CN110147722A - A kind of method for processing video frequency, video process apparatus and terminal device - Google Patents
A kind of method for processing video frequency, video process apparatus and terminal device Download PDFInfo
- Publication number
- CN110147722A CN110147722A CN201910288788.3A CN201910288788A CN110147722A CN 110147722 A CN110147722 A CN 110147722A CN 201910288788 A CN201910288788 A CN 201910288788A CN 110147722 A CN110147722 A CN 110147722A
- Authority
- CN
- China
- Prior art keywords
- video
- target
- frame
- processed
- target object
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 56
- 238000012545 processing Methods 0.000 title claims abstract description 37
- 230000008569 process Effects 0.000 title claims abstract description 22
- 238000013136 deep learning model Methods 0.000 claims abstract description 34
- 238000012549 training Methods 0.000 claims abstract description 28
- 238000004590 computer program Methods 0.000 claims description 20
- 230000002123 temporal effect Effects 0.000 claims description 6
- 235000013399 edible fruits Nutrition 0.000 claims 1
- 230000006870 function Effects 0.000 description 13
- 238000013135 deep learning Methods 0.000 description 12
- 238000010586 diagram Methods 0.000 description 8
- 238000013527 convolutional neural network Methods 0.000 description 6
- 230000004044 response Effects 0.000 description 4
- 230000006399 behavior Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 230000008859 change Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000005611 electricity Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000010485 coping Effects 0.000 description 1
- 238000003066 decision tree Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000001815 facial effect Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000007477 logistic regression Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000007618 network scheduling algorithm Methods 0.000 description 1
- 238000005192 partition Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 238000000926 separation method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/20—Scenes; Scene-specific elements in augmented reality scenes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/19—Recognition using electronic means
- G06V30/192—Recognition using electronic means using simultaneous comparisons or correlations of the image signals with a plurality of references
- G06V30/194—References adjustable by an adaptive method, e.g. learning
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Bioinformatics & Computational Biology (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Cheminformatics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Signal Processing (AREA)
- Databases & Information Systems (AREA)
- Television Signal Processing For Recording (AREA)
Abstract
The present invention is suitable for technical field of image processing, provides a kind of method for processing video frequency, video process apparatus and terminal device, the method for processing video frequency, comprising: obtain video to be processed;Determine the scene type of each video frame in the video to be processed;According to the scene type of each video frame, target identification is carried out to each video frame in the video to be processed by the first deep learning model after training, obtain target identification result, wherein, the target identification result indicates that the target object is corresponding with the scene type whether comprising target object in each video frame;According to the target identification as a result, multiple target frames of the identification video to be processed, wherein include the target object in the target frame;According to chronological order of each target frame in the video to be processed, successively multiple target frames are combined, obtain the target video.
Description
Technical field
The invention belongs to technical field of image processing more particularly to a kind of method for processing video frequency, video process apparatus and end
End equipment.
Background technique
Currently, video image technology is extensive in multiple fields (such as video monitoring, mobile terminal and social platform)
Using, and a large amount of video file is produced in application process.These video files usually contain a large amount of redundancy or not
Part and parcel, and the duration of these video files is often very long, causing user to need to take a long time could be from described
Required key message is inquired in video file.For example, user need from when a length of 24 hours monitor video in inquire
To some individual in the video section for executing specified activities, and currently, the mode that user can take is carried out to video file
Fast-forward play, and some the individual time corresponding to the video section for executing specified activities that can be will be seen that is marked,
To search required video section from the video file according to the corresponding time in subsequent use.But due to video file
In a large amount of redundancy or unessential part presence, the mode of the key message in inquiry video in this way takes a long time, use
Family is difficult to efficiently extract some key messages in video.
Summary of the invention
In view of this, the embodiment of the invention provides a kind of method for processing video frequency, video process apparatus and terminal device, it can
To identify and extract the key message in video (such as identify and extract the video section comprising target individual), to improve
User obtains the efficiency of the key message in video.
The first aspect of the embodiment of the present invention provides a kind of method for processing video frequency, comprising:
Obtain video to be processed;
Determine the scene type of each video frame in the video to be processed;
According to the scene type of each video frame, by the first deep learning model after training to the video to be processed
In each video frame carry out target identification, obtain target identification result, wherein the target identification result indicates each video
It whether include target object in frame, the target object is corresponding with the scene type;
According to the target identification as a result, multiple target frames of the identification video to be processed, wherein in the target frame
Include the target object;
According to chronological order of each target frame in the video to be processed, successively by multiple target frames into
Row combination, obtains the target video.
The second aspect of the embodiment of the present invention provides a kind of video process apparatus, comprising:
Module is obtained, for obtaining video to be processed;
Determining module, for determining the scene type of each video frame in the video to be processed;
First identification module passes through the first deep learning mould after training for the scene type according to each video frame
Type carries out target identification to each video frame in the video to be processed, obtains target identification result, wherein the target is known
Other result indicates that the target object is corresponding with the scene type whether comprising target object in each video frame;
Second identification module, for according to the target identification as a result, multiple target frames of the identification video to be processed,
It wherein, include the target object in the target frame;
Processing module successively will be more for the chronological order according to each target frame in the video to be processed
A target frame is combined, and obtains the target video.
The third aspect of the embodiment of the present invention provides a kind of terminal device, including memory, processor and is stored in
In the memory and the computer program that can run on the processor, when the processor executes the computer program
The step of realizing method as described above.
The fourth aspect of the embodiment of the present invention provides a kind of computer readable storage medium, the computer-readable storage
Media storage has the step of computer program, the computer program realizes method as described above when being executed by processor.
Existing beneficial effect is the embodiment of the present invention compared with prior art: in the embodiment of the present invention, obtaining to be processed
Video;Determine the scene type of each video frame in the video to be processed;According to the scene type of each video frame, pass through
The first deep learning model after training carries out target identification to each video frame in the video to be processed, obtains target and knows
Other result, wherein whether the target identification result indicates in each video frame to include target object, the target object and institute
It is corresponding to state scene type;According to the target identification as a result, multiple target frames of the identification video to be processed, wherein institute
It states in target frame comprising the target object;According to chronological order of each target frame in the video to be processed, according to
It is secondary to be combined multiple target frames, obtain the target video.The embodiment of the present invention is by determining each video frame
Scene type, and determined in each video frame by the deep learning model after training with the presence or absence of corresponding to the scene type
Target object can carry out targeted target identification according to different scene types, so that target identification result more has needle
To property, interference is less;According to the target identification as a result, identifying multiple target frames of the video to be processed, and according to described
Target frame obtains target video, and the key message that can accurately and efficiently extract in the video to be processed (is such as identified and mentioned
Take out the image section comprising target individual), and the target video being made of key message is obtained, and delete video to be processed
Middle redundancy, unessential video content, remains valuable video content, thus such as believing the key of monitor video
In the application scenarios such as breath lookup, user can carry out more efficient, more targeted processing according to the target video, be not necessarily to people
The operations such as work searches key message, editing synthesizes again.The embodiment of the present invention substantially increases user and obtains in video
The efficiency of key message, practicability and ease for use are stronger.
Detailed description of the invention
It to describe the technical solutions in the embodiments of the present invention more clearly, below will be to embodiment or description of the prior art
Needed in attached drawing be briefly described, it should be apparent that, the accompanying drawings in the following description is only of the invention some
Embodiment for those of ordinary skill in the art without creative efforts, can also be attached according to these
Figure obtains other attached drawings.
Fig. 1 is the implementation process schematic diagram for the method for processing video frequency that the embodiment of the present invention one provides;
Fig. 2 is the implementation process schematic diagram of method for processing video frequency provided by Embodiment 2 of the present invention;
Fig. 3 is the schematic diagram for the video process apparatus that the embodiment of the present invention three provides;
Fig. 4 is the schematic diagram for the terminal device that the embodiment of the present invention four provides.
Specific embodiment
In being described below, for illustration and not for limitation, the tool of such as particular system structure, technology etc is proposed
Body details, to understand thoroughly the embodiment of the present invention.However, it will be clear to one skilled in the art that there is no these specific
The present invention also may be implemented in the other embodiments of details.In other situations, it omits to well-known system, device, electricity
The detailed description of road and method, in case unnecessary details interferes description of the invention.
It should be appreciated that ought use in this specification and in the appended claims, term " includes " instruction is described special
Sign, entirety, step, operation, the presence of element and/or component, but be not precluded one or more of the other feature, entirety, step,
Operation, the presence or addition of element, component and/or its set.
It is also understood that mesh of the term used in this description of the invention merely for the sake of description specific embodiment
And be not intended to limit the present invention.As description of the invention and it is used in the attached claims, unless on
Other situations are hereafter clearly indicated, otherwise " one " of singular, "one" and "the" are intended to include plural form.
It will be further appreciated that the term "and/or" used in description of the invention and the appended claims is
Refer to any combination and all possible combinations of one or more of associated item listed, and including these combinations.
As used in this specification and in the appended claims, term " if " can be according to context quilt
Be construed to " when ... " or " once " or " in response to determination " or " in response to detecting ".Similarly, phrase " if it is determined that " or
" if detecting [described condition or event] " can be interpreted to mean according to context " once it is determined that " or " in response to true
It is fixed " or " once detecting [described condition or event] " or " in response to detecting [described condition or event] ".
In order to illustrate technical solutions according to the invention, the following is a description of specific embodiments.
Fig. 1 is the implementation process schematic diagram for the method for processing video frequency that the embodiment of the present invention one provides, the video as shown in Figure 1
Processing method may comprise steps of:
Step S101 obtains video to be processed.
The video to be processed can be the file for having carried out audio/video coding, and the coding mode of the video to be processed is
It is known, also, the coding mode is coding mode included in training data, wherein the training data is for instructing
Practice deep learning model.
Certainly, the video to be processed is also possible to the video not encoded, wherein each frame image includes complete
Image pixel information.
Step S102 determines the scene type of each video frame in the video to be processed.
In the embodiment of the present invention, the scene type of the video frame can be determined in several ways.For example, can pass through
Predetermined depth learning model determines the scene type of the video frame, it should be noted that the predetermined depth learning model can be with
It is identical as the first deep learning model after training described in step S103, it can also be different;The predetermined depth learning model
The scene type of each video frame in the video to be processed can be detected and determined one by one, it can also be according to a frame or multiframe
The scene type of the obtained video frame is detected to determine the subject scenes classification of the video to be processed (e.g., if more than pre-
If the scene type of the video frame of ratio is scene type A, it is determined that the subject scenes classification of the video to be processed is field
Scape classification A), and using the subject scenes classification as the scene type of each video frame of the video to be processed.In addition, can also
To determine each video frame in the video to be processed according to the title of the video to be processed, theme or default label
Scene type.
Step S103, according to the scene type of each video frame, by the first deep learning model after training to described
Each video frame in video to be processed carries out target identification, obtains target identification result, wherein the target identification result refers to
Show that the target object is corresponding with the scene type whether comprising target object in each video frame.
In the embodiment of the present invention, different scene types can correspond to different target objects.Such as, however, it is determined that the view
The scene type of frequency frame is airport scene, then target object corresponding to the scene of airport may include that aircraft, people, other machinery are set
Standby (small transport device in such as airport), other flying object objects;If it is determined that the scene type of the video frame is animal
Garden scene, then target object corresponding to the scene of zoo may include animal, people etc..The scene type and the object
The corresponding relationship of body can be pre-set.
Illustratively, the first deep learning model may include such as ResNet model, R-CNN model, Fast R-
CNN model etc., the first deep learning model can there are many, be not limited thereto.
Optionally, may include in the target identification result target object included in the video frame type,
The information such as the position of target object in the video frame.
Optionally, by training after the first deep learning model to each video frame in the video to be processed into
It further include the trained first deep learning model before row target identification.
Specifically, the step of training the first deep learning model may include:
Training data is obtained, the training data includes corresponding at least one training video and the training video
Identification information, the identification information may include in training video accurate information on target object (type of such as target object,
Position etc.);
The training data is detected by the first deep learning model, and obtains testing result;
The parameter that the first deep learning model is adjusted according to the testing result, until described by adjusting after
The testing result that one deep learning model obtains meets preset condition, and (value of such as corresponding loss function is less than preset threshold
Deng), and using the first deep learning model adjusted as the first deep learning model after training.
Step S104, according to the target identification as a result, multiple target frames of the identification video to be processed, wherein institute
It states in target frame comprising the target object.
The target frame can indicate the key message in the video to be processed, and the target frame of the video to be processed
Specific identification method can be arranged according to the needs of practical application scene.For example, it may be determined that any one frame includes described
The video frame of target object is the target frame.Furthermore, it is possible to according to the information on target object, according to chronological order according to
Comprising the otherness between the video frame of the target object, (otherness may include the difference of target object for secondary judgement
Property, and can ignoring for the otherness of non-targeted object), and determine according to the otherness target of the video to be processed
Frame.Illustratively, the otherness between the video frame comprising the target object can be identified by more traditional method.Example
Such as, for each video frame, if the image district where the positioning of target object described in the video frame, the target object
Domain size etc. changes relative to a upper video frame or next frame video frame;Alternatively, if target described in a certain video frame
The positioning of object, described image region the difference degree relative to a upper video frame or next frame video frame such as size
Meet specified preset condition (area of such as offset of positioning more than image-region in default bias amount threshold value, two frame video frames
The ratio of the difference area that accounts for one of image-region be more than default fractional threshold etc.), it is determined that the video frame is mesh
Frame is marked, until traversing all video frames in the video to be processed.It is of course also possible to be judged respectively by default classifier
Whether a video frame is target frame.Illustratively, the classifier may include decision tree, logistic regression, naive Bayesian, mind
Through network scheduling algorithm.
Illustrate a kind of implementation of step S104 with a specific example below.
Such as, however, it is determined that the scene type of each frame video frame of the video to be processed is airport scene, the machine
Target object corresponding to the scene of field is aircraft, it is assumed that includes 20 frame video frames in the video to be processed, wherein preceding 10 frame regards
Frequency frame does not include aircraft, it is determined that preceding 10 frame video is non-targeted frame;And the 11st to the 20th frame includes inspection in the video frame of aircraft
It measures the aircraft position that aircraft position of the 15th frame into the 20th frame is respectively relative to its former frame video frame and offset occurs, then
Determine that the 14th frame to the 20th frame video frame is the target frame of the video to be processed.
Step S105, according to chronological order of each target frame in the video to be processed, successively by multiple institutes
It states target frame to be combined, obtains the target video.
It can only include the target frame in the video to be processed in the embodiment of the present invention, in the target video, therefore,
It is considered that deleting redundancy in video to be processed, unessential video content in the target video, remain valuable
Video content, to search etc. in application scenarios in the key message to monitor video, user can be according to the mesh
Mark video and carry out more efficient, more targeted processing, without manually being searched key message, editing the behaviour such as synthesize again
Make.In the target video, the target frame corresponding timing node information in video to be processed can be retained, it can also not
Retain.
In addition, it is optional, after obtaining the target video, can also include:
The target video is stored into specified memory space.
In the embodiment of the present invention, the memory space, which can be, to be pre-set, and is also possible to obtain the target
After video, the specified memory space is determined according to the instruction operation of user, then the target video is stored to specified
In memory space.
Due to usually containing a large amount of redundancy or unessential part in video to be processed, and these parts can occupy greatly
The space of amount and resource, even if this resource occupation and wave can not be improved well by modes such as existing compressing files
The problem of taking.Therefore, in the prior art, directly the video to be processed is stored, a large amount of storage resource can be wasted.And
In the embodiment of the present invention, can store the target video so that storage video needed for space greatly reduce, so as to improve due to
Storage resource bears excessive problem caused by video is excessive.
Optionally, if including target object in the video frame of the video to be processed, the target identification result also refers to
Show the region where the target object and/or characteristic point position;
Correspondingly, it is described according to the target identification as a result, multiple target frames of the identification video to be processed, comprising:
Region and/or characteristic point position according to where the target object indicated by the target identification result, sentence
Whether each video frame broken in the video to be processed meets the first preset condition;
The video frame for meeting first preset condition is determined as to the target frame of the video to be processed.
In the embodiment of the present invention, region and/or characteristic point position where the target object may include the target
Top left co-ordinate, bottom right angular coordinate, center point coordinate and the target of image-region area, image-region where object
At least one of information such as the characteristic point coordinate of object.The characteristic point can be according to the target object and application scenarios
It determines, for example, the characteristic point may include the characteristic point of facial face if the target object is face.Described
It is pre-set that one preset condition can be user, also, first preset condition can there are many, for example, it may be if
Image-region area where the characteristic point coordinate of target object described in the video frame, the target object etc. is relative to upper
One video frame or next frame video frame change;Alternatively, if the characteristic point coordinate of target object described in a certain video frame,
Described image region area etc. meets specified preset condition relative to the difference degree of a upper video frame or next frame video frame
(as the offset of positioning accounts for one of them more than the difference of the area of image-region in default bias amount threshold value, two frame video frames
The ratio of the area of image-region is more than default fractional threshold etc.).
Wherein, optionally, the position where the target object according to indicated by the target identification result, judges institute
Whether each video frame stated in video to be processed meets the first preset condition, comprising:
For each video frame in the video to be processed including the target object, according to the target identification result
Region and/or characteristic point position where indicated target object, judge in the video frame, the target object relative to
The difference in the region and/or characteristic point position where target object in a upper video frame for the video frame or in next video frame
Whether off course degree meets the first preset condition;
Correspondingly, the target that the video frame for meeting first preset condition is determined as to the video to be processed
Frame, comprising:
If in the video frame, the target object is relative in a upper video frame for the video frame or next video frame
In target object where region and/or the difference degree of characteristic point position meet the first preset condition, then by the video frame
Target frame as the video to be processed.
Illustratively, in the embodiment of the present invention, the difference journey in region and/or characteristic point position where the target object
Degree may include one of characteristic point position degrees of offset, image-region variation degree etc. or a variety of, wherein described image
Regional change degree may include the variation degree of the area in described image region, or may include in described image region,
The part that changing occurs in pixel value accounts for the ratio size in described image region.At this point, illustrative, first preset condition can
The offset of characteristic point coordinate to be such as target object is more than default bias amount threshold value, image-region in two frame video frames
The ratio that the difference of area accounts for the area of one of image-region is more than default fractional threshold etc..First preset condition
It can be configured, be not limited thereto according to concrete application scene.
Optionally, the scene type of each video frame in the determination video to be processed, comprising:
Scene knowledge is carried out to each video frame in the video to be processed by the second deep learning model after training
Not, with the scene type of each video frame in the determination video to be processed;
Alternatively,
According to the title of the video to be processed, theme or default label, determine each in the video to be processed
The scene type of video frame.
In the embodiment of the present invention, the second deep learning model can be identical as the first deep learning model,
It can be different.Illustratively, if the second deep learning model is identical as the first deep learning model, then described
Two deep learning models (the first deep learning model in other words) may include two cascade deep learning submodels, and respectively
The scene type of each video frame in the video to be processed is determined by cascade deep learning submodel and to described
Each video frame in video to be processed carries out target identification.Assuming that the deep learning submodel is respectively the first deep learning
Submodel and the second deep learning submodel, then can by the first deep learning submodel described in the video input to be processed,
The first output of the first deep learning submodel is obtained as a result, output result instruction comes in the video to be processed
Each video frame scene Recognition, then the first of the first deep learning submodel the output result is inputted into the second depth
Submodel is practised, obtains the second output of the second deep learning submodel as a result, simultaneously can be according to the second output result
Obtain target identification result.Alternatively, the second deep learning model can also only include a convolutional neural networks model, and
The scene type and target identification result of each video frame are obtained by the convolutional neural networks model.
Illustratively, the second deep learning model may include such as ResNet model, R-CNN model, Fast R-
CNN model etc., the second deep learning model can there are many, be not limited thereto.
Optionally, after the multiple target frames for identifying the video to be processed, further includes:
The target frame is identified in the video to be processed.
Illustratively, timing node information, field corresponding to the target frame can be identified in the video to be processed
One of scape classification information, information on target object (type of such as target object, characteristic point, positioning, image-region) information
Or it is a variety of.The mode of the mark can there are many, for example, the target can be identified by rectangle frame of true-to-shape etc.
The image-region of object can identify timing node corresponding to the target frame in the time schedule item of the video to be processed
Etc..
Wherein, optionally, it when the continuous multiple frames video frame of the video to be processed is target frame, can identify described
First frame and last frame in continuous multiple frames video frame.
The embodiment of the present invention passes through the scene type for determining each video frame, and true by the deep learning model after training
With the presence or absence of target object corresponding to the scene type in fixed each video frame, can be had according to different scene types
Targetedly target identification, so that target identification result is more targeted, interference is less;According to the target identification as a result, knowing
Multiple target frames of the not described video to be processed, and target video is obtained according to the target frame, it can accurately and efficiently extract
To the key message (such as identify and extract the image section comprising target individual) in the video to be processed, and obtain by closing
The target video of key information composition, and redundancy in video to be processed, unessential video content are deleted, it remains valuable
Video content, to search etc. in application scenarios in the key message to monitor video, user can be according to the mesh
Mark video and carry out more efficient, more targeted processing, without manually being searched key message, editing the behaviour such as synthesize again
Make.The embodiment of the present invention substantially increases the efficiency that user obtains the key message of video, and practicability and ease for use are stronger.
On the basis of the above embodiments, Fig. 2 is the implementation process of method for processing video frequency provided by Embodiment 2 of the present invention
Schematic diagram, the method for processing video frequency as shown in Figure 2 may comprise steps of:
Step S201 obtains video to be processed;
Step S202 determines the scene type of each video frame in the video to be processed;
Step S203, according to the scene type of each video frame, by the first deep learning model after training to described
Each video frame in video to be processed carries out target identification, obtains target identification result, wherein the target identification result refers to
Show that the target object is corresponding with the scene type whether comprising target object in each video frame;
Step S204, according to the target identification as a result, multiple target frames of the identification video to be processed, wherein institute
It states in target frame comprising the target object;
Step S205, according to chronological order of each target frame in the video to be processed, successively by multiple institutes
It states target frame to be combined, obtains the target video;
The present embodiment step S201, S202, S203, S204, S205 and above-mentioned steps S101, S102, S103, S104,
S105 is identical, and for details, reference can be made to step S101, S102, S103, S104, S105 associated descriptions, and details are not described herein.
Step S206 establishes information index according to the target frame.
In the embodiment of the present invention, the information index may include being directed toward the target frame and/or being directed toward and the target
The link of the relevant information of frame or the inventory of pointer.The information index can be provided in direction specified data or node etc.
The pointer of appearance or link, to quickly access information relevant to the target frame by the information index.The information
Index can be set in interface associated and/or associated with the target video with the video to be processed, also, described
Information index can be directed toward the video to be processed and/or the target video.In addition, it should be noted that, the information rope
Draw can be embedded in the video to be processed perhaps in the target video or, or the video to be processed and
Information interface (such as database, table, text etc.) except target video, display mode can there are many.
Optionally, described according to the target frame, establish information index, comprising:
The information index list about the target frame is established, the information index list includes the target frame described
In the scene type information of temporal information, the target frame and the information on target object of the target frame in video to be processed
It is at least one.
In the embodiment of the present invention, the temporal information can indicate that the target frame is corresponding in the video to be processed
Timing node;Pass through the temporal information, it will be appreciated that the target frame is locating in the time shaft of the video to be processed
Time location so that user can understand the target frame in different time referential by the temporal information
Situation, without because by the target frame after being extracted in the video to be processed, segmentum intercalaris when being just lost original
Point information.The scene type information can indicate that the scene type of the target frame, the information on target object may include
The type of the target object, positioning, image-region etc. information relevant to the target object.
In the embodiment of the present invention, by establishing information index according to the target frame, it may be implemented to the target frame
The efficient retrieval of relevant information greatly improves user to the lookup of the key message in the video to be processed, editor and divides
The treatment effeciency of the operations such as analysis.
It should be understood that the size of the serial number of each step is not meant that the order of the execution order in above-described embodiment, each process
Execution sequence should be determined by its function and internal logic, the implementation process without coping with the embodiment of the present invention constitutes any limit
It is fixed.
Fig. 3 is the schematic diagram for the video process apparatus that the embodiment of the present invention three provides.For ease of description, illustrate only with
The relevant part of the embodiment of the present invention.
The video process apparatus 300 includes:
Module 301 is obtained, for obtaining video to be processed;
Determining module 302, for determining the scene type of each video frame in the video to be processed;
First identification module 303 passes through the first deep learning after training for the scene type according to each video frame
Model carries out target identification to each video frame in the video to be processed, obtains target identification result, wherein the target
Recognition result indicates that the target object is corresponding with the scene type whether comprising target object in each video frame;
Second identification module 304, for according to the target identification as a result, multiple targets of the identification video to be processed
Frame, wherein include the target object in the target frame;
Processing module 305 successively will for the chronological order according to each target frame in the video to be processed
Multiple target frames are combined, and obtain the target video.
Optionally, if including target object in the video frame of the video to be processed, the target identification result also refers to
Show the region where the target object and/or characteristic point position;
Correspondingly, second identification module 304 specifically includes:
Judging unit, for where the target object according to indicated by the target identification result region and/or
Characteristic point position, judges whether each video frame in the video to be processed meets the first preset condition;
Determination unit, for the video frame for meeting first preset condition to be determined as to the target of the video to be processed
Frame.
Optionally, the judging unit is specifically used for:
For each video frame in the video to be processed including the target object, according to the target identification result
Region and/or characteristic point position where indicated target object, judge in the video frame, the target object relative to
The difference in the region and/or characteristic point position where target object in a upper video frame for the video frame or in next video frame
Whether off course degree meets the first preset condition;
Correspondingly, the determination unit is specifically used for:
If in the video frame, the target object is relative in a upper video frame for the video frame or next video frame
In target object where region and/or the difference degree of characteristic point position meet the first preset condition, then by the video frame
Target frame as the video to be processed.
Optionally, the determining module 302 is specifically used for:
Scene knowledge is carried out to each video frame in the video to be processed by the second deep learning model after training
Not, with the scene type of each video frame in the determination video to be processed;
Alternatively,
According to the title of the video to be processed, theme or default label, determine each in the video to be processed
The scene type of video frame.
Optionally, the video process apparatus 300 further include:
Mark module, for identifying the target frame in the video to be processed.
Optionally, the video process apparatus 300 further include:
Index module, for establishing information index according to the target frame.
Optionally, the index module is specifically used for:
The information index list about the target frame is established, the information index list includes the target frame described
In the scene type information of temporal information, the target frame and the information on target object of the target frame in video to be processed
It is at least one.
The embodiment of the present invention passes through the scene type for determining each video frame, and true by the deep learning model after training
With the presence or absence of target object corresponding to the scene type in fixed each video frame, can be had according to different scene types
Targetedly target identification, so that target identification result is more targeted, interference is less;According to the target identification as a result, knowing
Multiple target frames of the not described video to be processed, and target video is obtained according to the target frame, it can accurately and efficiently extract
To the key message (such as identify and extract the image section comprising target individual) in the video to be processed, and obtain by closing
The target video of key information composition, and redundancy in video to be processed, unessential video content are deleted, it remains valuable
Video content, to search etc. in application scenarios in the key message to monitor video, user can be according to the mesh
Mark video and carry out more efficient, more targeted processing, without manually being searched key message, editing the behaviour such as synthesize again
Make.The embodiment of the present invention substantially increases the efficiency for the key message that user obtains in video, and practicability and ease for use are stronger.
Fig. 4 is the schematic diagram for the terminal device that the embodiment of the present invention four provides.As shown in figure 4, the terminal of the embodiment is set
Standby 4 include: processor 40, memory 41 and are stored in the meter that can be run in the memory 41 and on the processor 40
Calculation machine program 42.The processor 40 is realized when executing the computer program 42 in above-mentioned each method for processing video frequency embodiment
The step of, such as step 101 shown in FIG. 1 is to 105.Alternatively, realization when the processor 40 executes the computer program 42
The function of each module/unit in above-mentioned each Installation practice, such as the function of module 301 to 305 shown in Fig. 3.
Illustratively, the computer program 42 can be divided into one or more module/units, it is one or
Multiple module/units are stored in the memory 41, and are executed by the processor 40, to complete the present invention.Described one
A or multiple module/units can be the series of computation machine program instruction section that can complete specific function, which is used for
Implementation procedure of the computer program 42 in the terminal device 4 is described.For example, the computer program 42 can be divided
It is cut into and obtains module, determining module, the first identification module, the second identification module, processing module, each module concrete function is as follows:
Module is obtained, for obtaining video to be processed;
Determining module, for determining the scene type of each video frame in the video to be processed;
First identification module passes through the first deep learning mould after training for the scene type according to each video frame
Type carries out target identification to each video frame in the video to be processed, obtains target identification result, wherein the target is known
Other result indicates that the target object is corresponding with the scene type whether comprising target object in each video frame;
Second identification module, for according to the target identification as a result, multiple target frames of the identification video to be processed,
It wherein, include the target object in the target frame;
Processing module successively will be more for the chronological order according to each target frame in the video to be processed
A target frame is combined, and obtains the target video.
The terminal device 4 can be the calculating such as desktop PC, notebook, palm PC and cloud server and set
It is standby.The terminal device may include, but be not limited only to, processor 40, memory 41.It will be understood by those skilled in the art that Fig. 4
The only example of terminal device 4 does not constitute the restriction to terminal device 4, may include than illustrating more or fewer portions
Part perhaps combines certain components or different components, such as the terminal device can also include input-output equipment, net
Network access device, bus etc..
Alleged processor 40 can be central processing unit (Central Processing Unit, CPU), can also be
Other general processors, digital signal processor (Digital Signal Processor, DSP), specific integrated circuit
(Application Specific Integrated Circuit, ASIC), ready-made programmable gate array (Field-
Programmable Gate Array, FPGA) either other programmable logic device, discrete gate or transistor logic,
Discrete hardware components etc..General processor can be microprocessor or the processor is also possible to any conventional processor
Deng.
The memory 41 can be the internal storage unit of the terminal device 4, such as the hard disk or interior of terminal device 4
It deposits.The memory 41 is also possible to the External memory equipment of the terminal device 4, such as be equipped on the terminal device 4
Plug-in type hard disk, intelligent memory card (Smart Media Card, SMC), secure digital (Secure Digital, SD) card dodge
Deposit card (Flash Card) etc..Further, the memory 41 can also both include the storage inside list of the terminal device 4
Member also includes External memory equipment.The memory 41 is for storing needed for the computer program and the terminal device
Other programs and data.The memory 41 can be also used for temporarily storing the data that has exported or will export.
It is apparent to those skilled in the art that for convenience of description and succinctly, only with above-mentioned each function
Can unit, module division progress for example, in practical application, can according to need and by above-mentioned function distribution by different
Functional unit, module are completed, i.e., the internal structure of described device is divided into different functional unit or module, more than completing
The all or part of function of description.Each functional unit in embodiment, module can integrate in one processing unit, can also
To be that each unit physically exists alone, can also be integrated in one unit with two or more units, it is above-mentioned integrated
Unit both can take the form of hardware realization, can also realize in the form of software functional units.In addition, each function list
Member, the specific name of module are also only for convenience of distinguishing each other, the protection scope being not intended to limit this application.Above system
The specific work process of middle unit, module, can refer to corresponding processes in the foregoing method embodiment, and details are not described herein.
In the above-described embodiments, it all emphasizes particularly on different fields to the description of each embodiment, is not described in detail or remembers in some embodiment
The part of load may refer to the associated description of other embodiments.
Those of ordinary skill in the art may be aware that list described in conjunction with the examples disclosed in the embodiments of the present disclosure
Member and algorithm steps can be realized with the combination of electronic hardware or computer software and electronic hardware.These functions are actually
It is implemented in hardware or software, the specific application and design constraint depending on technical solution.Professional technician
Each specific application can be used different methods to achieve the described function, but this realization is it is not considered that exceed
The scope of the present invention.
In embodiment provided by the present invention, it should be understood that disclosed device/terminal device and method, it can be with
It realizes by another way.For example, device described above/terminal device embodiment is only schematical, for example, institute
The division of module or unit is stated, only a kind of logical function partition, there may be another division manner in actual implementation, such as
Multiple units or components can be combined or can be integrated into another system, or some features can be ignored or not executed.Separately
A bit, shown or discussed mutual coupling or direct-coupling or communication connection can be through some interfaces, device
Or the INDIRECT COUPLING or communication connection of unit, it can be electrical property, mechanical or other forms.
The unit as illustrated by the separation member may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.It can select some or all of unit therein according to the actual needs to realize the mesh of this embodiment scheme
's.
It, can also be in addition, the functional units in various embodiments of the present invention may be integrated into one processing unit
It is that each unit physically exists alone, can also be integrated in one unit with two or more units.Above-mentioned integrated list
Member both can take the form of hardware realization, can also realize in the form of software functional units.
If the integrated module/unit be realized in the form of SFU software functional unit and as independent product sale or
In use, can store in a computer readable storage medium.Based on this understanding, the present invention realizes above-mentioned implementation
All or part of the process in example method, can also instruct relevant hardware to complete, the meter by computer program
Calculation machine program can be stored in a computer readable storage medium, the computer program when being executed by processor, it can be achieved that on
The step of stating each embodiment of the method.Wherein, the computer program includes computer program code, the computer program
Code can be source code form, object identification code form, executable file or certain intermediate forms etc..Computer-readable Jie
Matter may include: can carry the computer program code any entity or device, recording medium, USB flash disk, mobile hard disk,
Magnetic disk, CD, computer storage, read-only memory (ROM, Read-Only Memory), random access memory (RAM,
Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium etc..It should be noted that described
The content that computer-readable medium includes can carry out increasing appropriate according to the requirement made laws in jurisdiction with patent practice
Subtract, such as does not include electric carrier signal and electricity according to legislation and patent practice, computer-readable medium in certain jurisdictions
Believe signal.
Embodiment described above is merely illustrative of the technical solution of the present invention, rather than its limitations;Although referring to aforementioned reality
Applying example, invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each
Technical solution documented by embodiment is modified or equivalent replacement of some of the technical features;And these are modified
Or replacement, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution should all
It is included within protection scope of the present invention.
Claims (10)
1. a kind of method for processing video frequency characterized by comprising
Obtain video to be processed;
Determine the scene type of each video frame in the video to be processed;
According to the scene type of each video frame, by the first deep learning model after training in the video to be processed
Each video frame carries out target identification, obtains target identification result, wherein the target identification result indicates in each video frame
It whether include target object, the target object is corresponding with the scene type;
According to the target identification as a result, multiple target frames of the identification video to be processed, wherein include in the target frame
The target object;
According to chronological order of each target frame in the video to be processed, multiple target frames are successively subjected to group
It closes, obtains the target video.
2. method for processing video frequency as described in claim 1, which is characterized in that if including in the video frame of the video to be processed
Target object, then the target identification result also indicates the region where the target object and/or characteristic point position;
Correspondingly, it is described according to the target identification as a result, multiple target frames of the identification video to be processed, comprising:
Region and/or characteristic point position according to where the target object indicated by the target identification result, judge institute
Whether each video frame stated in video to be processed meets the first preset condition;
The video frame for meeting first preset condition is determined as to the target frame of the video to be processed.
3. method for processing video frequency as claimed in claim 2, which is characterized in that described according to indicated by the target identification result
Target object where position, judge whether each video frame in the video to be processed meets the first preset condition, wrap
It includes:
It is signified according to the target identification result for each video frame in the video to be processed including the target object
Region and/or characteristic point position where the target object shown judge in the video frame, the target object is relative to the view
The difference journey in the region and/or characteristic point position where target object in a upper video frame for frequency frame or in next video frame
Whether degree meets the first preset condition;
Correspondingly, the target frame that the video frame for meeting first preset condition is determined as to the video to be processed, packet
It includes:
If in the video frame, the target object is relative in a upper video frame for the video frame or in next video frame
The difference degree in region and/or characteristic point position where target object meets the first preset condition, then using the video frame as
The target frame of the video to be processed.
4. method for processing video frequency as described in claim 1, which is characterized in that each in the determination video to be processed
The scene type of video frame, comprising:
Scene Recognition is carried out to each video frame in the video to be processed by the second deep learning model after training, with
Determine the scene type of each video frame in the video to be processed;
Alternatively,
According to the title of the video to be processed, theme or default label, each video in the video to be processed is determined
The scene type of frame.
5. method for processing video frequency as described in claim 1, which is characterized in that in the multiple targets for identifying the video to be processed
After frame, further includes:
The target frame is identified in the video to be processed.
6. the method for processing video frequency as described in claim 1 to 5 any one, which is characterized in that identifying the view to be processed
After multiple target frames of frequency, further includes:
According to the target frame, information index is established.
7. method for processing video frequency as claimed in claim 6, which is characterized in that it is described according to the target frame, establish information rope
Draw, comprising:
Establish information index list about the target frame, the information index list include the target frame described wait locate
In the information on target object of temporal information, the scene type information of the target frame and the target frame in reason video at least
It is a kind of.
8. a kind of video process apparatus characterized by comprising
Module is obtained, for obtaining video to be processed;
Determining module, for determining the scene type of each video frame in the video to be processed;
First identification module passes through the first deep learning model pair after training for the scene type according to each video frame
Each video frame in the video to be processed carries out target identification, obtains target identification result, wherein the target identification knot
Fruit indicates that the target object is corresponding with the scene type whether comprising target object in each video frame;
Second identification module, for according to the target identification as a result, multiple target frames of the identification video to be processed,
In, it include the target object in the target frame;
Processing module, for the chronological order according to each target frame in the video to be processed, successively by multiple institutes
It states target frame to be combined, obtains the target video.
9. a kind of terminal device, including memory, processor and storage are in the memory and can be on the processor
The computer program of operation, which is characterized in that the processor realizes such as claim 1 to 7 when executing the computer program
The step of any one method for processing video frequency.
10. a kind of computer readable storage medium, the computer-readable recording medium storage has computer program, and feature exists
In the step of realization method for processing video frequency as described in any one of claim 1 to 7 when the computer program is executed by processor
Suddenly.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910288788.3A CN110147722A (en) | 2019-04-11 | 2019-04-11 | A kind of method for processing video frequency, video process apparatus and terminal device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910288788.3A CN110147722A (en) | 2019-04-11 | 2019-04-11 | A kind of method for processing video frequency, video process apparatus and terminal device |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110147722A true CN110147722A (en) | 2019-08-20 |
Family
ID=67589739
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910288788.3A Pending CN110147722A (en) | 2019-04-11 | 2019-04-11 | A kind of method for processing video frequency, video process apparatus and terminal device |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110147722A (en) |
Cited By (27)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110569392A (en) * | 2019-08-28 | 2019-12-13 | 深圳市天视通电子科技有限公司 | multi-video processing system and method |
CN110659581A (en) * | 2019-08-29 | 2020-01-07 | 腾讯科技(深圳)有限公司 | Image processing method, device, equipment and storage medium |
CN110674781A (en) * | 2019-10-07 | 2020-01-10 | 黎剑猛 | Monitoring upgrading method and system |
CN110856014A (en) * | 2019-11-05 | 2020-02-28 | 北京奇艺世纪科技有限公司 | Moving image generation method, moving image generation device, electronic device, and storage medium |
CN110881141A (en) * | 2019-11-19 | 2020-03-13 | 浙江大华技术股份有限公司 | Video display method and device, storage medium and electronic device |
CN111010599A (en) * | 2019-12-18 | 2020-04-14 | 浙江大华技术股份有限公司 | Method and device for processing multi-scene video stream and computer equipment |
CN111048071A (en) * | 2019-11-11 | 2020-04-21 | 北京海益同展信息科技有限公司 | Voice data processing method and device, computer equipment and storage medium |
CN111209436A (en) * | 2020-01-10 | 2020-05-29 | 上海摩象网络科技有限公司 | Method and device for shooting material mark and electronic equipment |
CN111274441A (en) * | 2020-02-26 | 2020-06-12 | 赛特斯信息科技股份有限公司 | System and method for realizing maritime video data screening processing based on deep learning and target detection |
CN111416950A (en) * | 2020-03-26 | 2020-07-14 | 腾讯科技(深圳)有限公司 | Video processing method and device, storage medium and electronic equipment |
CN111445499A (en) * | 2020-03-25 | 2020-07-24 | 北京百度网讯科技有限公司 | Method and device for identifying target information |
CN111552837A (en) * | 2020-05-08 | 2020-08-18 | 深圳市英威诺科技有限公司 | Animal video tag automatic generation method based on deep learning, terminal and medium |
CN111901536A (en) * | 2020-08-04 | 2020-11-06 | 携程计算机技术(上海)有限公司 | Video editing method, system, device and storage medium based on scene recognition |
CN111930730A (en) * | 2020-07-28 | 2020-11-13 | 薛杨杨 | Data analysis method based on artificial intelligence and big data and block chain service platform |
CN112137591A (en) * | 2020-10-12 | 2020-12-29 | 平安科技(深圳)有限公司 | Target object position detection method, device, equipment and medium based on video stream |
CN112258513A (en) * | 2020-10-23 | 2021-01-22 | 岭东核电有限公司 | Nuclear power test video segmentation method and device, computer equipment and storage medium |
CN112380922A (en) * | 2020-10-23 | 2021-02-19 | 岭东核电有限公司 | Method and device for determining compound video frame, computer equipment and storage medium |
CN112954456A (en) * | 2021-03-29 | 2021-06-11 | 深圳康佳电子科技有限公司 | Video data processing method, terminal and computer readable storage medium |
CN113095194A (en) * | 2021-04-02 | 2021-07-09 | 北京车和家信息技术有限公司 | Image classification method and device, storage medium and electronic equipment |
CN113139428A (en) * | 2021-03-16 | 2021-07-20 | 西安天和防务技术股份有限公司 | Target identification method, edge device, frontier defense monitoring system and readable storage medium |
CN113329261A (en) * | 2021-08-02 | 2021-08-31 | 北京达佳互联信息技术有限公司 | Video processing method and device |
CN113496188A (en) * | 2020-04-08 | 2021-10-12 | 四零四科技股份有限公司 | Apparatus and method for processing video content analysis |
WO2021232978A1 (en) * | 2020-05-18 | 2021-11-25 | Oppo广东移动通信有限公司 | Video processing method and apparatus, electronic device and computer readable medium |
CN113949827A (en) * | 2021-09-30 | 2022-01-18 | 安徽尚趣玩网络科技有限公司 | Video content fusion method and device |
CN115171241A (en) * | 2022-06-30 | 2022-10-11 | 南京领行科技股份有限公司 | Video frame positioning method and device, electronic equipment and storage medium |
CN115334271A (en) * | 2022-08-10 | 2022-11-11 | 平安科技(深圳)有限公司 | High frame rate video generation method and device, electronic equipment and storage medium |
CN116680438A (en) * | 2023-05-13 | 2023-09-01 | 全景智联(武汉)科技有限公司 | Video concentration method, system, storage medium and electronic equipment |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103810195A (en) * | 2012-11-09 | 2014-05-21 | 中国电信股份有限公司 | Index generating method and system |
CN109213895A (en) * | 2017-07-05 | 2019-01-15 | 合网络技术(北京)有限公司 | A kind of generation method and device of video frequency abstract |
-
2019
- 2019-04-11 CN CN201910288788.3A patent/CN110147722A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103810195A (en) * | 2012-11-09 | 2014-05-21 | 中国电信股份有限公司 | Index generating method and system |
CN109213895A (en) * | 2017-07-05 | 2019-01-15 | 合网络技术(北京)有限公司 | A kind of generation method and device of video frequency abstract |
Non-Patent Citations (1)
Title |
---|
杨磊等: "《网络视频监控技术》", vol. 1, 30 June 2017, 中国传媒大学出版社, pages: 198 - 201 * |
Cited By (40)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110569392A (en) * | 2019-08-28 | 2019-12-13 | 深圳市天视通电子科技有限公司 | multi-video processing system and method |
CN110569392B (en) * | 2019-08-28 | 2023-01-10 | 深圳市天视通技术有限公司 | Multi-video processing system and method |
CN110659581A (en) * | 2019-08-29 | 2020-01-07 | 腾讯科技(深圳)有限公司 | Image processing method, device, equipment and storage medium |
CN110659581B (en) * | 2019-08-29 | 2024-02-20 | 腾讯科技(深圳)有限公司 | Image processing method, device, equipment and storage medium |
CN110674781B (en) * | 2019-10-07 | 2023-01-10 | 福建新宇电子有限公司 | Monitoring upgrading method and system |
CN110674781A (en) * | 2019-10-07 | 2020-01-10 | 黎剑猛 | Monitoring upgrading method and system |
CN110856014A (en) * | 2019-11-05 | 2020-02-28 | 北京奇艺世纪科技有限公司 | Moving image generation method, moving image generation device, electronic device, and storage medium |
CN111048071A (en) * | 2019-11-11 | 2020-04-21 | 北京海益同展信息科技有限公司 | Voice data processing method and device, computer equipment and storage medium |
CN110881141A (en) * | 2019-11-19 | 2020-03-13 | 浙江大华技术股份有限公司 | Video display method and device, storage medium and electronic device |
CN111010599A (en) * | 2019-12-18 | 2020-04-14 | 浙江大华技术股份有限公司 | Method and device for processing multi-scene video stream and computer equipment |
CN111209436A (en) * | 2020-01-10 | 2020-05-29 | 上海摩象网络科技有限公司 | Method and device for shooting material mark and electronic equipment |
CN111274441A (en) * | 2020-02-26 | 2020-06-12 | 赛特斯信息科技股份有限公司 | System and method for realizing maritime video data screening processing based on deep learning and target detection |
CN111445499A (en) * | 2020-03-25 | 2020-07-24 | 北京百度网讯科技有限公司 | Method and device for identifying target information |
CN111445499B (en) * | 2020-03-25 | 2023-07-18 | 北京百度网讯科技有限公司 | Method and device for identifying target information |
CN111416950A (en) * | 2020-03-26 | 2020-07-14 | 腾讯科技(深圳)有限公司 | Video processing method and device, storage medium and electronic equipment |
CN111416950B (en) * | 2020-03-26 | 2023-11-28 | 腾讯科技(深圳)有限公司 | Video processing method and device, storage medium and electronic equipment |
CN113496188B (en) * | 2020-04-08 | 2024-04-02 | 四零四科技股份有限公司 | Apparatus and method for processing video content analysis |
CN113496188A (en) * | 2020-04-08 | 2021-10-12 | 四零四科技股份有限公司 | Apparatus and method for processing video content analysis |
CN111552837A (en) * | 2020-05-08 | 2020-08-18 | 深圳市英威诺科技有限公司 | Animal video tag automatic generation method based on deep learning, terminal and medium |
WO2021232978A1 (en) * | 2020-05-18 | 2021-11-25 | Oppo广东移动通信有限公司 | Video processing method and apparatus, electronic device and computer readable medium |
CN111930730A (en) * | 2020-07-28 | 2020-11-13 | 薛杨杨 | Data analysis method based on artificial intelligence and big data and block chain service platform |
CN111901536A (en) * | 2020-08-04 | 2020-11-06 | 携程计算机技术(上海)有限公司 | Video editing method, system, device and storage medium based on scene recognition |
CN112137591A (en) * | 2020-10-12 | 2020-12-29 | 平安科技(深圳)有限公司 | Target object position detection method, device, equipment and medium based on video stream |
WO2021189911A1 (en) * | 2020-10-12 | 2021-09-30 | 平安科技(深圳)有限公司 | Target object position detection method and apparatus based on video stream, and device and medium |
CN112380922B (en) * | 2020-10-23 | 2024-03-22 | 岭东核电有限公司 | Method, device, computer equipment and storage medium for determining multiple video frames |
CN112380922A (en) * | 2020-10-23 | 2021-02-19 | 岭东核电有限公司 | Method and device for determining compound video frame, computer equipment and storage medium |
CN112258513A (en) * | 2020-10-23 | 2021-01-22 | 岭东核电有限公司 | Nuclear power test video segmentation method and device, computer equipment and storage medium |
CN113139428A (en) * | 2021-03-16 | 2021-07-20 | 西安天和防务技术股份有限公司 | Target identification method, edge device, frontier defense monitoring system and readable storage medium |
CN112954456A (en) * | 2021-03-29 | 2021-06-11 | 深圳康佳电子科技有限公司 | Video data processing method, terminal and computer readable storage medium |
CN113095194A (en) * | 2021-04-02 | 2021-07-09 | 北京车和家信息技术有限公司 | Image classification method and device, storage medium and electronic equipment |
CN113329261A (en) * | 2021-08-02 | 2021-08-31 | 北京达佳互联信息技术有限公司 | Video processing method and device |
CN113329261B (en) * | 2021-08-02 | 2021-12-07 | 北京达佳互联信息技术有限公司 | Video processing method and device |
CN113949827B (en) * | 2021-09-30 | 2023-04-07 | 安徽尚趣玩网络科技有限公司 | Video content fusion method and device |
CN113949827A (en) * | 2021-09-30 | 2022-01-18 | 安徽尚趣玩网络科技有限公司 | Video content fusion method and device |
CN115171241B (en) * | 2022-06-30 | 2024-02-06 | 南京领行科技股份有限公司 | Video frame positioning method and device, electronic equipment and storage medium |
CN115171241A (en) * | 2022-06-30 | 2022-10-11 | 南京领行科技股份有限公司 | Video frame positioning method and device, electronic equipment and storage medium |
CN115334271A (en) * | 2022-08-10 | 2022-11-11 | 平安科技(深圳)有限公司 | High frame rate video generation method and device, electronic equipment and storage medium |
CN115334271B (en) * | 2022-08-10 | 2024-05-07 | 平安科技(深圳)有限公司 | High-frame-rate video generation method and device, electronic equipment and storage medium |
CN116680438A (en) * | 2023-05-13 | 2023-09-01 | 全景智联(武汉)科技有限公司 | Video concentration method, system, storage medium and electronic equipment |
CN116680438B (en) * | 2023-05-13 | 2024-02-27 | 全景智联(武汉)科技有限公司 | Video concentration method, system, storage medium and electronic equipment |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110147722A (en) | A kind of method for processing video frequency, video process apparatus and terminal device | |
CN108427708B (en) | Data processing method, data processing apparatus, storage medium, and electronic apparatus | |
US10453165B1 (en) | Computer vision machine learning model execution service | |
CN110751224A (en) | Training method of video classification model, video classification method, device and equipment | |
CN111931809A (en) | Data processing method and device, storage medium and electronic equipment | |
CN112434178B (en) | Image classification method, device, electronic equipment and storage medium | |
CN105260487A (en) | Picture managing method and device | |
CN108334895A (en) | Sorting technique, device, storage medium and the electronic device of target data | |
CN109740530A (en) | Extracting method, device, equipment and the computer readable storage medium of video-frequency band | |
CN112732949A (en) | Service data labeling method and device, computer equipment and storage medium | |
CN112861894A (en) | Data stream classification method, device and system | |
CN113641797A (en) | Data processing method, device, equipment, storage medium and computer program product | |
CN110019849A (en) | A kind of video concern moment search method and device based on attention mechanism | |
CN109598289B (en) | Cross-platform data processing method, device, equipment and readable storage medium | |
CN112925899B (en) | Ordering model establishment method, case clue recommendation method, device and medium | |
CN114462582A (en) | Data processing method, device and equipment based on convolutional neural network model | |
CN112200862B (en) | Training method of target detection model, target detection method and device | |
CN116545740B (en) | Threat behavior analysis method and server based on big data | |
CN111368060A (en) | Self-learning method, device and system for conversation robot, electronic equipment and medium | |
CN112948251B (en) | Automatic software testing method and device | |
CN110071845A (en) | The method and device that a kind of pair of unknown applications are classified | |
CN108287817A (en) | A kind of information processing method and equipment | |
CN109840535B (en) | Method and device for realizing terrain classification | |
CN107992501A (en) | Social network information recognition methods, processing method and processing device | |
WO2021178643A1 (en) | An encoding device and method for utility-driven video compression |
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 |