CN106101771A - Method for processing video frequency, device and terminal - Google Patents
Method for processing video frequency, device and terminal Download PDFInfo
- Publication number
- CN106101771A CN106101771A CN201610483353.0A CN201610483353A CN106101771A CN 106101771 A CN106101771 A CN 106101771A CN 201610483353 A CN201610483353 A CN 201610483353A CN 106101771 A CN106101771 A CN 106101771A
- Authority
- CN
- China
- Prior art keywords
- head
- head portrait
- pending object
- video
- portrait
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/41—Structure of client; Structure of client peripherals
- H04N21/4104—Peripherals receiving signals from specially adapted client devices
- H04N21/4126—The peripheral being portable, e.g. PDAs or mobile phones
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/41—Structure of client; Structure of client peripherals
- H04N21/426—Internal components of the client ; Characteristics thereof
- H04N21/42653—Internal components of the client ; Characteristics thereof for processing graphics
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N21/00—Selective content distribution, e.g. interactive television or video on demand [VOD]
- H04N21/40—Client devices specifically adapted for the reception of or interaction with content, e.g. set-top-box [STB]; Operations thereof
- H04N21/43—Processing of content or additional data, e.g. demultiplexing additional data from a digital video stream; Elementary client operations, e.g. monitoring of home network or synchronising decoder's clock; Client middleware
- H04N21/431—Generation of visual interfaces for content selection or interaction; Content or additional data rendering
- H04N21/4318—Generation of visual interfaces for content selection or interaction; Content or additional data rendering by altering the content in the rendering process, e.g. blanking, blurring or masking an image region
Abstract
Embodiments providing a kind of method for processing video frequency, device and terminal, described method includes: according to the head portrait of object pending in described video, obtain the head pose of described pending object；The replacement head portrait that head pose is consistent with the head pose of described pending object is selected from head portrait storehouse；The head portrait of pending object described in two field picture by described video is replaced by described replacement head portrait.The head portrait of the pending object in two field pictures multiple in video can be replaced to the replacement head portrait that head pose is identical by the embodiment of the present invention automatically, allow users to only can be realized the purpose being replaced by the head portrait in multiple two field pictures of batch by little operation, and replace head portrait picture harmony degree in two field picture to improve after head portrait is replaced, the head portrait of pending object can be replaced with the replacement head portrait that head pose is identical, and then realize improving head portrait replacement efficiency, saving user time energy, the viewing quality of guarantee video pictures.
Description
Technical field
The present invention relates to field of computer technology, particularly relate to a kind of method for processing video frequency, device and terminal.
Background technology
Currently, the application of all kinds of short-sighted frequencies emerges in an endless stream, and some short Video Applications can record short-sighted frequency by photographic head, so
Afterwards according to the operation of user, the short-sighted frequency recorded being added various filter specially good effects, paster and background music etc., some short-sighted frequencies should
During can recording short-sighted frequency utilizing photographic head, at face, add specially good effect and stage property etc. according to the operation of user,
Also have some short Video Applications that user can be made according to text prompt lip-syncing performance etc. during recording.
But, these short Video Applications, adding specially good effect when, all must have user to participate in, and user needs manually will patch
Paper, specially good effect stage property etc. are positioned over relevant position, so, when operating each time, user be both needed to first to find paster to be placed and
Time location in whole short-sighted frequency and the picture positions in each frame picture such as specially good effect stage property, operate more complicated, numerous
Trivial, waste user time.
Summary of the invention
For overcoming problem present in correlation technique, the present invention provides a kind of method for processing video frequency, device and terminal.
First aspect according to embodiments of the present invention, it is provided that a kind of method for processing video frequency, including:
According to the head portrait of object pending in video, obtain the head pose of described pending object；
The replacement head portrait that head pose is consistent with the head pose of described pending object is selected from head portrait storehouse；
The head portrait of pending object described in two field picture by described video is replaced by described replacement head portrait.
Alternatively, described method also includes:
Obtain the head reference picture during different head attitude of pending object；
Judge in the two field picture of video, whether to there is the similarity with described head reference picture more than predetermined threshold value
Head portrait；
When there is the head portrait that the similarity with described head reference picture is more than predetermined threshold value, determine the frame of described video
Image includes the head portrait of described pending object.
Alternatively, described method also includes:
Obtain the head portrait of different objects in described video；
Judge whether to receive the selection operation of any one head portrait in described video；
When receiving described selection operation, the described head portrait selecting operation corresponding is defined as the head of pending object
Picture.
Alternatively, described method also includes:
Obtain the view of multiple angles of the head replacing object；
Utilize the replacement head portrait of the different head attitude replacing object described in the view generation of at least one angle；
The all replacement head portraits generated are stored in described head portrait storehouse.
Alternatively, the described head portrait according to object pending in described video, obtain the head appearance of described pending object
State, including:
Extract head feature and the feature locations of pending object in described two field picture；
Determine the described feature locations shift ratio relative to the contour edge of the head portrait of described pending object；
According to described shift ratio, determine the direction of visual lines of pending object described in described two field picture and shooting direction it
Between angle；
The head pose of described pending object is determined according to described angle.
Alternatively, described in the described two field picture by described video, the head portrait of pending object is replaced by described replacement head
Picture, including:
Obtain the head portrait of the described pending object band of position in described two field picture；
Delete the head portrait of described pending object in the described band of position；
Described replacement head portrait is added in the described band of position.
Alternatively, described in the described two field picture by described video, the head portrait of pending object is replaced by described replacement head
Picture, including:
Obtain the head portrait of the described pending object band of position in described two field picture；
By described replacement head portrait Overlapping display on the described band of position of described two field picture.
Second aspect according to embodiments of the present invention, it is provided that a kind of video process apparatus, including:
First acquisition module, for the head portrait according to object pending in video, obtains the head of described pending object
Attitude；
Select module, for selecting consistent the replacing of head pose of head pose and described pending object from head portrait storehouse
Change head portrait；
Changing module, the head portrait for pending object described in the two field picture by described video is replaced by described replacement head
Picture.
Alternatively, described device also includes:
Second acquisition module, head reference picture during for obtaining the different head attitude of pending object；
Search module, big with the similarity of described head reference picture for judging whether to exist in the two field picture of video
Head portrait in predetermined threshold value；
First determines module, for when there is the head portrait more than predetermined threshold value of the similarity with described head reference picture
Time, determine the head portrait including described pending object in the two field picture of described video.
Alternatively, described device also includes:
3rd acquisition module, for obtaining the head portrait of different objects in described video；
Judge module, operates the selection of any one head portrait in described video for judging whether to receive；
Second determines module, for when receiving described selection operation, is determined by the described head portrait selecting operation corresponding
Head portrait for pending object.
Alternatively, described device also includes:
4th acquisition module, for obtaining the view of multiple angles of the head replacing object；
Generation module, for utilizing replacing of the different head attitude replacing object described in the view generation of at least one angle
Change head portrait；
Memory module, for storing all replacement head portraits generated in described head portrait storehouse.
Alternatively, described acquisition module, including:
Extract submodule, for extracting head feature and the feature locations of pending object in described two field picture；
First determines submodule, for determining the described feature locations silhouette edge relative to the head portrait of described pending object
The shift ratio of edge；
Second determines submodule, for according to described shift ratio, determines pending object described in described two field picture
Angle between direction of visual lines and shooting direction；
3rd determines submodule, for determining the head pose of described pending object according to described angle.
Alternatively, described replacing module, including:
First obtains submodule, is used for the head portrait the obtaining described pending object band of position in described two field picture；
Delete submodule, for deleting the head portrait of the described pending object in the described band of position；
Add submodule, for being added in the described band of position by described replacement head portrait.
Alternatively, described replacing module, including:
Second obtains submodule, is used for the head portrait the obtaining described pending object band of position in described two field picture；
Display sub-module, is used for described replacement head portrait Overlapping display in the described band of position of described two field picture.
According to the third aspect of the invention we, it is provided that a kind of terminal, described terminal includes:
Processor；
For storing the memorizer of processor executable；
Wherein, described processor is configured to:
According to the head portrait of object pending in video, obtain the head pose of described pending object；
The replacement head portrait that head pose is consistent with the head pose of described pending object is selected from head portrait storehouse；
The head portrait of pending object described in two field picture by described video is replaced by described replacement head portrait.
The technical scheme that embodiments of the invention provide can include following beneficial effect:
The embodiment of the present invention, by the head portrait of pending object in the two field picture according to described video, obtains described pending
The head pose of object, selects the replacement head that head pose is consistent with the head pose of described pending object from head portrait storehouse
Picture, can be replaced by described replacement head portrait by the head portrait of the described pending object in described two field picture.
The method that the embodiment of the present invention provides, it is possible to automatically by the head of the pending object in two field pictures multiple in video
As replacing to the replacement head portrait that head pose is identical, allow users to only by little operation can realize in batches by multiple
The purpose that head portrait in two field picture is replaced, and in order to improve head portrait replace after replace head portrait picture in two field picture and
Humorous degree, can replace with the replacement head portrait that head pose is identical by the head portrait of pending object, and then realization improves head portrait and replaces
Change efficiency, save user time energy, the viewing quality of guarantee video pictures.
It should be appreciated that it is only exemplary and explanatory, not that above general description and details hereinafter describe
The present invention can be limited.
Accompanying drawing explanation
Accompanying drawing herein is merged in description and constitutes the part of this specification, it is shown that meet the enforcement of the present invention
Example, and for explaining the principle of the present invention together with description.
Fig. 1 is a kind of flow chart according to a kind of method for processing video frequency shown in an exemplary embodiment；
Fig. 2 is the another kind of flow chart according to a kind of method for processing video frequency shown in an exemplary embodiment；
Fig. 3 is the another kind of flow chart according to a kind of method for processing video frequency shown in an exemplary embodiment；
Fig. 4 is the another kind of flow chart according to a kind of method for processing video frequency shown in an exemplary embodiment；
Fig. 5 is the flow chart of step S101 in Fig. 1；
Fig. 6 is a kind of flow chart of step S103 in Fig. 1；
Fig. 7 is the another kind of flow chart of step S103 in Fig. 1；
Fig. 8 is the structure chart according to a kind of video process apparatus shown in an exemplary embodiment；
Fig. 9 is the block diagram according to a kind of terminal shown in an exemplary embodiment.
Detailed description of the invention
Here will illustrate exemplary embodiment in detail, its example represents in the accompanying drawings.Explained below relates to
During accompanying drawing, unless otherwise indicated, the same numbers in different accompanying drawings represents same or analogous key element.Following exemplary embodiment
Described in embodiment do not represent all embodiments consistent with the present invention.On the contrary, they are only with the most appended
The example of the apparatus and method that some aspects that described in detail in claims, the present invention are consistent.
As it is shown in figure 1, in one embodiment of the invention, it is provided that a kind of method for processing video frequency, described method include with
Lower step.
In step S101, according to the head portrait of object pending in video, obtain the head pose of described pending object.
In embodiments of the present invention, owing to a video there may be, multiple two field picture comprises pending object, institute
The head image of pending object in the multiple two field pictures in video is referred to, such as: work as Sun Wukong with the head portrait of pending object
During for pending object, the head image of Sun Wukong, head appearance in multiple two field pictures comprising Sun Wukong in Pilgrimage to the West video
State refers to the head of pending object attitude in the two field picture at its place, and attitude here can refer to the head of pending object
Portion is relative to the rotation attitude of camera lens, if head front is in the face of camera lens, left side of head are in the face of faced by camera lens, 45 degree, head right front
Camera lens, head rear are in the face of camera lens etc..
In this step, the two field picture comprising pending object can first be searched in all two field pictures of video, so
After in each two field picture, search the head portrait of pending object again, finally for each two field picture, extract therein pending
The head pose of object header.
In step s 102, from head portrait storehouse, select consistent the replacing of head pose of head pose and described pending object
Change head portrait.
In embodiments of the present invention, head portrait storehouse comprises the head portrait of the multiple head poses replacing object, the most heads
The head portrait of portion's attitude can be that the head portrait gathering the multiple angle of head replacing object obtains, it is also possible to is to gather several angles
Then the head portrait of degree generates according to several head portraits gathered, for example, it is possible to gather head front, left side of head face, head
The head portrait of four angles after right flank and head, then generates the head portrait of multiple angles, such as, generates according to these four head portraits
45 degree, head left front is in the face of multiple angles such as camera lenses faced by camera lens faced by camera lens, 35 degree, head right back and 45 degree, head right front
The head portrait of degree.
In this step, can be first according to the time order and function order of two field pictures comprising pending object all in video
First to the head pose of pending object in two field picture the earliest in time sequencing, then according to this head pose in head portrait storehouse
Find the replacement head portrait consistent with the head pose of this pending object, here unanimously refer to pending object with replace right
The head of elephant is identical, such as relative to the rotation attitude of camera lens: the head of the two is all head relative to the rotation attitude of camera lens
Front is in the face of camera lens, left side of head are in the face of camera lens faced by camera lens or 45 degree, head right back etc., identical finding attitude
After replacing head portrait, the replacement head portrait that the head portrait attitude pending with next two field picture is consistent can be continued to search for, until institute
The two field picture comprising pending object is had to be disposed.
In step s 103, the head portrait of the described pending object in the two field picture of described video is replaced by described replacement
Head portrait.
In this step, can according to two field pictures comprising pending object all in video time order and function order, by
The individual head portrait by the pending object in two field picture replaces with replacement head portrait, until all two field pictures comprising pending object are complete
Portion replaces complete.
The embodiment of the present invention, by the head portrait of pending object in the two field picture according to described video, obtains described pending
The head pose of object, selects the replacement head that head pose is consistent with the head pose of described pending object from head portrait storehouse
Picture, can be replaced by described replacement head portrait by the head portrait of the described pending object in described two field picture.
The method that the embodiment of the present invention provides, it is possible to automatically by the head of the pending object in two field pictures multiple in video
As replacing to the replacement head portrait that head pose is identical, allow users to only by little operation can realize in batches by multiple
The purpose that head portrait in two field picture is replaced, and in order to improve head portrait replace after replace head portrait picture in two field picture and
Humorous degree, can replace with the replacement head portrait that head pose is identical by the head portrait of pending object, and then realization improves head portrait and replaces
Change efficiency, save user time energy, the viewing quality of guarantee video pictures.
In actual applications, if all two field pictures in video all being obtained the head portrait of pending object, then due to certain
A little two field pictures do not comprise pending object, if the most still extracting the two field picture not comprising pending image, may result in
The wasting of resources, reduces head portrait and replaces efficiency, to this end, as in figure 2 it is shown, in another embodiment of the present invention, described method is also wrapped
Include step.
In step s 201, the head reference picture during different head attitude of pending object is obtained.
In embodiments of the present invention, head reference picture can refer to identify pending object head portrait with reference to figure
Picture, head reference picture here can be the head portrait of several different head attitudes of pending object, determines head reference
The mode of image can be directly several head portraits that user selects to be defined as head reference picture, it is also possible to is according to user
The head reference picture of multiple different head attitudes that several head portraits selected generate, it is also possible in user by selecting
When mode determines pending object, automatically from the head reference picture etc. of the different attitudes of the pending object of web search.
In this step, multiple head reference pictures of pending object can be obtained, and judge that multiple head is with reference to figure
Whether there is identical head reference picture in Xiang, if there is identical head reference picture, can only retain multiple identical
Head reference picture in one, delete remaining repeat head reference picture.
In step S202, it is judged that whether exist big with the similarity of described head reference picture in the two field picture of video
Head portrait in predetermined threshold value.
In this step, the head pose of multiple objects in two field picture can first be extracted, the most respectively according to these heads
Portion's attitude searches corresponding head reference picture, then by head reference identical with head pose respectively for the head portrait of multiple objects
Image compares, if the similarity of the head portrait of any one object and head reference picture is more than predetermined threshold value, then may determine that this
Two field picture exists the head portrait of pending object, in embodiments of the present invention, when relatively head portrait similarity, it is possible to use existing
The Similarity Measure algorithm etc. of image recognition in technology.
When there is the head portrait that the similarity with described head reference picture is more than predetermined threshold value, in step S203, really
The two field picture of fixed described video includes the head portrait of described pending object.
The embodiment of the present invention by obtain pending object different head attitude time head reference picture, it is judged that regarding
Whether the two field picture of frequency exists the head portrait more than predetermined threshold value of the similarity with described head reference picture, when existing with described
When the similarity of head reference picture is more than the head portrait of predetermined threshold value, it may be determined that the two field picture of described video includes described
The head portrait of pending object.
The method that the embodiment of the present invention provides, it is possible to automatically obtain and determine the frame figure comprising pending object in video
Picture, and then be easy to only the two field picture comprising pending object be processed, improve head portrait and replace efficiency.
In order to the pending object determined fast and accurately in video, as it is shown on figure 3, in the another enforcement of the present invention
In example, described method is further comprising the steps of.
In step S301, obtain the head portrait of different objects in described video.
In this step, can extract the head portrait of all different objects in video, the head portrait of each object can select spy
Determine angle, such as, select the head portrait etc. of head front.
In step s 302, it may be judged whether receive the selection operation of any one head portrait in described video.
In this step, it is possible to use display control shows all head portraits got, and display mode can be suspended in
Overlapping display on video playback interface, it is also possible to be in two field picture, highlights the head portrait of object, highlights and can refer to face
Color is prominent or brightness highlights etc., and whether the corresponding viewing area that then can detect these head portraits receives selection operation
Deng, selection operation can be with finger mouse single-click operation, finger touch operation or double mouse click operation etc..
When receiving described selection operation, in step S303, it is defined as treating by the described head portrait selecting operation corresponding
Process the head portrait of object.
In this step, when receiving selection operation in any one viewing area, can be corresponding by this viewing area
Head portrait be defined as the head portrait of pending object.
The embodiment of the present invention is by obtaining the head portrait of different objects in described video, it may be judged whether receive described video
In the selection operation of any one head portrait, when receive described select operation time, can be by the described head portrait selecting operation corresponding
It is defined as the head portrait of pending object.
The method that the embodiment of the present invention provides, it is possible to determine the pending object in video and pending fast and accurately
The head portrait of object, and then it is easy to head portrait replacement fast and accurately.
In actual applications, replacing the angled head portrait of object if gathered, will be the most loaded down with trivial details and time-consuming, to this end, such as
Shown in Fig. 4, in another embodiment of the present invention, described method is further comprising the steps of.
In step S401, obtain the view of multiple angles of the head replacing object.
In this step, can gather the view of predetermined quantity the angle of the head replacing object, such as, head is just
Face, left side of head face, right side of head face and head view below.
In step S402, utilize the replacing of different head attitude replacing object described in the view generation of at least one angle
Change head portrait.
In this step, the view gathered can be utilized image cropping, splice and the mode such as torsion processes, generate
Replace the replacement head portrait of object different head attitude, such as, when needing the replacement head portrait generating head front, can direct root
According to the view generation of the head front replacing object, when needing to generate 45 degree, head left front in the face of the replacement head portrait of camera lens,
Then can carry out image procossing generation according to the view in the view of head front and left side of head face.
In step S403, all replacement head portraits generated are stored in described head portrait storehouse.
In this step, the replacement head portrait of each generation can be respectively stored in head portrait storehouse, in storing process, for
It is easy to search, storage can be ranked up according to the head replacing object clockwise relative to the rotation attitude of camera lens,
Such as, according to head front, 10 degree, head right front, 20 degree, head right front ..., right side of head face ..., after head
Face ..., left side of head face ... carrying out store, in above sequence, the interval between each adjacent head pose is only for example
Illustrate, in being embodied as, can need to be set according to practical situation.
The embodiment of the present invention, by obtaining the view of multiple angles of the head replacing object, utilizes at least one angle
Replace the replacement head portrait of the different head attitude of object described in view generation, all replacement head portraits generated can be stored institute
State in head portrait storehouse.
The method that the embodiment of the present invention provides, it is possible to according to a small amount of view replacing object gathered, can generate
The replacement head portrait of multiple different head attitudes, simple to operate, it is simple to user uses, save user time energy.
In order to determine the head pose of pending object in each two field picture fast and accurately, as it is shown in figure 5, at this
In the another embodiment of invention, in described step S101, comprise the following steps.
In step S501, extract head feature and the feature of described head feature of pending object in described two field picture
Position.
In embodiments of the present invention, head feature can be that the face feature of head is (such as eye feature, nose feature, lip
Portion's feature, ear's feature, eyebrow feature), shape of face feature, hair line feature, hair feature etc., feature locations can be with head feature
Coordinate position, such as the coordinate position of the inner eye corner in eye feature, the coordinate position of the tail of the eye, the coordinate position of pupil,
Coordinate position of the lip corners of the mouth etc..
In this step, the head feature in two field picture can be identified by the way of image recognition, and by measuring picture
The mode of vegetarian refreshments determines the feature locations of each head feature.
In step S502, determine inclined relative to the contour edge of the head portrait of described pending object of described feature locations
Shifting ratio.
In embodiments of the present invention, contour edge can refer in this two field picture, the contour edge of head, such as front head
In the head portrait in portion, contour edge includes being positioned at top hair contour edge, middle part ear contour edge and the face mask of bottom
Edges etc., in the head portrait in left side of head face, are positioned at face mask edge and the right side of top hair contour edge, left portion and bottom
The hair contour edge etc. at middle part.
In this step, feature locations distance can be calculated and be positioned at the two of the contour edge being positioned at it on same level line
Ratio between the distance of individual intersection point, this ratio may act as shift ratio.
In step S503, according to described shift ratio, determine the sight line side of pending object described in described two field picture
Angle between shooting direction.
In embodiments of the present invention, the angle between direction of visual lines and shooting direction can refer to the intersection point at the two extended line
The angle at place, such as: when camera lens faced by the head portrait of pending object is front in a two field picture, can be by sight line and bat
Take the photograph the angle between direction and be defined as 180 degree, in order to distinguish right side and left side, then when head can be rotated to the right and shooting
The angle in direction is just, is negative with the angle of shooting direction, when before the right side of the head portrait of pending object when head rotates to the left
When aspect is to camera lens, the angle between sight line and shooting direction can be may determine that be 45 degree, when the head portrait of pending object
Left front in the face of camera lens time, the angle between sight line and shooting direction can be may determine that as-45 degree etc..
In step S504, determine the head pose of described pending object according to described angle.
In this step, can carry out converting calculating scheduling algorithm and process according to angle, finally determine the head of pending object
Portion's attitude.
The embodiment of the present invention is by extracting the head feature of pending object in described two field picture and described head feature
Feature locations, determines the described feature locations shift ratio relative to the contour edge of the head portrait of described pending object, according to
Described shift ratio, determines the angle between direction of visual lines and the shooting direction of pending object described in described two field picture, can
To determine the head pose of described pending object according to described angle.
The method that the embodiment of the present invention provides, it is possible to determine head pose automatically according to the head feature in head portrait, and
And determine the head pose of pending object in two field picture fast and accurately, simple to operate, Video processing efficiency is high.
As shown in Figure 6, in another embodiment of the present invention, described step S103, comprise the following steps.
In step s 601, the head portrait of the described pending object band of position in described two field picture is obtained.
In embodiments of the present invention, the band of position can be with finger picture band of position at place in two field picture.
In this step, the head portrait band of position in two field picture can be determined according to contouring head marginal area.
In step S602, delete the head portrait of described pending object in the described band of position.
In this step, it is possible to use the head of pending object in region, the delete position such as erasing rubber or crop tool
Picture.
In step S603, described replacement head portrait is added in the described band of position.
In this step, can add in the band of position in the way of the interpolation of bottom by replacing head portrait, say, that can
Replacement head portrait to be positioned at the bottom of two field picture, in this manner it is ensured that the contour edge replacing head portrait and the band of position added
Identical.
The embodiment of the present invention, by obtaining the head portrait band of position in described two field picture of described pending object, is being deleted
After the head portrait of the described pending object in the described band of position, described replacement head portrait can be added to the described band of position
In.
The method that the embodiment of the present invention provides, it is possible to show replacing head portrait in the band of position at head portrait place accurately
Illustrate.
As it is shown in fig. 7, in another embodiment of the present invention, described step S103, comprise the following steps.
In step s 701, the head portrait of the described pending object band of position in described two field picture is obtained.
In embodiments of the present invention, the band of position can be with finger picture band of position at place in two field picture.
In this step, the head portrait band of position in two field picture can be determined according to contouring head marginal area.
In step S702, by described replacement head portrait Overlapping display on the described band of position of described two field picture.
In this step, can will replace on the band of position shown in the broadcast interface of video that head portrait suspends,
In the playing process of video, can be along with the movement of the band of position, the mobile position replacing head portrait.
The embodiment of the present invention is by obtaining the head portrait band of position in described two field picture of described pending object, then
Can be by described replacement head portrait Overlapping display on the described band of position of described two field picture.
The method that the embodiment of the present invention provides, it is possible to show replacing head portrait in the band of position at head portrait place accurately
Illustrate.
As shown in Figure 8, in another embodiment of the present invention, it is provided that a kind of video process apparatus, including:
First acquisition module 801, for the head portrait according to object pending in video, obtains the head of described pending object
Portion's attitude；
Select module 802, for selecting head pose consistent with the head pose of described pending object from head portrait storehouse
Replacement head portrait；
Change module 803, for the head portrait of pending object described in two field picture by described video be replaced by described in replace
Change head portrait.
In another embodiment of the present invention, described device also includes:
Second acquisition module, head reference picture during for obtaining the different head attitude of pending object；
Search module, big with the similarity of described head reference picture for judging whether to exist in the two field picture of video
Head portrait in predetermined threshold value；
First determines module, for when there is the head portrait more than predetermined threshold value of the similarity with described head reference picture
Time, determine the head portrait including described pending object in the two field picture of described video.
In another embodiment of the present invention, described device also includes:
3rd acquisition module, for obtaining the head portrait of different objects in described video；
Judge module, operates the selection of any one head portrait in described video for judging whether to receive；
Second determines module, for when receiving described selection operation, is determined by the described head portrait selecting operation corresponding
Head portrait for pending object.
In another embodiment of the present invention, described device also includes:
4th acquisition module, for obtaining the view of multiple angles of the head replacing object；
Generation module, for utilizing replacing of the different head attitude replacing object described in the view generation of at least one angle
Change head portrait；
Memory module, for storing all replacement head portraits generated in described head portrait storehouse.
In another embodiment of the present invention, described acquisition module, including:
Extract submodule, for extracting head feature and the feature locations of pending object in described two field picture；
First determines submodule, for determining the described feature locations silhouette edge relative to the head portrait of described pending object
The shift ratio of edge；
Second determines submodule, for according to described shift ratio, determines pending object described in described two field picture
Angle between direction of visual lines and shooting direction；
3rd determines submodule, for determining the head pose of described pending object according to described angle.
In another embodiment of the present invention, described replacing module, including:
First obtains submodule, is used for the head portrait the obtaining described pending object band of position in described two field picture；
Delete submodule, for deleting the head portrait of the described pending object in the described band of position；
Add submodule, for being added in the described band of position by described replacement head portrait.
In another embodiment of the present invention, described replacing module, including:
Second obtains submodule, is used for the head portrait the obtaining described pending object band of position in described two field picture；
Display sub-module, is used for described replacement head portrait Overlapping display in the described band of position of described two field picture.
Fig. 9 is the block diagram according to a kind of application program erecting device shown in an exemplary embodiment.With reference to Fig. 9, this dress
Put and include:
Processor 21；
For storing the memorizer 22 of processor 21 executable instruction；
Wherein, described processor 21 is configured to:
According to the head portrait of object pending in video, obtain the head pose of described pending object；
The replacement head portrait that head pose is consistent with the head pose of described pending object is selected from head portrait storehouse；
The head portrait of pending object described in two field picture by described video is replaced by described replacement head portrait.
Those skilled in the art, after considering description and putting into practice invention disclosed herein, will readily occur to its of the present invention
Its embodiment.The application is intended to any modification, purposes or the adaptations of the present invention, these modification, purposes or
Person's adaptations is followed the general principle of the present invention and includes the undocumented common knowledge in the art of the present invention
Or conventional techniques means.Description and embodiments is considered only as exemplary, and true scope and spirit of the invention are by appended
Claim is pointed out.
It should be appreciated that the invention is not limited in precision architecture described above and illustrated in the accompanying drawings, and
And various modifications and changes can carried out without departing from the scope.The scope of the present invention is only limited by appended claim.
Claims (15)
1. a method for processing video frequency, it is characterised in that including:
According to the head portrait of object pending in video, obtain the head pose of described pending object；
The replacement head portrait that head pose is consistent with the head pose of described pending object is selected from head portrait storehouse；
The head portrait of pending object described in two field picture by described video is replaced by described replacement head portrait.
Method the most according to claim 1, it is characterised in that described method also includes:
Obtain the head reference picture during different head attitude of pending object；
Judge the head portrait that whether there is the similarity with described head reference picture in the two field picture of video more than predetermined threshold value；
When there is the head portrait that the similarity with described head reference picture is more than predetermined threshold value, determine the two field picture of described video
In include the head portrait of described pending object.
Method the most according to claim 1, it is characterised in that described method also includes:
Obtain the head portrait of different objects in described video；
Judge whether to receive the selection operation of any one head portrait in described video；
When receiving described selection operation, the described head portrait selecting operation corresponding is defined as the head portrait of pending object.
Method the most according to claim 1, it is characterised in that described method also includes:
Obtain the view of multiple angles of the head replacing object；
Utilize the replacement head portrait of the different head attitude replacing object described in the view generation of at least one angle；
The all replacement head portraits generated are stored in described head portrait storehouse.
Method for processing video frequency the most according to claim 1, it is characterised in that described according to object pending in described video
Head portrait, obtain the head pose of described pending object, including:
Extract head feature and the feature locations of pending object in described two field picture；
Determine the described feature locations shift ratio relative to the contour edge of the head portrait of described pending object；
According to described shift ratio, determine between direction of visual lines and the shooting direction of pending object described in described two field picture
Angle；
The head pose of described pending object is determined according to described angle.
Method for processing video frequency the most according to claim 1, it is characterised in that described in the described two field picture by described video
The head portrait of pending object is replaced by described replacement head portrait, including:
Obtain the head portrait of the described pending object band of position in described two field picture；
Delete the head portrait of described pending object in the described band of position；
Described replacement head portrait is added in the described band of position.
Method the most according to claim 1, it is characterised in that pending right described in the described two field picture by described video
The head portrait of elephant is replaced by described replacement head portrait, including:
Obtain the head portrait of the described pending object band of position in described two field picture；
By described replacement head portrait Overlapping display on the described band of position of described two field picture.
8. a video process apparatus, it is characterised in that including:
First acquisition module, for the head portrait according to object pending in video, obtains the head pose of described pending object；
Select module, for the replacement head that selection head pose from head portrait storehouse is consistent with the head pose of described pending object
Picture；
Changing module, the head portrait for pending object described in the two field picture by described video is replaced by described replacement head portrait.
Method the most according to claim 8, it is characterised in that described device also includes:
Second acquisition module, head reference picture during for obtaining the different head attitude of pending object；
Search module, for judging whether there is the similarity with described head reference picture in the two field picture of video more than pre-
If the head portrait of threshold value；
First determines module, during for being more than the head portrait of predetermined threshold value when the similarity of existence with described head reference picture, really
The two field picture of fixed described video includes the head portrait of described pending object.
Device the most according to claim 8, it is characterised in that described device also includes:
3rd acquisition module, for obtaining the head portrait of different objects in described video；
Judge module, operates the selection of any one head portrait in described video for judging whether to receive；
Second determines module, for when receiving described selection operation, is defined as treating by the described head portrait selecting operation corresponding
Process the head portrait of object.
11. devices according to claim 8, it is characterised in that described device also includes:
4th acquisition module, for obtaining the view of multiple angles of the head replacing object；
Generation module, for utilizing the replacement head of the different head attitude replacing object described in the view generation of at least one angle
Picture；
Memory module, for storing all replacement head portraits generated in described head portrait storehouse.
12. video process apparatus according to claim 8, it is characterised in that described acquisition module, including:
Extract submodule, for extracting head feature and the feature locations of pending object in described two field picture；
First determines submodule, for determining the described feature locations contour edge relative to the head portrait of described pending object
Shift ratio；
Second determines submodule, for according to described shift ratio, determines the sight line of pending object described in described two field picture
Angle between direction and shooting direction；
3rd determines submodule, for determining the head pose of described pending object according to described angle.
13. video process apparatus according to claim 8, it is characterised in that described replacing module, including:
First obtains submodule, is used for the head portrait the obtaining described pending object band of position in described two field picture；
Delete submodule, for deleting the head portrait of the described pending object in the described band of position；
Add submodule, for being added in the described band of position by described replacement head portrait.
14. devices according to claim 8, it is characterised in that described replacing module, including:
Second obtains submodule, is used for the head portrait the obtaining described pending object band of position in described two field picture；
Display sub-module, is used for described replacement head portrait Overlapping display in the described band of position of described two field picture.
15. 1 kinds of terminals, it is characterised in that described terminal includes:
Processor；
For storing the memorizer of processor executable；
Wherein, described processor is configured to:
According to the head portrait of object pending in video, obtain the head pose of described pending object；
The replacement head portrait that head pose is consistent with the head pose of described pending object is selected from head portrait storehouse；
The head portrait of pending object described in two field picture by described video is replaced by described replacement head portrait.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610483353.0A CN106101771A (en) | 2016-06-27 | 2016-06-27 | Method for processing video frequency, device and terminal |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610483353.0A CN106101771A (en) | 2016-06-27 | 2016-06-27 | Method for processing video frequency, device and terminal |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106101771A true CN106101771A (en) | 2016-11-09 |
Family
ID=57213662
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610483353.0A Pending CN106101771A (en) | 2016-06-27 | 2016-06-27 | Method for processing video frequency, device and terminal |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106101771A (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107508745A (en) * | 2017-08-11 | 2017-12-22 | 北京三快在线科技有限公司 | The association input method and device and computing device of prompting message |
| CN108966017A (en) * | 2018-08-24 | 2018-12-07 | 太平洋未来科技（深圳）有限公司 | Video generation method, device and electronic equipment |
| CN109429098A (en) * | 2017-08-24 | 2019-03-05 | 中兴通讯股份有限公司 | Method for processing video frequency, device and terminal |
| CN109697746A (en) * | 2018-11-26 | 2019-04-30 | 深圳艺达文化传媒有限公司 | Self-timer video cartoon head portrait stacking method and Related product |
| CN110266973A (en) * | 2019-07-19 | 2019-09-20 | 腾讯科技（深圳）有限公司 | Method for processing video frequency, device, computer readable storage medium and computer equipment |
| CN111225147A (en) * | 2020-01-17 | 2020-06-02 | 珠海格力电器股份有限公司 | Video shooting control method and device, storage medium and shooting equipment |
| CN111372136A (en) * | 2020-02-28 | 2020-07-03 | 上海万面智能科技有限公司 | Batch video scene content editing method and device |
| CN111491124A (en) * | 2020-04-17 | 2020-08-04 | 维沃移动通信有限公司 | Video processing method and device and electronic equipment |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2006097902A3 (en) * | 2005-03-18 | 2007-03-29 | Philips Intellectual Property | Method of performing face recognition |
| CN102196245A (en) * | 2011-04-07 | 2011-09-21 | 北京中星微电子有限公司 | Video play method and video play device based on character interaction |
| CN103258316A (en) * | 2013-03-29 | 2013-08-21 | 东莞宇龙通信科技有限公司 | Method and device for picture processing |
| CN105096354A (en) * | 2014-05-05 | 2015-11-25 | 腾讯科技（深圳）有限公司 | Image processing method and device |
| CN105469379A (en) * | 2014-09-04 | 2016-04-06 | 广东中星电子有限公司 | Video target area shielding method and device |
-
2016
- 2016-06-27 CN CN201610483353.0A patent/CN106101771A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2006097902A3 (en) * | 2005-03-18 | 2007-03-29 | Philips Intellectual Property | Method of performing face recognition |
| CN102196245A (en) * | 2011-04-07 | 2011-09-21 | 北京中星微电子有限公司 | Video play method and video play device based on character interaction |
| CN103258316A (en) * | 2013-03-29 | 2013-08-21 | 东莞宇龙通信科技有限公司 | Method and device for picture processing |
| CN105096354A (en) * | 2014-05-05 | 2015-11-25 | 腾讯科技（深圳）有限公司 | Image processing method and device |
| CN105469379A (en) * | 2014-09-04 | 2016-04-06 | 广东中星电子有限公司 | Video target area shielding method and device |
Non-Patent Citations (1)
| Title |
|---|
| DMITRI BITOUK等: "Face Swapping：Automatically Replacing Faces in Photographs", 《ACM TRANSACTIONS ON GRAPHICS》 * |
Cited By (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107508745A (en) * | 2017-08-11 | 2017-12-22 | 北京三快在线科技有限公司 | The association input method and device and computing device of prompting message |
| CN109429098A (en) * | 2017-08-24 | 2019-03-05 | 中兴通讯股份有限公司 | Method for processing video frequency, device and terminal |
| CN108966017A (en) * | 2018-08-24 | 2018-12-07 | 太平洋未来科技（深圳）有限公司 | Video generation method, device and electronic equipment |
| WO2020037681A1 (en) * | 2018-08-24 | 2020-02-27 | 太平洋未来科技（深圳）有限公司 | Video generation method and apparatus, and electronic device |
| CN108966017B (en) * | 2018-08-24 | 2021-02-12 | 太平洋未来科技（深圳）有限公司 | Video generation method and device and electronic equipment |
| CN109697746A (en) * | 2018-11-26 | 2019-04-30 | 深圳艺达文化传媒有限公司 | Self-timer video cartoon head portrait stacking method and Related product |
| CN110266973A (en) * | 2019-07-19 | 2019-09-20 | 腾讯科技（深圳）有限公司 | Method for processing video frequency, device, computer readable storage medium and computer equipment |
| CN110266973B (en) * | 2019-07-19 | 2020-08-25 | 腾讯科技（深圳）有限公司 | Video processing method, video processing device, computer-readable storage medium and computer equipment |
| CN111225147A (en) * | 2020-01-17 | 2020-06-02 | 珠海格力电器股份有限公司 | Video shooting control method and device, storage medium and shooting equipment |
| CN111372136A (en) * | 2020-02-28 | 2020-07-03 | 上海万面智能科技有限公司 | Batch video scene content editing method and device |
| CN111491124A (en) * | 2020-04-17 | 2020-08-04 | 维沃移动通信有限公司 | Video processing method and device and electronic equipment |
| CN111491124B (en) * | 2020-04-17 | 2023-02-17 | 维沃移动通信有限公司 | Video processing method and device and electronic equipment |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106101771A (en) | Method for processing video frequency, device and terminal | |
| CN104583902B (en) | The identification of improved gesture | |
| CN107590453B (en) | Processing method, device and equipment, the computer storage medium of augmented reality scene | |
| US6987520B2 (en) | Image region filling by exemplar-based inpainting | |
| EP2706507B1 (en) | Method and apparatus for generating morphing animation | |
| CN110533587A (en) | A kind of SLAM method of view-based access control model prior information and map recovery | |
| US10051180B1 (en) | Method and system for removing an obstructing object in a panoramic image | |
| CN103916587A (en) | Photographing device for producing composite image and method using the same | |
| CN101706793A (en) | Method and device for searching picture | |
| CN104813258A (en) | Data input device | |
| CN106331491A (en) | Photographing method and device | |
| CN105892685A (en) | Method and device for searching subjects of intelligent equipment | |
| US20220254143A1 (en) | Method and apparatus for determining item name, computer device, and storage medium | |
| CN106548185A (en) | A kind of foreground area determines method and apparatus | |
| CN110213493A (en) | Equipment imaging method, device, storage medium and electronic equipment | |
| WO2022001648A1 (en) | Image processing method and apparatus, and device and medium | |
| CN105513083A (en) | PTAM camera tracking method and device | |
| JP2007094679A (en) | Image analyzing device, image analyzing program and image analyzing program storage medium | |
| CN106648344A (en) | Screen content adjustment method and equipment | |
| CN106101540A (en) | Focusing determines method and device | |
| CN110463177A (en) | The bearing calibration of file and picture and device | |
| Xiong et al. | Snap angle prediction for 360 panoramas | |
| CN110213492A (en) | Equipment imaging method, device, storage medium and electronic equipment | |
| CN111951158B (en) | Unmanned aerial vehicle aerial image splicing interruption recovery method, device and storage medium | |
| WO2022111461A1 (en) | Recognition method and apparatus, and electronic device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20161109 |