CN108764067A - Video intercepting method, terminal, equipment and readable medium based on recognition of face - Google Patents
Video intercepting method, terminal, equipment and readable medium based on recognition of face Download PDFInfo
- Publication number
- CN108764067A CN108764067A CN201810431169.0A CN201810431169A CN108764067A CN 108764067 A CN108764067 A CN 108764067A CN 201810431169 A CN201810431169 A CN 201810431169A CN 108764067 A CN108764067 A CN 108764067A
- Authority
- CN
- China
- Prior art keywords
- image
- cluster
- feature vector
- face
- video
- 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
- 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
-
- 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
- G06V20/47—Detecting features for summarising 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
- G06V40/10—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
- G06V40/161—Detection; Localisation; Normalisation
-
- 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
- G06V40/168—Feature extraction; Face representation
-
- 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
- G06V40/172—Classification, e.g. identification
Abstract
Video intercepting method, terminal, equipment and readable medium provided in an embodiment of the present invention based on recognition of face, its terminal includes interception unit, detection unit, cluster cell and selection unit, it is based on face recognition technology, improve the accuracy and specific aim of video intercepting, and artificial screening is not needed, manpower and materials are saved.When it is applied to student's video intercepting, students and their parents can be effectively distinguished, improve the usage experience of Online class video recording.
Description
Technical field
The present invention relates to video intercepting technical fields.More particularly, to the video intercepting method based on recognition of face, end
End, equipment and readable medium.
Background technology
In video processing, generally require to intercept the image of some personage in video, or need interception one
Cover image of a most representative image as the video, but existing video intercepting generally requires artificial screening, it is time-consuming to take
Power, Shortcomings.
Invention content
At least one of in order to solve the above-mentioned technical problem, the first aspect of the invention provides a kind of based on face
The video intercepting method of identification, including：
Intercept the image of different frame in video；
Detect those images successively, and it includes multiple faces or the not no image of face to reject；
Clustering processing is carried out to the image for only including a face, obtains the image cluster of different personages；
Cover image of the image as the image cluster in one of image cluster is chosen, or chooses one of them
The multiple images of image cluster, and only include the corresponding personage's video of the image cluster by the chronological formation of multiple image.
In a preferred embodiment, the image progress clustering processing to only including a face includes：
Based on the feature of face in image, the feature vector of extraction reflection face characteristic；
According to the feature vector of face in image, multiple initial characteristics vectors are generated at random as barycenter, each barycenter pair
Answer a feature vector cluster；
Feature vector is calculated successively at a distance from all barycenter, and is classified to it where nearest barycenter
Feature vector cluster；
The averaged feature vector of each feature vector cluster is sought, and compared with the barycenter in this feature vector cluster, if differing
It causes, then using the averaged feature vector sought as new barycenter, feature vector is reclassified.
In another preferred embodiment, described image cluster includes student's image cluster, described to choose one of image
An image in cluster is configured as choosing the highest image of clarity, selection personage eyes most as the cover image of the personage
Big image or selection personage smiling face picture is as personage's cover image.
In yet another preferred embodiment, described image cluster includes student's image cluster, described to choose one of image
The multiple images of cluster, and be configured as the chronological video that formed of multiple image to choose all figures in student's image cluster
Picture, by the excellent time video of all chronological formation students of image.
Second aspect of the present invention provides a kind of video intercepting terminal based on recognition of face, including：
Interception unit intercepts the image of different frame in video；
Detection unit, detects those images successively, and it includes multiple faces or the not no image of face to reject；
Cluster cell carries out clustering processing to the image for only including a face, obtains the image cluster of different personages；
Selection unit is based on selection condition, chooses cover of the image as the personage in one of image cluster
Image, or the multiple images of one of image cluster are chosen, and only include the figure by the chronological formation of multiple image
As the corresponding personage's video of cluster.
In a preferred embodiment, the cluster cell includes：
Extraction unit, based on the feature of face in image, the feature vector of extraction reflection face characteristic；
Initial cell generates multiple initial characteristics vectors as barycenter, often at random according to the feature vector of face in image
A barycenter corresponds to a feature vector cluster；
Taxon, successively calculate feature vector with all barycenter at a distance from, and be classified to its distance recently
Feature vector cluster where barycenter；And
Comparing unit, seeks the averaged feature vector of each feature vector cluster, and with the barycenter ratio in this feature vector cluster
Compared with if inconsistent, using the averaged feature vector sought as new barycenter, feature vector is reclassified.
In another preferred embodiment, described eigenvector, the dimension of initial characteristics vector and averaged feature vector
Degree is 128 dimensions.
In yet another preferred embodiment, described image cluster includes student's image cluster, and the selection unit is configured as
The highest image of clarity, the image of selection personage's largest eyes or selection personage smiling face picture are chosen as personage's surface plot
Picture.
In yet another preferred embodiment, described image cluster includes student's image cluster, described to choose one of image
The multiple images of cluster, and be configured as the chronological video that formed of multiple image to choose all figures in student's image cluster
Picture, by the excellent time video of all chronological formation students of image.
Third aspect present invention provides a kind of computer equipment, including memory, processor and storage are on a memory
And the computer program that can be run on a processor,
The processor realizes method as described above when executing described program.
Fourth aspect present invention provides a kind of computer-readable medium, is stored thereon with computer program, which is located
Reason device realizes method as described above when executing.
Beneficial effects of the present invention are as follows：
Video intercepting method, terminal, equipment and readable medium provided in an embodiment of the present invention based on recognition of face, base
In face recognition technology, the accuracy and specific aim of video intercepting are improved, and does not need artificial screening, saves manpower object
Power.When it is applied to student's video intercepting, students and their parents can be effectively distinguished, improve Online class video recording uses body
It tests.
Description of the drawings
Specific embodiments of the present invention will be described in further detail below in conjunction with the accompanying drawings.
Fig. 1 shows the video intercepting method flow diagram according to an embodiment of the invention based on recognition of face.
Fig. 2 shows the idiographic flow schematic diagrams of S103 steps shown in Fig. 1.
Fig. 3 shows the video intercepting terminal structure schematic diagram according to an embodiment of the invention based on recognition of face.
Fig. 4 shows the concrete structure schematic diagram of cluster cell in Fig. 3.
Fig. 5 shows the structural representation of the computer equipment of the terminal device suitable for being used for realizing the embodiment of the present invention
Figure.
Specific implementation mode
In order to illustrate more clearly of the present invention, the present invention is done further with reference to preferred embodiments and drawings
It is bright.Similar component is indicated with identical reference numeral in attached drawing.It will be appreciated by those skilled in the art that institute is specific below
The content of description is illustrative and be not restrictive, and should not be limited the scope of the invention with this.
The various sectional views that embodiment is disclosed according to the present invention are shown in the accompanying drawings.These figures are not drawn to scale
, wherein for the purpose of clear expression, some details are magnified, and some details may be omitted.It is shown in the drawings
Various regions, the shape of layer and the relative size between them, position relationship are merely exemplary, in practice may be due to system
It makes tolerance or technology restriction and is deviated, and those skilled in the art may be additionally designed as required with not similar shape
Shape, size, the regions/layers of relative position.
In video processing, generally require to intercept the image of some personage in video, or need interception one
Cover image of a most representative image as the video, the prior art still have many deficiencies.For example, student's
In the video of classroom, the picture camera lens of parents of student is usually contained, when handling video, for example, selecting student
Cover photo, can generally intercept certain frames of video, but in these frames, may include parent, alternatively, student and family
It grows together, is needed at this time there is the frame of parent's face to exclude.
Fig. 1 shows the video intercepting method flow diagram according to an embodiment of the invention based on recognition of face, such as Fig. 1
Shown, this method comprises the following steps：
S101:Intercept the image of different frame in video.
In certain embodiments, video is the video of student's Online class, is attended class by recording student, from
And it supervises student and carries out on-line study.Include only student most of the time in the video, parent may also appear in video sometimes
In.
S102:Detect those images successively, and it includes multiple faces or the not no image of face to reject.
With in the embodiment that student attends class, truncated picture includes three kinds of situations：Only include student's (face), only
Including parent's (face) and including students and their parents (two faces), which can remove including students and their parents'
Situation.
S103:Clustering processing is carried out to the image for only including a face, obtains the image cluster of different personages.
Clustering processing can be understood as multiple and different images according to certain rule classification, specifically, incorporated by reference to Fig. 2,
The clustering processing includes：
S131:Based on the feature of face in image, the feature vector of extraction reflection face characteristic.
Since the image pixel size in video is consistent, it is assumed that it is 100*100, since each pixel has RGB tri- logical
Road, each pixel access then share 30,000 bytes of matrix meter of 3 100*100 as defeated by the byte representation of 0-255 ranges
Enter data, therefore, those input datas can be translated into characteristic value, this feature value by approximate function well known in the art
Generally use feature vector indicates that vectorial dimension is numerical value number therein, it is clear that is known when dimension is larger in a certain range
Other precision higher.
Preferably, feature vector is 128 dimensional vectors, such as indicates feature vector with faceID, i.e. first image can be with
It is expressed as faceID1, second graphical representation is faceID2, follow-up and so on.
S132:According to the feature vector of face in image, multiple initial characteristics vectors are generated at random as barycenter, Mei Gezhi
The heart corresponds to a feature vector cluster.
Initial characteristics vector can be set according to pre-stored standard picture, only include a people in the standard picture
Face can also determine an approximate initial characteristics vector according to the lumped values of the feature vector of face, in the present embodiment,
Initial characteristics vector can be calculated as faceID0.
S133:Feature vector is calculated successively at a distance from all barycenter, and is classified to it apart from nearest barycenter
The feature vector cluster at place.
Since feature vector is formed by multiple numerical value, the numerical value of each corresponding barycenter of numerical value has numerical difference,
It can be calculated every by calculating the distance of mean difference (such as standard deviation) this feature vector and barycenter of those numerical differences
A feature vector is classified to the spy with its distance nearest (calculated difference is minimum) at a distance from each barycenter, by feature vector
In the vectorial cluster of sign.
S134:The averaged feature vector of each feature vector cluster is sought, and compared with the barycenter in this feature vector cluster, if
It is inconsistent, then using the averaged feature vector sought as new barycenter, feature vector is reclassified.
Specifically, can be that simple adduction is average, or well known to a person skilled in the art the sides such as weighted average
Formula seeks averaged feature vector, and the embodiment of the present invention is without being limited thereto, if the averaged feature vector is completely the same (i.e. each with barycenter
The numerical value of a dimension is consistent), then confirm that the classification of this feature vector meets mark standard, if inconsistent, reclassifies, and should
Averaged feature vector repeats the step as new barycenter.
S104:Cover image of the image as the image cluster in one of image cluster is chosen, or chooses it
In an image cluster multiple images, and by multiple image it is chronological formation only regarded including the corresponding personage of the image cluster
Frequently.
Again by taking student attends class video recording online as an example, after carrying out S101-S103, student's class and parent can be sorted out
Class, found out from the image cluster of student's class a largest eyes (such as can be expressed as with feature vector the numerical value of certain dimensions compared with
Greatly), personage smiling face or the highest image of clarity are classified, or choose student as cover image in order to which video preserves
Multiple images in class image cluster, such as can be all images of student's class, the time point of all images is recorded, and should
All images form new video according to Time alignment, which is the excellent time video of student.
Video intercepting method provided in an embodiment of the present invention based on recognition of face is based on recognition of face, improves video
The accuracy and specific aim of interception, and artificial screening is not needed, save manpower and materials.When it is applied to student's video intercepting
When, students and their parents can be effectively distinguished, the usage experience of Online class video recording is improved.
Fig. 3 shows that another embodiment of the present invention provides the video intercepting terminal based on recognition of face, which can be
Mobile phone terminal, computer end etc. can be the front end used for user, or for the rear end of the making of producer, this hair
Bright embodiment is without being limited thereto.
In the embodiment, terminal specifically includes：
Interception unit 101 intercepts the image of different frame in video.
Specifically, video can be the video of student's Online class, is attended class by recording student, learned to supervise
It is raw to carry out on-line study.Include only student most of the time in the video, parent may also appear in video sometimes.
Detection unit 102, detects those images successively, and it includes multiple faces or the not no image of face to reject.
With in the embodiment that student attends class, truncated picture includes three kinds of situations：Only include student's (face), only
Including parent's (face) and including students and their parents (two faces), which can remove including students and their parents'
Situation.
Cluster cell 103 carries out clustering processing to the image for only including a face, obtains the image cluster of different personages.
Wherein, Fig. 4 shows the concrete structure unit of cluster cell, specifically includes：
Extraction unit 131, based on the feature of face in image, the feature vector of extraction reflection face characteristic.
Preferably, feature vector is 128 dimensional vectors.
Initial cell 132 generates multiple initial characteristics vectors as matter at random according to the feature vector of face in image
The heart, each barycenter correspond to a feature vector cluster.
Initial characteristics vector can be set according to pre-stored standard picture, only include a people in the standard picture
Face can also determine an approximate initial characteristics vector according to the lumped values of the feature vector of face.
Taxon 133 calculates feature vector at a distance from all barycenter, and is classified to its distance recently successively
Barycenter where feature vector cluster.
Since feature vector is formed by multiple numerical value, the numerical value of each corresponding barycenter of numerical value has numerical difference,
It can be calculated every by calculating the distance of mean difference (such as standard deviation) this feature vector and barycenter of those numerical differences
A feature vector is classified to the spy with its distance nearest (calculated difference is minimum) at a distance from each barycenter, by feature vector
In the vectorial cluster of sign.
Comparing unit 134, seeks the averaged feature vector of each feature vector cluster, and with the barycenter in this feature vector cluster
Compare, if inconsistent, using the averaged feature vector sought as new barycenter, feature vector is reclassified until average spy
Sign vector is consistent with barycenter.
By above-mentioned iterative manner, initial barycenter can be chosen in a certain range, without being accurate to a tool
Body vector, to which the selection of barycenter does not interfere with the accuracy of identification.
Selection unit 104 is based on selection condition, chooses envelope of the image as the personage in one of image cluster
Face image, or the multiple images of one of image cluster are chosen, and only including by the chronological formation of multiple image should
The corresponding personage's video of image cluster.
By taking student attends class video recording online as an example, student's class and parent's class can be sorted out by taxon, from student's class
Image cluster in find out a largest eyes (such as can be expressed as with feature vector certain dimensions numerical value it is larger), personage smiling face
Or the highest image of clarity is classified, or choose in student's class image cluster as cover image in order to which video preserves
Multiple images, such as can be all images of student's class, record the time point of all images, and by all images according to
Time alignment forms new video, which is the excellent time video of student.
Terminal provided in this embodiment intercepts video based on recognition of face, improves the accurate of video intercepting
Degree and specific aim, and artificial screening is not needed, save manpower and materials.When it is applied to student's video intercepting, Ke Yiyou
Effect distinguishes students and their parents, improves the usage experience of Online class video recording.
Further, some specific embodiments of the invention provide a kind of computer equipment, including memory, processor with
And the computer program that can be run on a memory and on a processor is stored, the processor is realized such as when executing described program
The upper method executed by terminal.
Below with reference to Fig. 5, it illustrates the computer equipments 500 suitable for the terminal device for realizing the embodiment of the present application
Structural schematic diagram.
As shown in figure 5, computer equipment 500 includes central processing unit (CPU) 501, it can be read-only according to being stored in
Program in memory (ROM) 502 is loaded into random access storage device (RAM) from storage section 508) program in 503
And execute various work appropriate and processing.In RAM503, also it is stored with system 500 and operates required various program sum numbers
According to.CPU501, ROM502 and RAM503 are connected with each other by bus 504.Input/output (I/O) interface 505 is also connected to
Bus 504.
It is connected to I/O interfaces 505 with lower component：Importation 506 including keyboard, mouse etc.；It is penetrated including such as cathode
The output par, c 507 of spool (CRT), liquid crystal display (LCD) etc. and loud speaker etc.；Storage section 508 including hard disk etc.；
And including such as LAN card, the communications portion 509 of the network interface card of modem etc..Communications portion 509 via such as because
The network of spy's net executes communication process.Driver 510 is also according to needing to be connected to I/O interfaces 506.Detachable media 511, such as
Disk, CD, magneto-optic disk, semiconductor memory etc. are mounted on driver 510, as needed in order to be read from thereon
Computer program be mounted as needed such as storage section 508.
Particularly, according to an embodiment of the invention, it may be implemented as computer above with reference to the process of flow chart description
Software program.For example, the embodiment of the present invention includes a kind of computer program product comprising be tangibly embodied in machine readable
Computer program on medium, the computer program include the program code for method shown in execution flow chart.At this
In the embodiment of sample, which can be downloaded and installed by communications portion 509 from network, and/or from removable
Medium 511 is unloaded to be mounted.
Flow chart in attached drawing and block diagram, it is illustrated that according to the system of various embodiments of the invention, method and computer journey
The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation
A part for a part for one module, program segment, or code of table, the module, program segment, or code includes one or more
Executable instruction for implementing the specified logical function.It should also be noted that in some implementations as replacements, institute in box
The function of mark can also be sent in a different order than that indicated in the drawings.Such as two boxes succeedingly indicated are actually
It can be basically executed in parallel, they can also execute in the opposite order sometimes, this is depended on the functions involved.Also it to note
Meaning, the combination of each box in block diagram and or flow chart and the box in block diagram and or flow chart can be with holding
The dedicated hardware based system of functions or operations as defined in row is realized, or can use specialized hardware and computer instruction
Combination realize.
Obviously, the above embodiment of the present invention be only to clearly illustrate example of the present invention, and not be pair
The restriction of embodiments of the present invention may be used also on the basis of the above description for those of ordinary skill in the art
To make other variations or changes in different ways, all embodiments can not be exhaustive here, it is every to belong to this hair
Row of the obvious changes or variations that bright technical solution is extended out still in protection scope of the present invention.
Claims (11)
1. a kind of video intercepting method based on recognition of face, which is characterized in that including：
Intercept the image of different frame in video；
Detect those images successively, and it includes multiple faces or the not no image of face to reject；
Clustering processing is carried out to the image for only including a face, obtains the image cluster of different personages；
Cover image of the image as the image cluster in one of image cluster is chosen, or chooses one of image
The multiple images of cluster, and only include the corresponding personage's video of the image cluster by the chronological formation of multiple image.
2. method according to claim 1, which is characterized in that the image to only including a face carries out clustering processing
Including：
Based on the feature of face in image, the feature vector of extraction reflection face characteristic；
According to the feature vector of face in image, multiple initial characteristics vectors are generated at random as barycenter, each barycenter corresponds to one
A feature vector cluster；
Feature vector is calculated successively at a distance from all barycenter, and is classified to it apart from the feature where nearest barycenter
Vectorial cluster；
The averaged feature vector of each feature vector cluster is sought, and compared with the barycenter in this feature vector cluster, if inconsistent,
Using the averaged feature vector sought as new barycenter, feature vector is reclassified.
3. method according to claim 1, which is characterized in that described image cluster includes student's image cluster, and the selection is wherein
An image in one image cluster is configured as choosing the highest image of clarity, chooses people as the cover image of the personage
The image of object largest eyes chooses personage smiling face picture as personage's cover image.
4. method according to claim 1, which is characterized in that described image cluster includes student's image cluster, and the selection is wherein
The multiple images of one image cluster, and the chronological formation video of multiple image is configured as choosing in student's image cluster
All images, by the excellent time video of all chronological formation students of image.
5. a kind of video intercepting terminal based on recognition of face, which is characterized in that including：
Interception unit intercepts the image of different frame in video；
Detection unit, detects those images successively, and it includes multiple faces or the not no image of face to reject；
Cluster cell carries out clustering processing to the image for only including a face, obtains the image cluster of different personages；
Selection unit is based on selection condition, chooses cover image of the image as the personage in one of image cluster,
Or the multiple images of one of image cluster are chosen, and only include the image cluster pair by the chronological formation of multiple image
The personage's video answered.
6. terminal according to claim 5, which is characterized in that the cluster cell includes：
Extraction unit, based on the feature of face in image, the feature vector of extraction reflection face characteristic；
Initial cell generates multiple initial characteristics vectors as barycenter, Mei Gezhi at random according to the feature vector of face in image
The heart corresponds to a feature vector cluster；
Taxon calculates feature vector at a distance from all barycenter, and is classified to it apart from nearest barycenter successively
The feature vector cluster at place；And
Comparing unit seeks the averaged feature vector of each feature vector cluster, and compared with the barycenter in this feature vector cluster, if
It is inconsistent, then using the averaged feature vector sought as new barycenter, feature vector is reclassified.
7. terminal according to claim 5, which is characterized in that described eigenvector, initial characteristics vector and average characteristics
The dimension of vector is 128 dimensions.
8. terminal according to claim 5, which is characterized in that described image cluster includes student's image cluster, the selection unit
It is configured as choosing the highest image of clarity, the image of selection personage's largest eyes or selection personage's smiling face's picture as people
Object cover image.
9. terminal according to claim 5, which is characterized in that described image cluster includes student's image cluster, and the selection is wherein
The multiple images of one image cluster, and the chronological formation video of multiple image is configured as choosing in student's image cluster
All images, by the excellent time video of all chronological formation students of image.
10. a kind of computer equipment, including memory, processor and storage can be run on a memory and on a processor
Computer program, which is characterized in that
The processor is realized when executing described program such as any one of claim 1-4 the method.
11. a kind of computer-readable medium, is stored thereon with computer program, which is characterized in that
It is realized such as any one of claim 1-4 the method when the program is executed by processor.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810431169.0A CN108764067A (en) | 2018-05-08 | 2018-05-08 | Video intercepting method, terminal, equipment and readable medium based on recognition of face |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810431169.0A CN108764067A (en) | 2018-05-08 | 2018-05-08 | Video intercepting method, terminal, equipment and readable medium based on recognition of face |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108764067A true CN108764067A (en) | 2018-11-06 |
Family
ID=64010381
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810431169.0A Pending CN108764067A (en) | 2018-05-08 | 2018-05-08 | Video intercepting method, terminal, equipment and readable medium based on recognition of face |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108764067A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111460226A (en) * | 2020-04-01 | 2020-07-28 | 山东云缦智能科技有限公司 | Video character retrieval method and retrieval system based on deep learning |
| CN111901627A (en) * | 2020-05-28 | 2020-11-06 | 北京大米科技有限公司 | Video processing method and device, storage medium and electronic equipment |
| CN112149650A (en) * | 2020-11-25 | 2020-12-29 | 南京爱照飞打影像科技有限公司 | System capable of rapidly and automatically delivering and trading images |
| CN114071244A (en) * | 2021-11-10 | 2022-02-18 | 广州博冠信息科技有限公司 | Method and device for generating live cover, computer storage medium and electronic equipment |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105631408A (en) * | 2015-12-21 | 2016-06-01 | 小米科技有限责任公司 | Video-based face album processing method and processing device |
| CN105760472A (en) * | 2016-02-06 | 2016-07-13 | 中国农业大学 | Video retrieval method and system |
-
2018
- 2018-05-08 CN CN201810431169.0A patent/CN108764067A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105631408A (en) * | 2015-12-21 | 2016-06-01 | 小米科技有限责任公司 | Video-based face album processing method and processing device |
| CN105760472A (en) * | 2016-02-06 | 2016-07-13 | 中国农业大学 | Video retrieval method and system |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111460226A (en) * | 2020-04-01 | 2020-07-28 | 山东云缦智能科技有限公司 | Video character retrieval method and retrieval system based on deep learning |
| CN111901627A (en) * | 2020-05-28 | 2020-11-06 | 北京大米科技有限公司 | Video processing method and device, storage medium and electronic equipment |
| CN111901627B (en) * | 2020-05-28 | 2022-12-30 | 北京大米科技有限公司 | Video processing method and device, storage medium and electronic equipment |
| CN112149650A (en) * | 2020-11-25 | 2020-12-29 | 南京爱照飞打影像科技有限公司 | System capable of rapidly and automatically delivering and trading images |
| CN114071244A (en) * | 2021-11-10 | 2022-02-18 | 广州博冠信息科技有限公司 | Method and device for generating live cover, computer storage medium and electronic equipment |
| CN114071244B (en) * | 2021-11-10 | 2022-11-04 | 广州博冠信息科技有限公司 | Method and device for generating live cover, computer storage medium and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110544258B (en) | Image segmentation method and device, electronic equipment and storage medium | |
| WO2021057848A1 (en) | Network training method, image processing method, network, terminal device and medium | |
| CN107452010B (en) | Automatic cutout algorithm and device | |
| CN108764067A (en) | Video intercepting method, terminal, equipment and readable medium based on recognition of face | |
| CN105118048B (en) | The recognition methods of reproduction certificate picture and device | |
| EP2889367B1 (en) | Image processing device, program, image processing method, computer-readable medium, and image processing system | |
| Sarrafzadeh et al. | Selection of the best features for leukocytes classification in blood smear microscopic images | |
| US10169861B2 (en) | Image processing apparatus, non-transitory computer readable medium, and image processing method | |
| CN108846401A (en) | Commodity detect terminal, method, system and computer equipment, readable medium | |
| WO2015107722A1 (en) | Detection control device, program, detection system, storage medium and detection control method | |
| CN112348765A (en) | Data enhancement method and device, computer readable storage medium and terminal equipment | |
| CN110136198A (en) | Image processing method and its device, equipment and storage medium | |
| CN108769803A (en) | Recognition methods, method of cutting out, system, equipment with frame video and medium | |
| CN108769634A (en) | A kind of image processing method, image processing apparatus and terminal device | |
| CN111080595A (en) | Image processing method, image processing device, electronic equipment and computer readable medium | |
| CN110751218A (en) | Image classification method, image classification device and terminal equipment | |
| CN108764139A (en) | A kind of method for detecting human face, mobile terminal and computer readable storage medium | |
| CN111144215B (en) | Image processing method, device, electronic equipment and storage medium | |
| CN110348516B (en) | Data processing method, data processing device, storage medium and electronic equipment | |
| CN108932703B (en) | Picture processing method, picture processing device and terminal equipment | |
| CN113763370A (en) | Digital pathological image processing method and device, electronic equipment and storage medium | |
| US10146042B2 (en) | Image processing apparatus, storage medium, and image processing method | |
| CN111797922B (en) | Text image classification method and device | |
| CN114663418A (en) | Image processing method and device, storage medium and electronic equipment | |
| CN112434547B (en) | User identity auditing method and device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181106 |
|
| RJ01 | Rejection of invention patent application after publication |