similar programming language.Program code can
Fully to execute, partly execute on the user computer on the user computer, be executed as an independent software package,
Part executes on the remote computer or executes on a remote computer or server completely on the user computer for part.
In situations involving remote computers, remote computer can pass through the network of any kind --- including local area network (LAN)
Or wide area network (WAN)-is connected to subscriber computer, or, it may be connected to outer computer (such as utilize Internet service
Provider is connected by internet).
Flow chart and block diagram in attached drawing are illustrated according to the system of the various embodiments of the disclosure, method and computer journey
The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation
A part of one module, program segment or code of table, a part of the module, program segment or code include one or more use
The executable instruction of the logic function as defined in realizing.It should also be noted that in some implementations as replacements, being marked in box
The function of note can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are actually
It can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it to infuse
Meaning, the combination of each box in block diagram and or flow chart and the box in block diagram and or flow chart can be with holding
The dedicated hardware based system of functions or operations as defined in row is realized, or can use specialized hardware and computer instruction
Combination realize.
Being described in unit involved in the embodiment of the present disclosure can be realized by way of software, can also be by hard
The mode of part is realized.Wherein, the title of unit does not constitute the restriction to the unit itself under certain conditions, for example, the
One acquiring unit is also described as " obtaining the unit of at least two internet protocol addresses ".
It should be appreciated that each section of the disclosure can be realized with hardware, software, firmware or their combination.
The above, the only specific embodiment of the disclosure, but the protection scope of the disclosure is not limited thereto, it is any
Those familiar with the art is in the technical scope that the disclosure discloses, and any changes or substitutions that can be easily thought of, all answers
Cover within the protection scope of the disclosure.Therefore, the protection scope of the disclosure should be subject to the protection scope in claims.
Claims (12)
1. a kind of video cover generation method characterized by comprising
Target video is parsed, multiple parsing images are obtained；
Clustering processing is carried out to the multiple parsing image, obtains multiple candidate cover images；
Based on the critical point detection executed to the target object for including in the candidate cover image as a result, to the multiple candidate
The target object for including in cover image executes line-of-sight detection and detection of opening eyes, to filter out the candidate envelope for not meeting preset requirement
Face image；
Based on image quality evaluation network to the quality score of candidate's cover image remaining after filtering, the remaining time after filtering
It selects and chooses final cover image of the target image as the target video in cover image.
2. obtaining multiple parsings the method according to claim 1, wherein described parse target video
Image, comprising:
All video frame performance objectives detection to including in the target video；
It is based on target detection as a result, whether judging to form in the video frame of target video comprising the target object；
If so, setting the parsing image for the video frame comprising the target object.
3. the method according to claim 1, wherein it is described to the multiple parsing image carry out clustering processing,
Obtain multiple candidate cover images, comprising:
The cluster calculation of k class is executed to the multiple parsing image；
Select the image for meeting preset condition as the candidate cover image in each cluster.
4. according to the method described in claim 3, it is characterized in that, the cluster for executing k class to the multiple parsing image
It calculates, comprising:
It is initial cluster center that k sample point is chosen in the multiple parsing image, is denoted as z1 (l), z2 (l) ... ... zk
(l), iteration serial number l=1；
All samples are assigned in k class ω j (k) representated by each cluster centre using Nearest Neighbor Method, it is all kinds of included
Sample number is Nj (l)；
All kinds of centers of gravity is calculated, the center of gravity being calculated is determined as to new cluster centre；
For iteration j, judge whether the value of zj (l+1) and zj (l) is identical, when zj (l+1) ≠ zj (l) Shi Jixu iteration meter
It calculates, as zj (l+1)=zj (l), stops iterative calculation.
5. the method according to claim 1, wherein described to the mesh for including in the multiple candidate cover image
Before marking object execution line-of-sight detection and detection of opening eyes, the method also includes:
Quality evaluation is executed to the multiple candidate cover image using pre-set convolutional neural networks；
Based on the quality evaluation as a result, determining the quality score of each image in the multiple candidate cover image.
6. the method according to claim 1, wherein described to the mesh for including in the multiple candidate cover image
It marks object and executes line-of-sight detection and detection of opening eyes, comprising:
Obtain the critical point detection result executed to the target object for including in candidate cover image；
Based on the critical point detection as a result, determining pupil key point, eyeball key point and eye profile key point；
Based on the pupil key point, eyeball key point and eye profile key point, modeling is executed to the eyeball of the target object
Operation；
Based on modelling operability as a result, being detected the target object is executed line-of-sight detection and be opened eyes.
7. according to the method described in claim 6, it is characterized in that, it is described based on modelling operability as a result, come to the target
Object executes line-of-sight detection and detection of opening eyes, comprising:
Go out the key point of eyeball using CLNF model inspection；
Three-dimensional modeling is carried out to eyes；
It connects origin and forms a ray to pupil center, its intersection point with eyeball is calculated, by eyeball center to intersection point direction
Vector is as direction of visual lines.
8. according to the method described in claim 6, it is characterized in that, it is described based on modelling operability as a result, come to the target
Object executes line-of-sight detection and detection of opening eyes, comprising:
By modelling operability as a result, calculating the ratio of width to height of external eyes profile；
Judge whether the target object is in eyes-open state by judging whether described the ratio of width to height is greater than preset threshold.
9. the method according to claim 1, wherein the image quality evaluation network that is based on is to remaining after filtering
Candidate cover image quality score, from after filtering it is remaining candidate cover image in choose target image as the target
The final cover image of video, comprising:
The highest candidate cover image of quality score is chosen from candidate's cover image remaining after filtering；
Using the highest candidate cover image of the quality score as the final cover image of the target video.
10. a kind of video cover generating means characterized by comprising
Parsing module obtains multiple parsing images for parsing to target video；
Cluster module obtains multiple candidate cover images for carrying out clustering processing to the multiple parsing image；
Filtering module, for based on the critical point detection executed to the target object for including in the candidate cover image as a result,
Line-of-sight detection is executed and detection of opening eyes to the target object for including in the multiple candidate cover image, with filter out do not meet it is pre-
If it is required that candidate cover image；
Selecting module, for based on image quality evaluation network to after filtering it is remaining candidate cover image quality score, from
Final cover image of the target image as the target video is chosen in remaining candidate's cover image after filtering.
11. a kind of electronic equipment, which is characterized in that the electronic equipment includes:
At least one processor；And
The memory being connect at least one described processor communication；Wherein,
The memory is stored with the instruction that can be executed by least one described processor, and described instruction is by described at least one
It manages device to execute, so that at least one described processor is able to carry out the generation of video cover described in aforementioned any claim 1-9
Method.
12. a kind of non-transient computer readable storage medium, which stores computer instruction,
The computer instruction is for making the computer execute video cover generation method described in aforementioned any claim 1-9.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910622565.6A CN110381368A (en) | 2019-07-11 | 2019-07-11 | Video cover generation method, device and electronic equipment |
| PCT/CN2020/096719 WO2021004247A1 (en) | 2019-07-11 | 2020-06-18 | Method and apparatus for generating video cover and electronic device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910622565.6A CN110381368A (en) | 2019-07-11 | 2019-07-11 | Video cover generation method, device and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110381368A true CN110381368A (en) | 2019-10-25 |
Family
ID=68252598
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910622565.6A Pending CN110381368A (en) | 2019-07-11 | 2019-07-11 | Video cover generation method, device and electronic equipment |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN110381368A (en) |
| WO (1) | WO2021004247A1 (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110929070A (en) * | 2019-12-09 | 2020-03-27 | 北京字节跳动网络技术有限公司 | Image processing method, image processing device, electronic equipment and storage medium |
| CN111062314A (en) * | 2019-12-13 | 2020-04-24 | 腾讯科技（深圳）有限公司 | Image selection method and device, computer readable storage medium and electronic equipment |
| CN111464833A (en) * | 2020-03-23 | 2020-07-28 | 腾讯科技（深圳）有限公司 | Target image generation method, target image generation device, medium, and electronic apparatus |
| CN111491182A (en) * | 2020-04-23 | 2020-08-04 | 百度在线网络技术（北京）有限公司 | Method and device for video cover storage and analysis |
| CN111859022A (en) * | 2020-07-07 | 2020-10-30 | 咪咕文化科技有限公司 | Cover generation method, electronic device and computer-readable storage medium |
| WO2021004247A1 (en) * | 2019-07-11 | 2021-01-14 | 北京字节跳动网络技术有限公司 | Method and apparatus for generating video cover and electronic device |
| CN112308864A (en) * | 2020-10-27 | 2021-02-02 | 北京达佳互联信息技术有限公司 | Image processing method and device, electronic equipment and storage medium |
| CN112749711A (en) * | 2020-08-04 | 2021-05-04 | 腾讯科技（深圳）有限公司 | Video acquisition method and device and storage medium |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113179421B (en) * | 2021-04-01 | 2023-03-10 | 影石创新科技股份有限公司 | Video cover selection method and device, computer equipment and storage medium |
| CN113723242B (en) * | 2021-08-20 | 2024-04-26 | 湖南全航信息通信有限公司 | Visual lie detection method based on video terminal, electronic equipment and medium |
| CN114297433B (en) * | 2021-12-28 | 2024-04-19 | 抖音视界有限公司 | Method, device, equipment and storage medium for searching question and answer result |
| CN114449362B (en) * | 2022-03-17 | 2023-08-22 | 腾讯科技（上海）有限公司 | Video cover selection method, device, equipment and storage medium |
| CN115134677A (en) * | 2022-05-30 | 2022-09-30 | 一点灵犀信息技术(广州)有限公司 | Video cover selection method and device, electronic equipment and computer storage medium |
| CN117689782A (en) * | 2024-02-02 | 2024-03-12 | 腾讯科技（深圳）有限公司 | Method, device, equipment and storage medium for generating poster image |
Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110047163A1 (en) * | 2009-08-24 | 2011-02-24 | Google Inc. | Relevance-Based Image Selection |
| CN102830793A (en) * | 2011-06-16 | 2012-12-19 | 北京三星通信技术研究有限公司 | Sight tracking method and sight tracking device |
| EP2669894A1 (en) * | 2012-05-31 | 2013-12-04 | Samsung Electronics Co., Ltd | Method for creating thumbnail images of videos and an electronic device for display thereof |
| CN105894031A (en) * | 2016-03-31 | 2016-08-24 | 青岛海信移动通信技术股份有限公司 | Photo selection method and photo selection device |
| CN106529406A (en) * | 2016-09-30 | 2017-03-22 | 广州华多网络科技有限公司 | Method and device for acquiring video abstract image |
| CN107147939A (en) * | 2017-05-05 | 2017-09-08 | 百度在线网络技术（北京）有限公司 | Method and apparatus for adjusting net cast front cover |
| CN108108684A (en) * | 2017-12-15 | 2018-06-01 | 杭州电子科技大学 | A kind of attention detection method for merging line-of-sight detection |
| CN108733676A (en) * | 2017-04-14 | 2018-11-02 | 合信息技术（北京）有限公司 | The extracting method and device of video thumbnails |
| CN108833939A (en) * | 2018-06-20 | 2018-11-16 | 北京优酷科技有限公司 | Generate the method and device of the poster of video |
| CN108833942A (en) * | 2018-06-28 | 2018-11-16 | 北京达佳互联信息技术有限公司 | Video cover choosing method, device, computer equipment and storage medium |
| CN109165301A (en) * | 2018-09-13 | 2019-01-08 | 北京字节跳动网络技术有限公司 | Video cover selection method, device and computer readable storage medium |
| CN109583292A (en) * | 2018-10-11 | 2019-04-05 | 杭州电子科技大学 | A kind of visibility region detection method |
| CN110555426A (en) * | 2019-09-11 | 2019-12-10 | 北京儒博科技有限公司 | Sight line detection method, device, equipment and storage medium |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20100052676A (en) * | 2008-11-11 | 2010-05-20 | 삼성전자주식회사 | Apparatus for albuming contents and method thereof |
| CN107392159A (en) * | 2017-07-27 | 2017-11-24 | 竹间智能科技（上海）有限公司 | A kind of facial focus detecting system and method |
| CN107832725A (en) * | 2017-11-17 | 2018-03-23 | 北京奇虎科技有限公司 | Video front cover extracting method and device based on evaluation index |
| CN108875524B (en) * | 2018-01-02 | 2021-03-02 | 北京旷视科技有限公司 | Sight estimation method, device, system and storage medium |
| CN109002812A (en) * | 2018-08-08 | 2018-12-14 | 北京未来媒体科技股份有限公司 | A kind of method and device of intelligent recognition video cover |
| CN109635747A (en) * | 2018-12-14 | 2019-04-16 | 央视国际网络无锡有限公司 | The automatic abstracting method of video cover and device |
| CN110381368A (en) * | 2019-07-11 | 2019-10-25 | 北京字节跳动网络技术有限公司 | Video cover generation method, device and electronic equipment |
-
2019
- 2019-07-11 CN CN201910622565.6A patent/CN110381368A/en active Pending
-
2020
- 2020-06-18 WO PCT/CN2020/096719 patent/WO2021004247A1/en active Application Filing
Patent Citations (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110047163A1 (en) * | 2009-08-24 | 2011-02-24 | Google Inc. | Relevance-Based Image Selection |
| CN102830793A (en) * | 2011-06-16 | 2012-12-19 | 北京三星通信技术研究有限公司 | Sight tracking method and sight tracking device |
| EP2669894A1 (en) * | 2012-05-31 | 2013-12-04 | Samsung Electronics Co., Ltd | Method for creating thumbnail images of videos and an electronic device for display thereof |
| CN105894031A (en) * | 2016-03-31 | 2016-08-24 | 青岛海信移动通信技术股份有限公司 | Photo selection method and photo selection device |
| CN106529406A (en) * | 2016-09-30 | 2017-03-22 | 广州华多网络科技有限公司 | Method and device for acquiring video abstract image |
| CN108733676A (en) * | 2017-04-14 | 2018-11-02 | 合信息技术（北京）有限公司 | The extracting method and device of video thumbnails |
| CN107147939A (en) * | 2017-05-05 | 2017-09-08 | 百度在线网络技术（北京）有限公司 | Method and apparatus for adjusting net cast front cover |
| CN108108684A (en) * | 2017-12-15 | 2018-06-01 | 杭州电子科技大学 | A kind of attention detection method for merging line-of-sight detection |
| CN108833939A (en) * | 2018-06-20 | 2018-11-16 | 北京优酷科技有限公司 | Generate the method and device of the poster of video |
| CN108833942A (en) * | 2018-06-28 | 2018-11-16 | 北京达佳互联信息技术有限公司 | Video cover choosing method, device, computer equipment and storage medium |
| CN109165301A (en) * | 2018-09-13 | 2019-01-08 | 北京字节跳动网络技术有限公司 | Video cover selection method, device and computer readable storage medium |
| CN109583292A (en) * | 2018-10-11 | 2019-04-05 | 杭州电子科技大学 | A kind of visibility region detection method |
| CN110555426A (en) * | 2019-09-11 | 2019-12-10 | 北京儒博科技有限公司 | Sight line detection method, device, equipment and storage medium |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021004247A1 (en) * | 2019-07-11 | 2021-01-14 | 北京字节跳动网络技术有限公司 | Method and apparatus for generating video cover and electronic device |
| CN110929070A (en) * | 2019-12-09 | 2020-03-27 | 北京字节跳动网络技术有限公司 | Image processing method, image processing device, electronic equipment and storage medium |
| CN111062314A (en) * | 2019-12-13 | 2020-04-24 | 腾讯科技（深圳）有限公司 | Image selection method and device, computer readable storage medium and electronic equipment |
| CN111062314B (en) * | 2019-12-13 | 2021-11-02 | 腾讯科技（深圳）有限公司 | Image selection method and device, computer readable storage medium and electronic equipment |
| CN111464833A (en) * | 2020-03-23 | 2020-07-28 | 腾讯科技（深圳）有限公司 | Target image generation method, target image generation device, medium, and electronic apparatus |
| CN111464833B (en) * | 2020-03-23 | 2023-08-04 | 腾讯科技（深圳）有限公司 | Target image generation method, target image generation device, medium and electronic device |
| CN111491182A (en) * | 2020-04-23 | 2020-08-04 | 百度在线网络技术（北京）有限公司 | Method and device for video cover storage and analysis |
| CN111859022A (en) * | 2020-07-07 | 2020-10-30 | 咪咕文化科技有限公司 | Cover generation method, electronic device and computer-readable storage medium |
| CN112749711A (en) * | 2020-08-04 | 2021-05-04 | 腾讯科技（深圳）有限公司 | Video acquisition method and device and storage medium |
| CN112749711B (en) * | 2020-08-04 | 2023-08-25 | 腾讯科技（深圳）有限公司 | Video acquisition method and device and storage medium |
| CN112308864A (en) * | 2020-10-27 | 2021-02-02 | 北京达佳互联信息技术有限公司 | Image processing method and device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2021004247A1 (en) | 2021-01-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110381368A (en) | Video cover generation method, device and electronic equipment | |
| CN110399848A (en) | Video cover generation method, device and electronic equipment | |
| CN110321958A (en) | Training method, the video similarity of neural network model determine method | |
| CN108491816A (en) | The method and apparatus for carrying out target following in video | |
| CN109299348A (en) | A kind of data query method, apparatus, electronic equipment and storage medium | |
| CN110267097A (en) | Video pushing method, device and electronic equipment based on characteristic of division | |
| CN103999032A (en) | Interestingness scoring of areas of interest included in a display element | |
| CN110222726A (en) | Image processing method, device and electronic equipment | |
| CN110278447A (en) | Video pushing method, device and electronic equipment based on continuous feature | |
| CN109495784A (en) | Information-pushing method, device, electronic equipment and computer readable storage medium | |
| CN110363753A (en) | Image quality measure method, apparatus and electronic equipment | |
| CN110674349B (en) | Video POI (Point of interest) identification method and device and electronic equipment | |
| CN110415276A (en) | Motion information calculation method, device and electronic equipment | |
| CN110189394A (en) | Shape of the mouth as one speaks generation method, device and electronic equipment | |
| CN110399847A (en) | Extraction method of key frame, device and electronic equipment | |
| CN110070063A (en) | Action identification method, device and the electronic equipment of target object | |
| CN110288037A (en) | Image processing method, device and electronic equipment | |
| CN110287810A (en) | Car door motion detection method, device and computer readable storage medium | |
| CN110287350A (en) | Image search method, device and electronic equipment | |
| CN110198473A (en) | Method for processing video frequency, device, electronic equipment and computer readable storage medium | |
| CN113096055B (en) | Training method and device for image generation model, electronic equipment and storage medium | |
| CN110188782A (en) | Image similarity determines method, apparatus, electronic equipment and readable storage medium storing program for executing | |
| CN110008926A (en) | The method and apparatus at age for identification | |
| CN107169005B (en) | POI recommendation method, device, equipment and computer readable storage medium | |
| CN111598923B (en) | Target tracking method and device, computer equipment and storage medium |
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: 20191025 |