CN108921136A - Video marker method and device, storage medium, terminal - Google Patents
Video marker method and device, storage medium, terminal Download PDFInfo
- Publication number
- CN108921136A CN108921136A CN201810863800.4A CN201810863800A CN108921136A CN 108921136 A CN108921136 A CN 108921136A CN 201810863800 A CN201810863800 A CN 201810863800A CN 108921136 A CN108921136 A CN 108921136A
- Authority
- CN
- China
- Prior art keywords
- video
- recognition result
- default
- feature
- marked
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/174—Facial expression recognition
-
- 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
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
- G06V20/42—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items of sport video content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Computational Linguistics (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Image Analysis (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
A kind of video marker method and device, storage medium, terminal, video marker method include：Obtain video file to be processed；Feature identification is carried out to the video content of the video file, to obtain recognition result；Judge whether the recognition result includes default feature；When the recognition result includes the default feature, the video content comprising the default feature is marked.The efficiency and accuracy of video screening can be improved in technical solution of the present invention.
Description
Technical field
The present invention relates to technical field of video processing more particularly to a kind of video marker method and devices, storage medium, end
End.
Background technique
In the prior art, user after obtaining the video recorded needs that video content is checked and screened frame by frame, with
The video content for being unsatisfactory for custom condition is rejected, guarantees the quality of video content.
But time-consuming and laborious, efficiency is lower is manually checked frame by frame to the screening of video content needs；Meanwhile the selection result
With certain artificial subjectivity, screening effect is poor.
Summary of the invention
Present invention solves the technical problem that being how to improve the efficiency and accuracy of video screening.
In order to solve the above technical problems, the embodiment of the present invention provides a kind of video marker method, video marker method includes：
Obtain video file to be processed；Feature identification is carried out to the video content of the video file, to obtain recognition result；Judgement
Whether the recognition result includes default feature；When the recognition result includes the default feature, to comprising described default
The video content of feature is marked.
Optionally, the described pair of video content comprising the default feature be marked including：In the recognition result pair
The position that axis is played locating for the video content answered is marked.
Optionally, packet is marked in the position that axis is played locating for the corresponding video content of the recognition result
It includes：The initial position of broadcasting axis and end position locating for the video content are marked.
Optionally, the default feature includes default text, and the video content to the video file identifies
Including：To in the video file voice or text identify that the recognition result includes text；It is described in the identification
As a result include the default feature when, to the video content comprising the default feature be marked including：It is tied in the identification
When text in fruit includes the default text, the video content comprising the default text is marked.
Optionally, the default feature includes deliberate action, and the video content to the video file identifies
Including：Humanoid in the video file is identified, the recognition result includes humanoid movement；It is described to be tied in the identification
Fruit include the default feature when, to the video content comprising the default feature be marked including：In the recognition result
In humanoid movement it is consistent with the deliberate action when, the video content comprising the humanoid movement is marked.
Optionally, the default feature includes presetting expression or default face, in the video to the video file
Appearance carries out identification：Face in the video file is identified, the recognition result includes face；It is described described
Recognition result include the default feature when, to the video content comprising the default feature be marked including：In the knowledge
The expression of the face in the face recognition result consistent or described with the default face in other result and the preset table
When feelings are consistent, the video content comprising the face is marked.
Optionally, the default feature includes predeterminated frequency, and the video content to the video file identifies
Including：Track in the video file is identified, the recognition result includes sound frequency；It is described to be tied in the identification
Fruit include the default feature when, to the video content comprising the default feature be marked including：In the recognition result
In sound frequency it is consistent with the predeterminated frequency when, the video content comprising the predeterminated frequency is marked.
In order to solve the above technical problems, the embodiment of the invention also discloses a kind of video marker device, video marker device
Including：Video file obtains module, suitable for obtaining video file to be processed；Feature recognition module is suitable for the video text
The video content of part carries out feature identification, to obtain recognition result；Judgment module, suitable for judge the recognition result whether include
Default feature；Mark module is suitable for when the recognition result includes the default feature, to the view comprising the default feature
Frequency content is marked.
Optionally, the mark module includes：Axis marking unit is played, is suitable in the corresponding video of the recognition result
The position for holding locating broadcasting axis is marked.
Optionally, initial position and the end for playing axis marking unit to axis is played locating for the video content
Position is marked.
Optionally, the default feature includes default text, and the feature recognition module includes：Word recognition unit is fitted
In in the video file voice or text identify that the recognition result includes text；The mark module includes：
When first marking unit suitable for the text in the recognition result includes the default text, to including the default text
Video content be marked.
Optionally, the default feature includes deliberate action, and the feature recognition module includes：Action recognition unit is fitted
It is identified in humanoid in the video file, the recognition result includes humanoid movement；The mark module includes：The
Two marking units, when consistent with the deliberate action suitable for the humanoid movement in the recognition result, to including the people
The video content of shape movement is marked.
Optionally, the default feature includes default expression or default face, the feature recognition module include：Face is known
Other unit, suitable for identifying to the face in the video file, the recognition result includes face；The mark module packet
It includes：Third marking unit, suitable for the identification knot consistent or described with the default face of the face in the recognition result
When the expression of face in fruit is consistent with the default expression, the video content comprising the face is marked.
Optionally, the default feature includes predeterminated frequency, and the feature recognition module includes：Track recognition unit is fitted
It is identified in the track in the video file, the recognition result includes sound frequency；The mark module includes：The
Four marking units, when consistent with the predeterminated frequency suitable for the sound frequency in the recognition result, to comprising described pre-
If the video content of frequency is marked.
The embodiment of the invention also discloses a kind of storage mediums, are stored thereon with computer instruction, the computer instruction
The step of video marker method is executed when operation.
The embodiment of the invention also discloses a kind of terminal, including memory and processor, being stored on the memory can
The computer instruction run on the processor, the processor execute the video marker when running the computer instruction
The step of method.
Compared with prior art, the technical solution of the embodiment of the present invention has the advantages that：
Technical solution of the present invention obtains video file to be processed；Feature is carried out to the video content of the video file
Identification, to obtain recognition result；Judge whether the recognition result includes default feature；It include described pre- in the recognition result
If when feature, the video content comprising the default feature is marked.Technical solution of the present invention passes through to video to be processed
Feature identification is carried out, and video content is marked by recognition result and the inclusion relation determination of default feature, so that packet
Video content containing default feature is screened out, is further processed for user；It avoids and artificially looks into the prior art
Efficiency brought by seeing and accuracy problem improve efficiency and accuracy to video screening.
Further, the position that axis is played locating for the corresponding video content of the recognition result is marked.This hair
Bright technical solution is marked video content by the broadcasting shaft position in video file, is conducive to user's dragging broadcasting axis and looks into
The video content for seeing label, improve labeled video content checks convenience.
Further, the default feature includes default text or the default feature includes deliberate action, Huo Zhesuo
State that default feature includes deliberate action or the default feature includes predeterminated frequency.Technical solution of the present invention is more by being arranged
The default feature of kind, and feature identification correspondingly is carried out to video content in different ways, it realizes and video content is carried out
The rich and diversity of label, expands the application range of video marker method.
Detailed description of the invention
Fig. 1 is a kind of flow chart of video marker method of the embodiment of the present invention；
Fig. 2 is a kind of specific embodiment flow chart of video marker method of the embodiment of the present invention；
Fig. 3 is the specific embodiment flow chart of another kind video marker method of the embodiment of the present invention；
Fig. 4 is the specific embodiment flow chart of another video marker method of the embodiment of the present invention；
Fig. 5 is the specific embodiment flow chart of another video marker method of the embodiment of the present invention；
Fig. 6 is a kind of structural schematic diagram of video marker device of the embodiment of the present invention；
Fig. 7 is a kind of structural schematic diagram of the specific embodiment of video marker device of the embodiment of the present invention.
Specific embodiment
As described in the background art, the screening needs of video content are manually checked frame by frame in the prior art, it is time-consuming and laborious,
Efficiency is lower；Meanwhile the selection result has certain artificial subjectivity, screening effect is poor.
Technical solution of the present invention passes through recognition result and default feature by carrying out feature identification to video to be processed
Inclusion relation determination video content is marked so that comprising preset feature video content be screened out, for user into
The further processing of row；Efficiency brought by artificially checking in the prior art and accuracy problem are avoided, is improved to video
The efficiency and accuracy of screening.
To make the above purposes, features and advantages of the invention more obvious and understandable, with reference to the accompanying drawing to the present invention
Specific embodiment be described in detail.
Fig. 1 is a kind of flow chart of video marker method of the embodiment of the present invention.
Video marker method shown in Fig. 1 may comprise steps of：
Step S101：Obtain video file to be processed；
Step S102：Feature identification is carried out to the video content of the video file, to obtain recognition result；
Step S103：Judge whether the recognition result includes default feature；
Step S104：When the recognition result includes the default feature, in the video comprising the default feature
Appearance is marked.
In specific implementation, video file to be processed, which can be, to be prerecorded, and is also possible to enroll in real time.Example
Such as, in the case where quality of instruction monitors scene, the instructional video enrolled in instructional space can be obtained in real time.
For the video prerecorded, can be pre-stored in database.Step S101 can be adjusted from the database
Take video file to be processed.
In the specific implementation of step S102, carrying out feature identification to video content can refer in extraction video content
Feature, recognition result may include the feature of video content.Specifically, when video content includes face, it can be in video
Face in appearance is identified；When video content includes humanoid, humanoid in video content can be identified；Video content
When including voice, the voice in video content can be identified.
It is understood that video content may include any enforceable feature, and can be to any in video content
Enforceable feature carries out feature identification, and the embodiment of the present invention is without limitation.
After obtaining the recognition result of video content, in the specific implementation of step S103 and step S104, it can be determined that
Whether recognition result includes default feature.If recognition result includes default feature, to the video comprising the default feature
Content is marked, namely the corresponding video content of recognition result is marked.
Specifically, default feature can be it is pre-set.Specifically, default feature can be deliberate action, preset
Face, default text etc..Under different application scenarios, different default features can be set.
The embodiment of the present invention passes through the packet of recognition result and default feature by carrying out feature identification to video to be processed
Video content is marked in the determination containing relationship, so that being screened out comprising the video content for presetting feature, carries out for user
Further processing；Efficiency brought by artificially checking in the prior art and accuracy problem are avoided, improves and video is sieved
The efficiency and accuracy of choosing.
In a specific embodiment of the invention, about the specific mark mode to video content, it can be in the identification
As a result the position that axis is played locating for corresponding video content is marked.
Specifically, playing axis has time attribute.The position that axis is played locating for video content is determined for regarding
The time location of frequency content plays the position that axis is played locating for axis to the video content by dragging, can play described
Video content is checked for user.The broadcasting axis is referred to as time shaft.
Specifically, it is marked can be in the position for playing axis and be marked using the shapes such as arrow or point.More specifically
Ground, the color of mark shape can be any enforceable color such as red, yellow.
The embodiment of the present invention is marked video content by the broadcasting shaft position in video file, and video may be implemented
The intuitive of content-label；Be conducive to user and drag the video content that broadcasting axis checks label, improves in labeled video
That holds checks convenience.
It should be noted that any other enforceable mode can also be to the mark mode of video content, such as
Default video frame etc. is inserted into video content comprising the default feature, the embodiment of the present invention is without limitation.
Further, the initial position of broadcasting axis and end position locating for the video content are marked.
Specifically, the video content comprising the default feature may include multiple continuous video frames.In this feelings
Under condition, when the video content is marked, it can exist to video content in the initial position and video content for playing axis
The end position for playing axis is marked.In other words, video content can correspond to the video content in the initial position for playing axis
Start frame, video content can correspond to the end frame of the video content in the end position for playing axis.
In turn, user plays axis to labeled initial position and end position by dragging, can check labeled
Video content, the convenience that the intuitive and user for realizing label are checked.
The present invention is a kind of about in the specific embodiment of step S103 and step S104, referring to figure 2., video marker side
Method may comprise steps of：
Step S201：To in the video file voice or text identify that the recognition result includes text；
Step S202：When text in the recognition result includes the default text, to including the default text
Video content be marked.
In specific implementation, if in video file including voice, speech recognition is carried out to the voice in video file；People
Include subtitle in work video file, then Text region is carried out to the subtitle in video file.If in video file including voice
And letter, then one of which can be identified, both of which can also be identified, and two kinds of recognition results progress are whole
Conjunction and duplicate removal, obtain final recognition result.
It is equal that the recognition result for identifying and being identified to the subtitle in video file is carried out to the voice in video file
Including text.
Default text can be preset by user.It, can be in video file to be processed by the way that default text is arranged
Determine the video content comprising default text.So as to which the video content is marked.
In specific application scenarios, it is dirty word that default text, which can be set, then packet can be determined through the above way
Video content containing dirty word, and be marked.User can be quickly found after getting the video file after label and include
The video content of dirty word, so as to delete the video content, or using the video content as evidence to video text
People in part is monitored.
It is understood that speech recognition and Text region can be realized using any enforceable technology or algorithm,
The embodiment of the present invention is without limitation.
It is of the invention another about in the specific embodiment of step S103 and step S104, referring to figure 3., video marker
Method may comprise steps of：
Step S301：Humanoid in the video file is identified, the recognition result includes humanoid movement；
Step S302：When humanoid movement in the recognition result is consistent with the deliberate action, to comprising described
The video content of humanoid movement is marked.
In specific implementation, if video file include it is humanoid, to video file carry out Human detection.It is wrapped in recognition result
It includes humanoid in the video file.Further, humanoid to indicate to act, then identify it is humanoid after, can determine view
Movement in frequency file.
Deliberate action can be preset by user.It, can be in video file to be processed by the way that deliberate action is arranged
Determine the video content comprising deliberate action.So as to which the video content is marked.
In specific application scenarios, in teaching monitoring scene, it is to raise one's hand, push and shove that deliberate action, which can be set,.So
The video content comprising raising one's hand or pushing and shoving can be determined through the above way, and is marked.User is after getting label
After video file, it can be quickly found the video content comprising raising one's hand or pushing and shoving, so as to include in the video raised one's hand
It takes in and makees splendid contents, will be deleted comprising the video content pushed and shoved, or push and shove the user of movement to execution in video file
It is monitored.
It is understood that Human detection can realize that the present invention is implemented using any enforceable technology or algorithm
Example is without limitation.
The present invention another about in the specific embodiment of step S103 and step S104, referring to figure 4., video marker
Method may comprise steps of：
Step S401：Face in the video file is identified, the recognition result includes face；
Step S402：The face recognition result consistent or described with the default face in the recognition result
In face expression it is consistent with the default expression when, the video content comprising the face is marked.
It may include face in video file in specific implementation.It, can be with by being identified to the face in video file
The identity of face is obtained, or can determine the expression of face.
Default expression or default face can be configured in advance by user.By the way that default face is arranged, can help to use
Family navigates to the video content comprising default face in video file, namely user is helped to look for people.By the way that default expression is arranged,
The video content comprising default expression can be determined in video file.
In a specific application scenarios, default face can be the face to be looked for of user, such as Missing Persons, criminal
Deng video file can be market monitor video, road monitor video etc..Since monitor video is usually long, such as usually
It is the monitor video in three days, one week or one month, therefore packet can be quickly determined in monitor video through the above way
Video content containing default face, the problem of avoiding low efficiency brought by artificial checking monitoring video.
In another specific application scenarios, default expression can be laugh, and video file can be instructional video.It is logical
Aforesaid way is crossed, can determine that comprising human face expression be the video content laughed, so that it is determined that the student played on classroom；Or
Person is the quality of instruction that the video content laughed can be used for evaluating the video file comprising human face expression.
It is understood that default expression is also possible to any other enforceable expression, such as smile, is angry, is sad,
Yawn etc., the embodiment of the present invention is without limitation.
The present invention another about in the specific embodiment of step S103 and step S104, referring to figure 5., video marker
Method may comprise steps of：
Step S501：Track in the video file is identified, the recognition result includes sound frequency；
Step S502：When sound frequency in the recognition result is consistent with the predeterminated frequency, to comprising described
The video content of predeterminated frequency is marked.
In the present embodiment, when video file includes track, the track in video file can be identified.And pass through by
The frequency of track in video file is compared with predeterminated frequency, can determine the video content comprising predeterminated frequency, go forward side by side
Line flag.
Predeterminated frequency, which can be, to be predefined and is arranged.
In specific application scenarios, video file can be instructional video.Predeterminated frequency can be teaching person, such as always
The track frequency of teacher.By determining the video content comprising predeterminated frequency, the teaching duration of teaching person can be determined.Alternatively,
Video content not comprising predeterminated frequency can be marked, to determine the duration that do not teach, and for evaluating and testing quality of instruction.
It should be noted that step shown in Fig. 2 to Fig. 5 is that step S101 and step S102 shown in Fig. 1 are held later
Capable；Specific embodiment about step S101 and step S102 please refers to previous embodiment, and details are not described herein again.
In a preferred embodiment of the invention, it can also include the following steps after step S104 shown in FIG. 1：It will label
The video content of completion is sent to default terminal device.
In specific implementation, default terminal device can be the equipment that relevant person in charge uses, and pass through the video after marking
Content is sent to associated user, and associated user can be made to be monitored using the video content, or for promotional display etc..
Please refer to Fig. 6, video marker device 60 may include video file obtain module 601, feature recognition module 602,
Judgment module 603 and mark module 604.
Wherein, video file obtains module 601 and is suitable for obtaining video file to be processed；Feature recognition module 602 is suitable for
Feature identification is carried out to the video content of the video file, to obtain recognition result；Judgment module 603 is suitable for judging the knowledge
Whether other result includes default feature；Mark module 604 be suitable for the recognition result include the default feature when, to comprising
The video content of the default feature is marked.
The embodiment of the present invention passes through the packet of recognition result and default feature by carrying out feature identification to video to be processed
Video content is marked in the determination containing relationship, so that being screened out comprising the video content for presetting feature, carries out for user
Further processing；Efficiency brought by artificially checking in the prior art and accuracy problem are avoided, improves and video is sieved
The efficiency and accuracy of choosing.
In a specific embodiment of the invention, mark module 604 may include playing axis marking unit, be suitable for described
The position that axis is played locating for the corresponding video content of recognition result is marked.
Further, the axis marking unit that plays is to initial position and the knot for playing axis locating for the video content
Beam position is marked.
Fig. 7 is please referred to, feature recognition module 602 and mark module 604 can be respectively provided with multiple specific units, multiple tools
Body unit can be applied in combination, and can be used for different application scenarios.
Specifically, the feature recognition module 602 shown in Fig. 6 may include word recognition unit 6021, action recognition list
Member 6022, face identification unit 6023 and/or track recognition unit 6024.Mark module 604 may include the first marking unit
6041, the second marking unit 6042, third marking unit 6043 and/or the 4th marking unit 6044.
In a specific embodiment of the invention, feature recognition module 602 may include word recognition unit 6021, be suitable for
To in the video file voice or text identify that the recognition result includes text.Mark module 604 may include
When first marking unit 6041 suitable for the text in the recognition result includes the default text, to comprising described default
The video content of text is marked.
In another embodiment of the present invention, the feature recognition module 602 may include action recognition unit
6022, suitable for identifying to humanoid in the video file, the recognition result includes humanoid movement.Mark module 604
It may include the second marking unit 6042, it is consistent with the deliberate action suitable for the humanoid movement in the recognition result
When, the video content comprising the humanoid movement is marked.
In another specific embodiment of the invention, the feature recognition module 602 may include face identification unit
6023, suitable for identifying to the face in the video file, the recognition result includes face.Mark module 604 can be with
Including third marking unit 6043, suitable in the recognition result face and the default face it is consistent or described
When the expression of face in recognition result is consistent with the default expression, the video content comprising the face is marked
Note.
In another specific embodiment of the invention, the feature recognition module 602 may include track recognition unit
6024, suitable for identifying to the track in the video file, the recognition result includes sound frequency.Mark module 604
It may include the 4th marking unit 6044, it is consistent with the predeterminated frequency suitable for the sound frequency in the recognition result
When, the video content comprising the predeterminated frequency is marked.
It is understood that multiple units shown in Fig. 7 are also possible to any other enforceable combination, the present invention is real
It is without limitation to apply example.For example, word recognition unit 6021 and the first marking unit 6041 and action recognition unit 6022
It is applied in combination with the second marking unit 6042；Face identification unit 6023 and third marking unit 6043 and track recognition unit
6024 and the 4th marking unit 6044 be applied in combination.
Working principle, more contents of working method about the video marker device 60, are referred to Fig. 1 to Fig. 5
In associated description, which is not described herein again.
The embodiment of the invention also discloses a kind of storage mediums, are stored thereon with computer instruction, the computer instruction
The step of video marker method shown in Fig. 1 to Fig. 5 can be executed when operation.The storage medium may include ROM, RAM,
Disk or CD etc..The storage medium can also include non-volatility memorizer (non-volatile) or non-transient
(non-transitory) memory etc..
The embodiment of the invention also discloses a kind of terminal, the terminal may include memory and processor, the storage
The computer instruction that can be run on the processor is stored on device.The processor can be with when running the computer instruction
1 is executed to shown in Fig. 5 the step of video marker method.The terminal includes but is not limited to mobile phone, computer, tablet computer
Equal terminal devices.
Although present disclosure is as above, present invention is not limited to this.Anyone skilled in the art are not departing from this
It in the spirit and scope of invention, can make various changes or modifications, therefore protection scope of the present invention should be with claim institute
Subject to the range of restriction.
Claims (16)
1. a kind of video marker method, which is characterized in that including：
Obtain video file to be processed；
Feature identification is carried out to the video content of the video file, to obtain recognition result；
Judge whether the recognition result includes default feature；
When the recognition result includes the default feature, the video content comprising the default feature is marked.
2. video marker method according to claim 1, which is characterized in that the described pair of video comprising the default feature
Content be marked including：
The position that axis is played locating for the corresponding video content of the recognition result is marked.
3. video marker method according to claim 2, which is characterized in that described in the corresponding video of the recognition result
Locating for content play axis position be marked including：
The initial position of broadcasting axis and end position locating for the video content are marked.
4. video marker method according to claim 1, which is characterized in that the default feature includes default text, institute
State to the video content of the video file carry out identification include：
To in the video file voice or text identify that the recognition result includes text；
It is described the recognition result include the default feature when, the video content comprising the default feature is marked
Including：
When text in the recognition result includes the default text, the video content comprising the default text is carried out
Label.
5. video marker method according to claim 1, which is characterized in that the default feature includes deliberate action, institute
State to the video content of the video file carry out identification include：
Humanoid in the video file is identified, the recognition result includes humanoid movement；
It is described the recognition result include the default feature when, the video content comprising the default feature is marked
Including：
When humanoid movement in the recognition result is consistent with the deliberate action, to the video comprising the humanoid movement
Content is marked.
6. video marker method according to claim 1, which is characterized in that the default feature includes default expression or pre-
If face, the video content to the video file carries out identification and includes：
Face in the video file is identified, the recognition result includes face；
It is described the recognition result include the default feature when, the video content comprising the default feature is marked
Including：
The expression of face in the recognition result consistent or described with the default face of the face in the recognition result
When consistent with the default expression, the video content comprising the face is marked.
7. video marker method according to claim 1, which is characterized in that the default feature includes predeterminated frequency, institute
State to the video content of the video file carry out identification include：
Track in the video file is identified, the recognition result includes sound frequency；
It is described the recognition result include the default feature when, the video content comprising the default feature is marked
Including：
When sound frequency in the recognition result is consistent with the predeterminated frequency, to the video comprising the predeterminated frequency
Content is marked.
8. a kind of video marker device, which is characterized in that including：
Video file obtains module, suitable for obtaining video file to be processed；
Feature recognition module carries out feature identification suitable for the video content to the video file, to obtain recognition result；
Judgment module, suitable for judging whether the recognition result includes default feature；
Mark module is suitable for when the recognition result includes the default feature, in the video comprising the default feature
Appearance is marked.
9. video marker device according to claim 8, which is characterized in that the mark module includes：
Axis marking unit is played, the position suitable for playing axis locating for the corresponding video content of the recognition result is marked
Note.
10. video marker device according to claim 9, which is characterized in that the broadcasting axis marking unit is to the view
The initial position of axis is played locating for frequency content and end position is marked.
11. video marker device according to claim 8, which is characterized in that the default feature includes default text, institute
Stating feature recognition module includes：
Word recognition unit, suitable for in the video file voice or text identify, the recognition result include text
Word；
The mark module includes：
When first marking unit suitable for the text in the recognition result includes the default text, to comprising described default
The video content of text is marked.
12. video marker device according to claim 8, which is characterized in that the default feature includes deliberate action, institute
Stating feature recognition module includes：
Action recognition unit, suitable for identifying to humanoid in the video file, the recognition result includes humanoid movement；
The mark module includes：
Second marking unit, when consistent with the deliberate action suitable for the humanoid movement in the recognition result, to comprising
The video content of the humanoid movement is marked.
13. video marker device according to claim 8, which is characterized in that the default feature include default expression or
Default face, the feature recognition module include：
Face identification unit, suitable for identifying to the face in the video file, the recognition result includes face；
The mark module includes：
Third marking unit, suitable for the identification consistent or described with the default face of the face in the recognition result
As a result when the expression of the face in is consistent with the default expression, the video content comprising the face is marked.
14. video marker device according to claim 8, which is characterized in that the default feature includes predeterminated frequency, institute
Stating feature recognition module includes：
Track recognition unit, suitable for identifying to the track in the video file, the recognition result includes sound frequency；
The mark module includes：
4th marking unit, when consistent with the predeterminated frequency suitable for the sound frequency in the recognition result, to comprising
The video content of the predeterminated frequency is marked.
15. a kind of storage medium, is stored thereon with computer instruction, which is characterized in that the computer instruction executes when running
Described in any one of claims 1 to 7 the step of video marker method.
16. a kind of terminal, including memory and processor, the meter that can be run on the processor is stored on the memory
Calculation machine instruction, which is characterized in that perform claim requires any one of 1 to 7 institute when the processor runs the computer instruction
The step of stating video marker method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810863800.4A CN108921136A (en) | 2018-08-01 | 2018-08-01 | Video marker method and device, storage medium, terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810863800.4A CN108921136A (en) | 2018-08-01 | 2018-08-01 | Video marker method and device, storage medium, terminal |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108921136A true CN108921136A (en) | 2018-11-30 |
Family
ID=64394217
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810863800.4A Pending CN108921136A (en) | 2018-08-01 | 2018-08-01 | Video marker method and device, storage medium, terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108921136A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110598695A (en) * | 2019-08-15 | 2019-12-20 | 北京搜狗科技发展有限公司 | Medicine identification method and device for medicine identification |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104952451A (en) * | 2015-06-08 | 2015-09-30 | 广东欧珀移动通信有限公司 | Sound recording processing method and sound recording processing device |
| CN105611413A (en) * | 2015-12-24 | 2016-05-25 | 小米科技有限责任公司 | Method and device for adding video clip class markers |
| CN106878759A (en) * | 2017-02-13 | 2017-06-20 | 广东小天才科技有限公司 | Many seat in the plane Shot change method and system in a kind of live teaching broadcast |
| CN107529068A (en) * | 2016-06-21 | 2017-12-29 | 北京新岸线网络技术有限公司 | Video content discrimination method and system |
| US20180114071A1 (en) * | 2016-10-21 | 2018-04-26 | Nokia Technologies Oy | Method for analysing media content |
| CN108271069A (en) * | 2017-12-11 | 2018-07-10 | 北京奇艺世纪科技有限公司 | The segment filter method and device of a kind of video frequency program |
-
2018
- 2018-08-01 CN CN201810863800.4A patent/CN108921136A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104952451A (en) * | 2015-06-08 | 2015-09-30 | 广东欧珀移动通信有限公司 | Sound recording processing method and sound recording processing device |
| CN105611413A (en) * | 2015-12-24 | 2016-05-25 | 小米科技有限责任公司 | Method and device for adding video clip class markers |
| CN107529068A (en) * | 2016-06-21 | 2017-12-29 | 北京新岸线网络技术有限公司 | Video content discrimination method and system |
| US20180114071A1 (en) * | 2016-10-21 | 2018-04-26 | Nokia Technologies Oy | Method for analysing media content |
| CN106878759A (en) * | 2017-02-13 | 2017-06-20 | 广东小天才科技有限公司 | Many seat in the plane Shot change method and system in a kind of live teaching broadcast |
| CN108271069A (en) * | 2017-12-11 | 2018-07-10 | 北京奇艺世纪科技有限公司 | The segment filter method and device of a kind of video frequency program |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110598695A (en) * | 2019-08-15 | 2019-12-20 | 北京搜狗科技发展有限公司 | Medicine identification method and device for medicine identification |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106128212B (en) | Learning calligraphy system and method based on augmented reality | |
| US20160267808A1 (en) | Augmented Reality | |
| KR20210144658A (en) | Video processing method and apparatus, electronic device and storage medium | |
| CN108898115B (en) | Data processing method, storage medium and electronic device | |
| CN106296690A (en) | The method for evaluating quality of picture material and device | |
| WO2020214316A1 (en) | Artificial intelligence-based generation of event evaluation report | |
| Lakshmiprabha et al. | [Poster] An augmented and virtual reality system for training autistic children | |
| CN108766113B (en) | Method and device for monitoring classroom performance of students | |
| CN114298497A (en) | Evaluation method and device for classroom teaching quality of teacher | |
| CN110225394A (en) | The processing method and processing device of recorded broadcast video knowledge point | |
| CN115205764B (en) | Online learning concentration monitoring method, system and medium based on machine vision | |
| US20160240006A1 (en) | Evaluation of augmented reality skins | |
| CN112949461A (en) | Learning state analysis method and device and electronic equipment | |
| Nyaga et al. | Sign language gesture recognition through computer vision | |
| CN111144321A (en) | Concentration degree detection method, device, equipment and storage medium | |
| CN111353439A (en) | Method, device, system and equipment for analyzing teaching behaviors | |
| CN109240931A (en) | Problem feedback information treating method and apparatus | |
| CN108921136A (en) | Video marker method and device, storage medium, terminal | |
| Quintana et al. | Object and gesture recognition to assist children with autism during the discrimination training | |
| CN114299777A (en) | Virtual reality industrial simulation training system | |
| CN107364256B (en) | Intelligent pen, intelligent learning system and working method thereof | |
| US11335084B2 (en) | Image object anomaly detection | |
| CN112306601A (en) | Application interaction method and device, electronic equipment and storage medium | |
| CN110019862B (en) | Courseware recommendation method, device, equipment and storage medium | |
| CN111081090B (en) | Information output method and learning device in point-to-read scene |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20181130 |
|
| WD01 | Invention patent application deemed withdrawn after publication |