CN107341784A - A kind of expression moving method and electronic equipment - Google Patents
A kind of expression moving method and electronic equipment Download PDFInfo
- Publication number
- CN107341784A CN107341784A CN201610283525.XA CN201610283525A CN107341784A CN 107341784 A CN107341784 A CN 107341784A CN 201610283525 A CN201610283525 A CN 201610283525A CN 107341784 A CN107341784 A CN 107341784A
- Authority
- CN
- China
- Prior art keywords
- user
- point parameter
- poker
- faced
- face
- 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
- 230000014509 gene expression Effects 0.000 title claims abstract description 34
- 238000000034 method Methods 0.000 title claims abstract description 32
- 210000000887 face Anatomy 0.000 claims description 28
- 206010053238 Amimia Diseases 0.000 claims description 18
- 230000001815 facial effect Effects 0.000 claims description 15
- 230000002411 adverse Effects 0.000 abstract description 7
- 230000000694 effects Effects 0.000 abstract description 6
- 238000012545 processing Methods 0.000 abstract description 4
- 239000011159 matrix material Substances 0.000 description 34
- 230000007704 transition Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 238000013519 translation Methods 0.000 description 6
- 230000005012 migration Effects 0.000 description 5
- 238000013508 migration Methods 0.000 description 5
- 241000208340 Araliaceae Species 0.000 description 4
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 4
- 235000003140 Panax quinquefolius Nutrition 0.000 description 4
- 235000008434 ginseng Nutrition 0.000 description 4
- 230000008859 change Effects 0.000 description 2
- 238000012163 sequencing technique Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000000151 deposition Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000008921 facial expression Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
- G06T3/4023—Scaling of whole images or parts thereof, e.g. expanding or contracting based on decimating pixels or lines of pixels; based on inserting pixels or lines of pixels
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2200/00—Indexing scheme for image data processing or generation, in general
- G06T2200/04—Indexing scheme for image data processing or generation, in general involving 3D image data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20212—Image combination
- G06T2207/20221—Image fusion; Image merging
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
- G06T2207/30201—Face
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2210/00—Indexing scheme for image generation or computer graphics
- G06T2210/22—Cropping
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Graphics (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The embodiments of the invention provide a kind of expression moving method and electronic equipment, belong to image processing field.The expression moving method includes:Obtain just poker-faced face feature point parameter corresponding with user;User's face characteristic point parameter is obtained from instant video frame;Calculate the deviation between just poker-faced face feature point parameter corresponding to the user's face characteristic point parameter and described and user;According to the deviation, the new position of the characteristic point of calculating standard faces model;According to the new position of the characteristic point of the standard faces model, target image is driven.By introducing standard faces model, adverse effect caused by being migrated because of each discrepant shape of face of user and other face features to expression can be reduced.
Description
Technical field
The present invention relates to image processing field, more particularly to a kind of expression moving method and electronic equipment.
Background technology
With the development of technology, expression migration becomes to be becoming increasingly popular.Current expression migrating technology is usual
The facial expression of user is directly migrated on target image.Due to the shape of face of different user and the difference of face
It is different, easily cause different user and make identical expression, and occur the situation of different expressions on target image,
Cause Consumer's Experience poor.
The content of the invention
Adversely affected to reduce in expression migration caused by user's otherness, the embodiments of the invention provide
A kind of expression moving method and electronic equipment.
According to the first aspect of the invention, there is provided a kind of expression moving method, methods described include:
Obtain just poker-faced face feature point parameter corresponding with user;
User's face characteristic point parameter is obtained from instant video frame;
Calculate the user's face characteristic point parameter and the just poker-faced face feature point corresponding with user
Deviation between parameter;
According to the deviation, the new position of the characteristic point of calculating standard faces model;
According to the new position of the characteristic point of the standard faces model, target image is driven.
With reference to the first aspect of the present invention, in the first possible implementation, the acquisition and user couple
The just poker-faced face feature point parameter answered includes:
Obtain the just poker-faced face feature point parameter corresponding with user prestored.
With reference to the first aspect of the present invention, in second of possible implementation, the acquisition and user couple
The just poker-faced face feature point parameter answered includes:
Just poker-faced face feature point parameter corresponding with user is obtained from instant video frame.
With reference to second of possible implementation of the first aspect of the present invention, in the third possible realization side
Obtaining just poker-faced face feature point parameter corresponding with user in formula, in the frame from instant video includes:
Every prefixed time interval, instant video frame is obtained;
Human face characteristic point parameter corresponding with user is obtained from the instant video frame;
Judge whether the human face characteristic point parameter corresponding with user meets just poker-faced condition;
If it is, it is special that the human face characteristic point parameter corresponding with user is output into just poker-faced face
Sign point parameter queue;And
When the queue reaches preset length, the just poker-faced face feature point parameter queue is calculated
Average characteristics point parameter is as just poker-faced face feature point parameter corresponding with user.
With reference to the third possible implementation of the first aspect of the present invention, the 4th kind of possible realization side
In formula, the just poker-faced condition includes:
Face it is amimia, facial for front, face is symmetrical and opens eyes and shuts up.
With reference to the third possible implementation of the first aspect of the present invention, the 5th kind of possible realization side
In formula, the average characteristics point parameter for calculating the just poker-faced face feature point parameter queue as with
Just poker-faced face feature point parameter includes corresponding to user:
Calculate the weighted average characteristic point parameter of the just poker-faced face feature point parameter queue as with
Just poker-faced face feature point parameter corresponding to family.
It is described from instant video frame in the 6th kind of possible implementation with reference to the first aspect of the present invention
Middle acquisition user's face characteristic point parameter includes:
The user's face characteristic point parameter of frontization is obtained from the instant video frame.
With reference to the first aspect of the present invention, in the 7th kind of possible implementation, methods described also includes:
The renewal just poker-faced face feature point parameter corresponding with user.
According to the second aspect of the invention, there is provided a kind of electronic equipment, the electronic equipment include:
First acquisition module, for obtaining just poker-faced face feature point parameter corresponding with user;
Second acquisition module, for obtaining user's face characteristic point parameter from instant video frame;
Deviation computing module, for calculate the user's face characteristic point parameter with it is described it is corresponding with user just
Deviation between poker-faced face feature point parameter;
New position computing module, for according to the deviation, the new position of the characteristic point of calculating standard faces model
Put;
Drive module, for the new position of the characteristic point according to the standard faces model, drive target image.
With reference to the second aspect of the present invention, in the first possible implementation, first acquisition module
It is specifically used for:
Obtain the just poker-faced face feature point parameter corresponding with user prestored.
With reference to the second aspect of the present invention, in second of possible implementation, first acquisition module
It is specifically used for:
Just poker-faced face feature point parameter corresponding with user is obtained from instant video frame.
With reference to second of possible implementation of the second aspect of the present invention, in the third possible realization side
In formula, first acquisition module is specifically used for:
Every prefixed time interval, instant video frame is obtained;
Human face characteristic point parameter corresponding with user is obtained from the instant video frame;
Judge whether the human face characteristic point parameter corresponding with user meets just poker-faced condition;
If it is, it is special that the human face characteristic point parameter corresponding with user is output into just poker-faced face
Sign point parameter queue;And
When the queue reaches preset length, the just poker-faced face feature point parameter queue is calculated
Average characteristics point parameter is as just poker-faced face feature point parameter corresponding with user.
With reference to the third possible implementation of the second aspect of the present invention, the 4th kind of possible realization side
In formula, the just poker-faced condition includes:
Face it is amimia, facial for front, face is symmetrical and opens eyes and shuts up.
With reference to second of possible implementation of the second aspect of the present invention, the 5th kind of possible realization side
In formula, the computing module is specifically used for:
Calculate the weighted average characteristic point parameter of the just poker-faced face feature point parameter queue as with
Just poker-faced face feature point parameter corresponding to family.
With reference to the second aspect of the present invention, in the 6th kind of possible implementation, second acquisition module
It is specifically used for:
The user's face characteristic point parameter of frontization is obtained from the instant video frame.
With reference to the second aspect of the present invention, in the 7th kind of possible implementation, the electronic equipment also wraps
Include:
Update module, for updating the just poker-faced face feature point parameter corresponding with user.
According to the third aspect of the invention we, there is provided a kind of electronic equipment, the electronic equipment include memory,
Camera and the processor being connected with memory and camera, wherein, memory is used to store batch processing
Code, the program code that processor calls memory to be stored are used to perform following operation:
Obtain just poker-faced face feature point parameter corresponding with user;
User's face characteristic point parameter is obtained from instant video frame;
Calculate the user's face characteristic point parameter and the just poker-faced face feature point corresponding with user
Deviation between parameter;
According to the deviation, the new position of the characteristic point of calculating standard faces model;
According to the new position of the characteristic point of the standard faces model, target image is driven.
With reference to the third aspect of the present invention, in the first possible implementation, processor calls memory
The program code stored is used to perform following operation:
Obtain the just poker-faced face feature point parameter corresponding with user prestored.
With reference to the third aspect of the present invention, in second of possible implementation, processor calls memory
The program code stored is used to perform following operation:
Just poker-faced face feature point parameter corresponding with user is obtained from instant video frame.
With reference to second of possible implementation of the third aspect of the present invention, in the third possible realization side
In formula, the program code that processor calls memory to be stored is used to perform following operation:
Every prefixed time interval, instant video frame is obtained;
Human face characteristic point parameter corresponding with user is obtained from the instant video frame;
Judge whether the human face characteristic point parameter corresponding with user meets just poker-faced condition;
If it is, it is special that the human face characteristic point parameter corresponding with user is output into just poker-faced face
Sign point parameter queue;And
When the queue reaches preset length, the just poker-faced face feature point parameter queue is calculated
Average characteristics point parameter is as just poker-faced face feature point parameter corresponding with user.
With reference to the third possible implementation of the third aspect of the present invention, the 4th kind of possible realization side
In formula, the just poker-faced condition includes:
Face it is amimia, facial for front, face is symmetrical and opens eyes and shuts up.
With reference to the third possible implementation of the third aspect of the present invention, the 5th kind of possible realization side
In formula, the program code that processor calls memory to be stored is used to perform following operation:
Calculate the weighted average characteristic point parameter of the just poker-faced face feature point parameter queue as with
Just poker-faced face feature point parameter corresponding to family.
With reference to the third aspect of the present invention, in the 6th kind of possible implementation, processor calls memory
The program code stored is used to perform following operation:
The user's face characteristic point parameter of frontization is obtained from the instant video frame.
With reference to the third aspect of the present invention, in the 7th kind of possible implementation, processor calls memory
The program code stored is used to perform following operation:
The renewal just poker-faced face feature point parameter corresponding with user.
The embodiments of the invention provide a kind of expression moving method and electronic equipment.By special according to user's face
The deviation between just poker-faced face feature point parameter corresponding to point parameter and described and user is levied, to calculate
The new position of the characteristic point of standard faces, and target image is driven according to new position, it can reduce because using
Each discrepant shape of face in family and other face features and to expression migrate caused by adverse effect.
Brief description of the drawings
Technical scheme in order to illustrate the embodiments of the present invention more clearly, institute in being described below to embodiment
The accompanying drawing needed to use is briefly described, it should be apparent that, drawings in the following description are only the present invention
Some embodiments, for those of ordinary skill in the art, on the premise of not paying creative work,
Other accompanying drawings can also be obtained according to these accompanying drawings.
Fig. 1 shows a kind of flow chart of expression moving method according to embodiments of the present invention;
Fig. 2, which is shown, obtains positive nothing corresponding with user in the frame according to embodiments of the present invention from instant video
The flow chart of expressive facial features point parameter;
Fig. 3 shows a kind of flow chart of expression moving method according to embodiments of the present invention;
Fig. 4 shows the block diagram of a kind of electronic equipment according to embodiments of the present invention;
Fig. 5 shows the block diagram of the first acquisition module according to embodiments of the present invention;
Fig. 6 shows the block diagram of a kind of electronic equipment according to embodiments of the present invention;
Fig. 7 shows the block diagram of a kind of electronic equipment according to embodiments of the present invention.
Embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with the embodiment of the present invention
In accompanying drawing, the technical scheme in the embodiment of the present invention is clearly and completely described, it is clear that retouched
The embodiment stated is only part of the embodiment of the present invention, rather than whole embodiments.Based in the present invention
Embodiment, those of ordinary skill in the art obtained under the premise of creative work is not made it is all its
His embodiment, belongs to the scope of protection of the invention.
The embodiment of the invention discloses a kind of expression moving method, this method can be applied to electronic equipment, electronics
Equipment can be smart mobile phone, tablet PC, notebook, desktop computer or special game control
Device processed etc..
Fig. 1 shows a kind of expression moving method according to embodiments of the present invention.As shown in figure 1, this method
Comprise the following steps:
Step 101:Obtain just poker-faced face feature point parameter corresponding with user.
Just poker-faced face feature point parameter includes the two-dimensional coordinate of just poker-faced face feature point, such as
The two-dimensional coordinate of 66 just poker-faced face feature points.
In one example, the step includes obtaining the just poker-faced face corresponding with user prestored
Characteristic point parameter.Prestoring here can include being locally stored in advance and long-range storage in advance.
In another example, the step includes obtaining front corresponding with user from instant video frame without table
Feelings face feature point parameter.
Step 102:User's face characteristic point parameter is obtained from instant video frame.
Specifically, can be according to active appearance models (Active Appearance Model), active shape mould
Type (Active Shape Model) or other modes, user's face feature is obtained from instant video frame
Point parameter.User's face characteristic point parameter includes the two-dimensional coordinate of user's face characteristic point, such as 66 features
The two-dimensional coordinate of point.
In one example, the step includes the user's face characteristic point that frontization is obtained from instant video frame
Parameter.Specifically, the process can be:
A, the two-dimensional coordinate of user's face characteristic point is obtained.
B, the three-dimensional coordinate of the characteristic point of default three-dimensional face model is obtained.
Specifically, the step can be:
The characteristic point of the default three-dimensional face model is calculated according to training and recurrence;
Obtain this coordinate and preset three-dimensional coordinate corresponding to the characteristic point of three-dimensional face model.
C, the two-dimensional coordinate of user's face characteristic point is extended into three-dimensional coordinate by way of zero padding;
D, according to the coordinate of the three-dimensional feature point after extension, the seat of the three-dimensional feature point of default three-dimensional face model
Mark, preset formula and preparatory condition, obtain spin matrix, scaled matrix and translation matrix.
Exemplary, preset formula is:
Wherein, R represents spin matrix, and S represents scaled matrix, and T represents translation matrix, and M is three-dimensional face
The coordinates matrix of the three-dimensional feature point of model, X3For the three-dimensional after the two-dimensional coordinate matrix-expand of human face characteristic point
Coordinates matrix;
Preparatory condition includes:
Preparatory condition one:
RRT=I,
Wherein, R represents spin matrix;
Preparatory condition two:
||Σ(xi+xi')||2< ε1,
Wherein, xi∈ X, xi' ∈ X, X be the face characteristic point two-dimensional coordinate matrix, xiAnd xi' it is the people
The value of the abscissa of axisymmetric 2 points of left and right on the face, ε1For default first threshold;
Preparatory condition three:
||Σ(yi-yi')||2< ε2,
Wherein, yi∈ X, yi' ∈ X, X be the face characteristic point two-dimensional coordinate matrix, yiAnd yi' it is the people
The value of the ordinate of axisymmetric 2 points of left and right on the face, ε2For default Second Threshold.
Formula can be obtained according to the preset formula and preparatory condition:
According to formula (3) and given M, X3Can be in the hope of corresponding R, S, T.
E, according to the two-dimensional coordinate of user's face characteristic point and the spin matrix, scaled matrix and translation matrix
Obtain the two-dimensional coordinate of the user's face characteristic point of frontization.
Specifically, the process can be:
Can be in the hope of the two-dimensional coordinate square of the user's face characteristic point of frontization according to R, S, T and the X tried to achieve
Battle array X', calculation formula are:
X'=SRX+T,
Wherein, X is the two-dimensional coordinate matrix of user's face characteristic point.
Step 103:Calculate the user's face characteristic point parameter and the just poker-faced face corresponding with user
Deviation between portion's characteristic point parameter.
Specifically, the two-dimensional coordinate of each user's face characteristic point is subtracted into front corresponding with user accordingly
The two-dimensional coordinate of amimia face feature point, obtain deviation corresponding to each characteristic point.Using 66 characteristic points as
Example, then obtain 66 deviations corresponding to 66 characteristic points.General 66 characteristic points can form matrix, thus
Corresponding 66 deviations can also form matrix.
The deviation being calculated can reflect the expression information of user.
Step 104:According to the deviation, the new position of the characteristic point of calculating standard faces model.
Specifically, deviation corresponding to each characteristic point is added to the position of the character pair point in standard faces model
Put, the new position of the character pair point in standard faces model is calculated.
By the deviation according to reflection user's expression information, to calculate the character pair point in standard faces model
New position so that only reflect user's expression information in standard faces model, and eliminate user's face in itself
Adverse effect caused by individual uniqueness.
Step 105:According to the new position of the characteristic point of the standard faces model, target image is driven.
Specifically, the step can be:
According to the corresponding relation and standard between the characteristic point of standard faces model and the characteristic point of target image
The new position of the characteristic point of faceform, determine the new position of the characteristic point of target image.Standard faces model
Characteristic point and target image characteristic point between corresponding relation can prestore, or immediately meter
Calculate.
According to triangulation and linear interpolation, the new position of other pixels of target image is determined.Can be pre-
Delaunay (Delaunay) triangulation is first carried out to target image according to characteristic point, obtained every in target image
The triangle that characteristic point corresponding to one pixel is formed, and the coordinate position of each pixel can
Linearly expressed with three feature summits as the triangle where it.Triangulation can be precalculated and deposit
Storage, or instant computing.Utilize the new position of characteristic point and each pixel and triangle around it
Characteristic point between linear relationship, the new position of each pixel can be obtained by linear interpolation.This
Sample just drives whole target image so that it shows the expression of user.
It is noted that show that step 101 and step 102 do not have sequencing in figure, in other examples
In, step 101 and 102 can also have sequencing, such as step 101 is held before step 102
OK, or step 101 performs after step 102.
The embodiments of the invention provide a kind of expression moving method.By according to user's face characteristic point parameter with
Deviation between just poker-faced face feature point parameter corresponding to described and user, to calculate standard faces
The new position of characteristic point, and target image is driven according to new position, it can reduce because user is each variant
Shape of face and other face features and to expression migrate caused by adverse effect.
As shown in Fig. 2 in one embodiment, positive nothing corresponding with user is obtained from instant video frame
Expressive facial features point parameter may include following steps:
201:Every prefixed time interval, instant video frame is obtained.
Prefixed time interval can be set according to being actually needed, such as 0.15 second or other numerical value.
202:Human face characteristic point parameter corresponding with user is obtained from the instant video frame.
Specifically, can be according to active appearance models (Active Appearance Model), active shape mould
Type (Active Shape Model) or other modes, obtained from instant video frame corresponding with user
Human face characteristic point parameter.Characteristic point parameter includes the two-dimensional coordinate of characteristic point, such as the two dimension of 66 characteristic points
Coordinate.
203:Judge whether the human face characteristic point parameter corresponding with user meets just poker-faced condition.
Specifically, just poker-faced condition may include amimia face, frontal face, face is symmetrical and opens
Eye is shut up.
In one example, judge whether human face characteristic point parameter corresponding with user meets that this is just poker-faced
Condition may include whether serial judgement human face characteristic point parameter corresponding with user meets that face is amimia, facial
Front, face are symmetrical and open eyes and shut up, and when the sub- condition of aforementioned four is judged to being, judge with
Human face characteristic point parameter corresponding to user meets just poker-faced condition, and when any sub- condition criterion is no,
Judge that human face characteristic point parameter corresponding with user is unsatisfactory for just poker-faced condition and no longer carries out following son
Condition judgment.Certainly, serial order can be according to the amimia, frontal face of face, face symmetrically with
And the open eyes order shut up or other orders.
In another example, judge whether human face characteristic point parameter corresponding with user meets the front without table
Feelings condition may include whether parallel judgement human face characteristic point parameter corresponding with user meets facial amimia, face
Portion front, face is symmetrical and eye opening is shut up, and when four subconditional judged results are to meet, judges
Human face characteristic point parameter corresponding with user meets just poker-faced condition, when any subconditional judged result
During to be unsatisfactory for, judge that human face characteristic point parameter corresponding with user is unsatisfactory for just poker-faced condition.
Certainly, just poker-faced condition can also be other appropraite conditions.
204:If it is, the human face characteristic point parameter corresponding with user is output to just poker-faced face
Portion's characteristic point parameter queue.
For example, the two-dimensional coordinate of 66 characteristic points is output to the just poker-faced characteristic point parameter queue.Just
Poker-faced face feature point parameter queue can be regular length or active length.
205:When queue reaches preset length, the just poker-faced face feature point parameter queue is calculated
Average characteristics point parameter is as just poker-faced face feature point parameter corresponding with user.
Preset length may, for example, be 1,2,3 or other numerical value.Tool of the embodiment of the present invention to preset length
Body numerical value is not limited.
In one example, the step may include to calculate the common of just poker-faced face feature point parameter queue
Average characteristics point parameter is as just poker-faced face feature point parameter corresponding with user.It is therein common flat
The coefficient all same of each element in queue is referred to, numerical value is 1/ preset length.
In one example, the step may include the weighting for calculating just poker-faced face feature point parameter queue
Average characteristics point parameter is as just poker-faced face feature point parameter corresponding with user.Weight coefficient can be with
Adjusted according to preset length and characteristic point newness degree.For example, the weight coefficient of old characteristic point can be higher than
The weight coefficient of new characteristic point, and the difference of preset length may be such that the weight coefficient of characteristic point reduces.
For example, when preset length is 2, the weight coefficient of the two-dimensional coordinate of old 66 characteristic points can be 0.6,
The weight coefficient of the two-dimensional coordinate of 66 new characteristic points is 0.4, and when preset length is 3, old 66
The weight coefficient of the two-dimensional coordinate of characteristic point can be 0.5, the weighting of the two-dimensional coordinate of 66 new characteristic points
Coefficient is 0.3, and the weight coefficient of the two-dimensional coordinate of newest 66 characteristic points is 0.2, and certainly, the above is only
It is citing, weight coefficient can be other numerical value.The embodiment of the present invention is not limited specific weight coefficient
It is fixed.
By obtaining average just poker-faced characteristic point parameter, the just poker-faced feature of user can be caused
Point is smoother, avoids the noise of indivedual just poker-faced characteristic points.
As shown in figure 3, in one embodiment, the above method also includes:
301:The renewal just poker-faced face feature point parameter corresponding with user.
Specifically, the renewal process may include step 201-204.Step 201-204 details can be found in above
It is described.After step 204, the renewal process also includes:When queue often increases preset length, calculate
The average characteristics point parameter of the just poker-faced face feature point parameter queue, join in the average characteristics point
Enter row interpolation between several and described just poker-faced face feature point parameter corresponding to user, obtain transition
Just poker-faced face feature point parameter, and by the just poker-faced face feature point parameter of the transition with
The average characteristics point parameter gradually substitutes the just poker-faced face feature point parameter corresponding with user.
For example, the average characteristics point parameter calculated is N, the just poker-faced face feature point ginseng corresponding with user
Number is M, and the just poker-faced face feature point parameter of transition of acquisition is N1, N2, then N is first used in the next frame1
Substitute M and be used for follow-up mathematic interpolation, N is used in next frame again2Instead of N1For follow-up mathematic interpolation,
In frame afterwards N is replaced using N2For follow-up mathematic interpolation, two groups of Interims have merely just been lifted
Point parameter, it is to be appreciated that, the group number of Interim point parameter can be with more or less.To be illustrated
It is that preset length here can be identical or different with preset length above-mentioned.By in average characteristics
Enter row interpolation between point parameter and just poker-faced face feature point parameter corresponding to user old before, obtain
Take the just poker-faced face feature point parameter of transition, it is possible to achieve seamlessly transitting between them, so as to keep away
Exempt from average characteristics point parameter directly replacing old just poker-faced face feature point parameter corresponding with user and
Caused by feature point mutation, so as to improve Consumer's Experience.
Fig. 4 shows a kind of block diagram of electronic equipment for expression migration according to embodiments of the present invention.Such as
Shown in Fig. 4, the electronic equipment is included with lower module:
First acquisition module 401, for obtaining just poker-faced face feature point parameter corresponding with user.
Just poker-faced face feature point parameter includes the two-dimensional coordinate of just poker-faced face feature point, such as
The two-dimensional coordinate of 66 just poker-faced face feature points.
In one example, the first acquisition module 401 is used to obtain the front corresponding with user prestored
Amimia face feature point parameter.Prestoring here can include being locally stored and remotely depositing in advance in advance
Storage.
In another example, it is corresponding with user that the first acquisition module 401 is used for the acquisition from instant video frame
Just poker-faced face feature point parameter.
Second acquisition module 402, for obtaining user's face characteristic point parameter from instant video frame.
Specifically, the second acquisition module 402 can be according to active appearance models (Active Appearance
Model), active shape model (Active Shape Model) or other modes, from instant video frame
Middle acquisition user's face characteristic point parameter.User's face characteristic point parameter includes the two dimension of user's face characteristic point
Coordinate, such as the two-dimensional coordinate of 66 characteristic points.
In one example, the second acquisition module 402 is used for the user that frontization is obtained from instant video frame
Face feature point parameter.Specifically, the second acquisition module 402 is used for:
A, the two-dimensional coordinate of user's face characteristic point is obtained.
B, the three-dimensional coordinate of the characteristic point of default three-dimensional face model is obtained.
Specifically, the step can be:
The characteristic point of the default three-dimensional face model is calculated according to training and recurrence;
Obtain this coordinate and preset three-dimensional coordinate corresponding to the characteristic point of three-dimensional face model.
C, the two-dimensional coordinate of user's face characteristic point is extended into three-dimensional coordinate by way of zero padding;
D, according to the coordinate of the three-dimensional feature point after extension, the seat of the three-dimensional feature point of default three-dimensional face model
Mark, preset formula and preparatory condition, obtain spin matrix, scaled matrix and translation matrix.
Exemplary, preset formula is:
Wherein, R represents spin matrix, and S represents scaled matrix, and T represents translation matrix, and M is three-dimensional face
The coordinates matrix of the three-dimensional feature point of model, X3For the three-dimensional after the two-dimensional coordinate matrix-expand of human face characteristic point
Coordinates matrix;
Preparatory condition includes:
Preparatory condition one:
RRT=I,
Wherein, R represents spin matrix;
Preparatory condition two:
||Σ(xi+xi')||2< ε1,
Wherein, xi∈ X, xi' ∈ X, X be the face characteristic point two-dimensional coordinate matrix, xiAnd xi' it is the people
The value of the abscissa of axisymmetric 2 points of left and right on the face, ε1For default first threshold;
Preparatory condition three:
||Σ(yi-yi')||2< ε2,
Wherein, yi∈ X, yi' ∈ X, X be the face characteristic point two-dimensional coordinate matrix, yiAnd yi' it is the people
The value of the ordinate of axisymmetric 2 points of left and right on the face, ε2For default Second Threshold.
Formula can be obtained according to the preset formula and preparatory condition:
According to formula (3) and given M, X3Can be in the hope of corresponding R, S, T.
E, according to the two-dimensional coordinate of user's face characteristic point and the spin matrix, scaled matrix and translation matrix
Obtain the two-dimensional coordinate of the user's face characteristic point of frontization.
Specifically, the process can be:
Can be in the hope of the two-dimensional coordinate square of the user's face characteristic point of frontization according to R, S, T and the X tried to achieve
Battle array X', calculation formula are:
X'=SRX+T,
Wherein, X is the two-dimensional coordinate matrix of user's face characteristic point.
Deviation computing module 403, for calculate the user's face characteristic point parameter with it is described corresponding with user
Deviation between just poker-faced face feature point parameter.
Specifically, deviation computing module 403 is used to the two-dimensional coordinate of each user's face characteristic point subtracting phase
The two-dimensional coordinate for the just poker-faced face feature point corresponding with user answered, is obtained corresponding to each characteristic point
Deviation.By taking 66 characteristic points as an example, then 66 deviations corresponding to 66 characteristic points are obtained.General 66
Characteristic point can form matrix, thus corresponding 66 deviations can also form matrix.
By calculating user's face characteristic point parameter between just poker-faced characteristic point parameter corresponding with user
Deviation, can only retain user and do caused change during expression, and eliminate user's face in itself individual solely
Characteristic.
New position computing module 404, for according to the deviation, calculate standard faces model characteristic point it is new
Position.
Specifically, new position computing module 404 is used to deviation corresponding to each characteristic point adding standard faces
The position of character pair point in model, the new position of the character pair point in standard faces model is calculated.
Target image drive module 405, for the new position of the characteristic point according to the standard faces model, drive
Moving-target image.
Specifically, target image drive module 405 is used for characteristic point and target figure according to standard faces model
The new position of the characteristic point of corresponding relation and standard faces model between the characteristic point of picture, determines target figure
The new position of the characteristic point of picture.It is corresponding between the characteristic point of standard faces model and the characteristic point of target image
Relation can prestore, or instant computing.
Target image drive module 305 is additionally operable to, according to triangulation and linear interpolation, determine target image
The new position of other pixels.Delaunay (Delaunay) can be carried out to target image previously according to characteristic point
Triangulation, the triangle that characteristic point corresponding to each pixel is formed in target image is obtained, and
The coordinate position of each pixel can linearly be expressed as three feature summits of the triangle where it.
Triangulation can be precalculated and stored, or instant computing.Using characteristic point new position and
Each pixel and the linear relationship around it between characteristic point of triangle, can be obtained by linear interpolation
To the new position of each pixel.Thus drive whole target image so that it shows the table of user
Feelings.
The embodiments of the invention provide a kind of electronic equipment for expression migration.By special according to user's face
The deviation between just poker-faced face feature point parameter corresponding to point parameter and described and user is levied, to calculate
The new position of the characteristic point of standard faces, and target image is driven according to new position, it can reduce because using
Each discrepant shape of face in family and other face features and to expression migrate caused by adverse effect.
Fig. 5 shows the block diagram of the first acquisition module according to embodiments of the present invention.As shown in figure 5, first
Acquisition module 401 includes following submodule:
Frame of video acquisition submodule 501, for every prefixed time interval, obtaining instant video frame.
Prefixed time interval can be set according to being actually needed, such as 0.15 second or other numerical value.
Characteristic point acquisition submodule 502, it is special for obtaining face corresponding with user from the instant video frame
Sign point parameter.
Specifically, characteristic point acquisition submodule 502 can be according to active appearance models (Active Appearance
Model), active shape model (Active Shape Model) or other modes, from instant video frame
It is middle to obtain human face characteristic point parameter corresponding with user.Characteristic point parameter includes the two-dimensional coordinate of characteristic point, example
Such as the two-dimensional coordinate of 66 characteristic points.
Judging submodule 503, for judging whether the human face characteristic point parameter corresponding with user meets front
Amimia condition.
Specifically, just poker-faced condition may include amimia face, frontal face, face is symmetrical and opens
Eye is shut up.
In one example, judging submodule 503 is used to serially judge human face characteristic point ginseng corresponding with user
Whether number meets that amimia face, frontal face, face are symmetrical and eye opening is shut up, and in aforementioned four
When sub- condition is judged to being, judge that human face characteristic point parameter corresponding with user meets just poker-faced condition,
And when any sub- condition criterion is no, judge that human face characteristic point parameter corresponding with user is unsatisfactory for positive nothing
Expression condition and no longer carry out following sub- condition judgment.Certainly, serial order can be according to facial nothing
Expression, frontal face, facial symmetrical and the open eyes order shut up or other orders.
In another example, judging submodule 503 is used to judge human face characteristic point corresponding with user parallel
Whether parameter meets that amimia face, frontal face, face are symmetrical and eye opening is shut up, when four sub- conditions
Judged result be to judge human face characteristic point parameter satisfaction corresponding with user just poker-faced bar when meeting
Part, when any subconditional judged result is to be unsatisfactory for, judge human face characteristic point parameter corresponding with user
It is unsatisfactory for just poker-faced condition.
Certainly, just poker-faced condition can also be other appropraite conditions.
Queue submodule 504 is added, for if it is determined that the judgement of submodule 503 is, then by described and user couple
The human face characteristic point parameter answered is output to just poker-faced face feature point parameter queue.
Just poker-faced face feature point parameter queue can be regular length or active length.
Average characteristics point calculating sub module 505, for when the queue reaches preset length, calculating to be described just
The average characteristics point parameter of poker-faced face feature point parameter queue is as corresponding with user just poker-faced
Face feature point parameter.
Preset length may, for example, be 1,2,3 or other numerical value.Tool of the embodiment of the present invention to preset length
Body numerical value is not limited.
In one example, average characteristics point calculating sub module 505 is used to calculate just poker-faced facial characteristics
The common average characteristic point parameter of point parameter queue is as just poker-faced face feature point ginseng corresponding with user
Number.Common average therein refers to the coefficient all same of each element in queue, and numerical value is 1/ default length
Degree.
In one example, average characteristics point calculating sub module 505 is used to calculate just poker-faced facial characteristics
The weighted average characteristic point parameter of point parameter queue is as just poker-faced face feature point ginseng corresponding with user
Number.Weight coefficient can adjust according to preset length and characteristic point newness degree.For example, old characteristic point
Weight coefficient can be higher than the weight coefficient of new characteristic point, and the change of preset length may be such that characteristic point
Weight coefficient reduce.For example, when preset length is 2, the weight coefficient of old characteristic point can be 0.6, newly
The weight coefficient of characteristic point is 0.4, and when preset length is 3, the weight coefficient of old characteristic point can be 0.5,
The weight coefficient of new feature point is 0.3, and the weight coefficient of latest features point is 0.2, and certainly, the above is simply lifted
Example, weight coefficient can be other numerical value.The embodiment of the present invention is not limited to specific weight coefficient.
By obtaining average just poker-faced characteristic point parameter, the just poker-faced feature of user can be caused
Point is smoother, avoids the noise of indivedual just poker-faced characteristic points.
As shown in fig. 6, in one embodiment, above-mentioned electronic equipment also includes:
Update module 601, for updating the just poker-faced face feature point parameter corresponding with user.
Specifically, the update module 601 may include submodule 401-404.Submodule 401-204 details can
Referring to described above.The update module 601 also includes:Average computation submodule, for often increasing when queue
During preset length, the average characteristics point parameter of the just poker-faced face feature point parameter queue is calculated;Cross
Acquisition submodule is crossed, in the average characteristics point parameter and the just poker-faced face corresponding with user
Enter row interpolation between portion's characteristic point parameter, obtain the just poker-faced face feature point parameter of transition;And more
New submodule, for the just poker-faced face feature point parameter of the transition and the average characteristics point to be joined
Number gradually substitutes the just poker-faced face feature point parameter corresponding with user.For example, what is calculated is averaged
Characteristic point parameter is N, and the just poker-faced face feature point parameter corresponding with user is M, the mistake of acquisition
It is N to cross just poker-faced face feature point parameter1, N2, then N is first used in the next frame1Substitute M and be used for follow-up
Mathematic interpolation, N is used in next frame again2Instead of N1For follow-up mathematic interpolation, make in frame afterwards
N is replaced with N2For follow-up mathematic interpolation, two groups of Interim point parameters have merely just been lifted, but will
Understand, the group number of Interim point parameter can be with more or less.It is noted that here default
Length can be identical or different with preset length above-mentioned.By average characteristics point parameter and before it is old
Just poker-faced face feature point parameter corresponding with user between enter row interpolation, obtain the positive nothing of transition
Expressive facial features point parameter, it is possible to achieve seamlessly transitting between them, so as to avoid average characteristics point
Parameter directly replace old just poker-faced face feature point parameter corresponding with user and caused by characteristic point dash forward
Become, so as to improve Consumer's Experience.
Fig. 7 shows a kind of electronic equipment implemented according to the present invention.Shown in reference picture 7, the electronics is set
It is standby to include memory 701, camera 702 and the processor being connected with memory 701 and camera 702
703, wherein, memory 701 is used to store batch processing code, and processor 703 calls the institute of memory 701
The program code of storage is used to perform following operation:
Obtain just poker-faced face feature point parameter corresponding with user;
User's face characteristic point parameter is obtained from instant video frame;
Calculate the user's face characteristic point parameter and the just poker-faced face feature point corresponding with user
Deviation between parameter;
According to the deviation, the new position of the characteristic point of calculating standard faces model;
According to the new position of the characteristic point of the standard faces model, target image is driven.
Optionally, the program code that processor 703 calls memory 701 to be stored is used to perform following operation:
Obtain the just poker-faced face feature point parameter corresponding with user prestored.
Optionally, the program code that processor 703 calls memory 701 to be stored is used to perform following operation:
Just poker-faced face feature point parameter corresponding with user is obtained from instant video frame.
Optionally, the program code that processor 703 calls memory 701 to be stored is used to perform following operation:
Every prefixed time interval, instant video frame is obtained;
Human face characteristic point parameter corresponding with user is obtained from the instant video frame;
Judge whether the human face characteristic point parameter corresponding with user meets just poker-faced condition;
If it is, it is special that the human face characteristic point parameter corresponding with user is output into just poker-faced face
Sign point parameter queue;And
When the queue reaches preset length, the just poker-faced face feature point parameter queue is calculated
Average characteristics point parameter is as just poker-faced face feature point parameter corresponding with user.
Optionally, the just poker-faced condition includes:
Face it is amimia, facial for front, face is symmetrical and opens eyes and shuts up.
Optionally, the program code that processor 703 calls memory 701 to be stored is used to perform following operation:
Calculate the weighted average characteristic point parameter of the just poker-faced face feature point parameter queue as with
Just poker-faced face feature point parameter corresponding to family.
Optionally, the program code that processor 703 calls memory 701 to be stored is used to perform following operation:
The user's face characteristic point parameter of frontization is obtained from the instant video frame.
Optionally, the program code that processor 703 calls memory 701 to be stored is used to perform following operation:
The renewal just poker-faced face feature point parameter corresponding with user.
The embodiments of the invention provide a kind of electronic equipment.By according to user's face characteristic point parameter with it is described
Deviation between just poker-faced face feature point parameter corresponding to user, to calculate the feature of standard faces
The new position of point, and target image is driven according to new position, it can reduce because each discrepant face of user
Type and other face features and to expression migrate caused by adverse effect.
Above-mentioned all optional technical schemes, any combination can be used to form the alternative embodiment of the present invention,
This is no longer repeated one by one.
It should be noted that:Above-described embodiment provide electronic equipment when performing expression moving method, only with
The division progress of above-mentioned each functional module, can be as needed and by above-mentioned work(for example, in practical application
It can distribute and be completed by different functional modules, i.e., the internal structure of equipment is divided into different functional modules,
To complete all or part of function described above.In addition, electronic equipment and table that above-described embodiment provides
Feelings migration embodiment belongs to same design, and its specific implementation process refers to embodiment of the method, repeats no more here.
One of ordinary skill in the art will appreciate that realizing all or part of step of above-described embodiment can pass through
Hardware is completed, and by program the hardware of correlation can also be instructed to complete, described program can be stored in
In a kind of computer-readable recording medium, storage medium mentioned above can be read-only storage, disk or
CD etc..
Presently preferred embodiments of the present invention is the foregoing is only, is not intended to limit the invention, it is all the present invention's
Within spirit and principle, any modification, equivalent substitution and improvements made etc., it should be included in the present invention's
Within protection domain.
Claims (10)
1. a kind of expression moving method, it is characterised in that methods described includes:
Obtain just poker-faced face feature point parameter corresponding with user;
User's face characteristic point parameter is obtained from instant video frame;
Calculate the user's face characteristic point parameter and the just poker-faced face feature point corresponding with user
Deviation between parameter;
According to the deviation, the new position of the characteristic point of calculating standard faces model;
According to the new position of the characteristic point of the standard faces model, target image is driven.
2. according to the method for claim 1, it is characterised in that described to obtain front corresponding with user
Amimia face feature point parameter includes:
Obtain the just poker-faced face feature point parameter corresponding with user prestored.
3. according to the method for claim 1, it is characterised in that described to obtain front corresponding with user
Amimia face feature point parameter includes:
Just poker-faced face feature point parameter corresponding with user is obtained from instant video frame.
4. according to the method for claim 3, it is characterised in that in the frame from instant video obtain with
Just poker-faced face feature point parameter includes corresponding to user:
Every prefixed time interval, instant video frame is obtained;
Human face characteristic point parameter corresponding with user is obtained from the instant video frame;
Judge whether the human face characteristic point parameter corresponding with user meets just poker-faced condition;
If it is, it is special that the human face characteristic point parameter corresponding with user is output into just poker-faced face
Sign point parameter queue;And
When the queue reaches preset length, the just poker-faced face feature point parameter queue is calculated
Average characteristics point parameter is as just poker-faced face feature point parameter corresponding with user.
5. according to the method for claim 4, it is characterised in that the just poker-faced condition includes:
Face it is amimia, facial for front, face is symmetrical and opens eyes and shuts up.
6. a kind of electronic equipment, it is characterised in that the electronic equipment includes:
First acquisition module, for obtaining just poker-faced face feature point parameter corresponding with user;
Second acquisition module, for obtaining user's face characteristic point parameter from instant video frame;
Deviation computing module, for calculate the user's face characteristic point parameter with it is described it is corresponding with user just
Deviation between poker-faced face feature point parameter;
New position computing module, for according to the deviation, the new position of the characteristic point of calculating standard faces model
Put;
Drive module, for the new position of the characteristic point according to the standard faces model, drive target image.
7. electronic equipment according to claim 6, it is characterised in that first acquisition module is specific
For:
Obtain the just poker-faced face feature point parameter corresponding with user prestored.
8. electronic equipment according to claim 6, it is characterised in that first acquisition module is specific
For:
Just poker-faced face feature point parameter corresponding with user is obtained from instant video frame.
9. electronic equipment according to claim 8, it is characterised in that first acquisition module is specific
For:
Every prefixed time interval, instant video frame is obtained;
Human face characteristic point parameter corresponding with user is obtained from the instant video frame;
Judge whether the human face characteristic point parameter corresponding with user meets just poker-faced condition;
If it is, it is special that the human face characteristic point parameter corresponding with user is output into just poker-faced face
Sign point parameter queue;And
When the queue reaches preset length, the just poker-faced face feature point parameter queue is calculated
Average characteristics point parameter is as just poker-faced face feature point parameter corresponding with user.
10. electronic equipment according to claim 9, it is characterised in that the just poker-faced condition
Including:
Face it is amimia, facial for front, face is symmetrical and opens eyes and shuts up.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610283525.XA CN107341784A (en) | 2016-04-29 | 2016-04-29 | A kind of expression moving method and electronic equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610283525.XA CN107341784A (en) | 2016-04-29 | 2016-04-29 | A kind of expression moving method and electronic equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107341784A true CN107341784A (en) | 2017-11-10 |
Family
ID=60222868
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610283525.XA Pending CN107341784A (en) | 2016-04-29 | 2016-04-29 | A kind of expression moving method and electronic equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107341784A (en) |
Cited By (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108399383A (en) * | 2018-02-14 | 2018-08-14 | 深圳市商汤科技有限公司 | Expression moving method, device storage medium and program |
CN109376671A (en) * | 2018-10-30 | 2019-02-22 | 北京市商汤科技开发有限公司 | Image processing method, electronic equipment and computer-readable medium |
CN109977739A (en) * | 2017-12-28 | 2019-07-05 | 广东欧珀移动通信有限公司 | Image processing method, device, storage medium and electronic equipment |
CN110941332A (en) * | 2019-11-06 | 2020-03-31 | 北京百度网讯科技有限公司 | Expression driving method and device, electronic equipment and storage medium |
CN111314627A (en) * | 2020-02-24 | 2020-06-19 | 北京百度网讯科技有限公司 | Method and apparatus for processing video frames |
CN112528835A (en) * | 2020-12-08 | 2021-03-19 | 北京百度网讯科技有限公司 | Training method, recognition method and device of expression prediction model and electronic equipment |
CN113269700A (en) * | 2021-04-29 | 2021-08-17 | 北京达佳互联信息技术有限公司 | Video generation method and device, electronic equipment and storage medium |
WO2021218650A1 (en) * | 2020-04-29 | 2021-11-04 | 百果园技术(新加坡)有限公司 | Adaptive rigid prior model training method and training apparatus, and face tracking method and tracking apparatus |
CN115331292A (en) * | 2022-08-17 | 2022-11-11 | 武汉元紫东科技有限公司 | Facial image-based emotion recognition method and device and computer storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20040056857A1 (en) * | 2002-04-24 | 2004-03-25 | Zhengyou Zhang | System and method for expression mapping |
CN101499132A (en) * | 2009-03-12 | 2009-08-05 | 广东药学院 | Three-dimensional transformation search method for extracting characteristic points in human face image |
CN101944238A (en) * | 2010-09-27 | 2011-01-12 | 浙江大学 | Data driving face expression synthesis method based on Laplace transformation |
CN103268623A (en) * | 2013-06-18 | 2013-08-28 | 西安电子科技大学 | Static human face expression synthesizing method based on frequency domain analysis |
-
2016
- 2016-04-29 CN CN201610283525.XA patent/CN107341784A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20040056857A1 (en) * | 2002-04-24 | 2004-03-25 | Zhengyou Zhang | System and method for expression mapping |
CN101499132A (en) * | 2009-03-12 | 2009-08-05 | 广东药学院 | Three-dimensional transformation search method for extracting characteristic points in human face image |
CN101944238A (en) * | 2010-09-27 | 2011-01-12 | 浙江大学 | Data driving face expression synthesis method based on Laplace transformation |
CN103268623A (en) * | 2013-06-18 | 2013-08-28 | 西安电子科技大学 | Static human face expression synthesizing method based on frequency domain analysis |
Non-Patent Citations (2)
Title |
---|
ZICHENG LIU,等: "Expressive Expression Mapping with Ratio Images", 《IN PROCEEDINGS OF THE 28TH ANNUAL CONFERENCE ON COMPUTER GRAPHICS AND INTERACTIVE TECHNIQUES》 * |
蒲倩: "人脸表情迁移与分析方法研究", 《中国优秀硕士学位论文全文数据库》 * |
Cited By (14)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109977739A (en) * | 2017-12-28 | 2019-07-05 | 广东欧珀移动通信有限公司 | Image processing method, device, storage medium and electronic equipment |
CN108399383B (en) * | 2018-02-14 | 2021-03-23 | 深圳市商汤科技有限公司 | Expression migration method, device storage medium, and program |
CN108399383A (en) * | 2018-02-14 | 2018-08-14 | 深圳市商汤科技有限公司 | Expression moving method, device storage medium and program |
CN109376671A (en) * | 2018-10-30 | 2019-02-22 | 北京市商汤科技开发有限公司 | Image processing method, electronic equipment and computer-readable medium |
CN109376671B (en) * | 2018-10-30 | 2022-06-21 | 北京市商汤科技开发有限公司 | Image processing method, electronic device, and computer-readable medium |
CN110941332A (en) * | 2019-11-06 | 2020-03-31 | 北京百度网讯科技有限公司 | Expression driving method and device, electronic equipment and storage medium |
CN111314627B (en) * | 2020-02-24 | 2022-03-11 | 北京百度网讯科技有限公司 | Method and apparatus for processing video frames |
CN111314627A (en) * | 2020-02-24 | 2020-06-19 | 北京百度网讯科技有限公司 | Method and apparatus for processing video frames |
WO2021218650A1 (en) * | 2020-04-29 | 2021-11-04 | 百果园技术(新加坡)有限公司 | Adaptive rigid prior model training method and training apparatus, and face tracking method and tracking apparatus |
CN112528835A (en) * | 2020-12-08 | 2021-03-19 | 北京百度网讯科技有限公司 | Training method, recognition method and device of expression prediction model and electronic equipment |
CN112528835B (en) * | 2020-12-08 | 2023-07-04 | 北京百度网讯科技有限公司 | Training method and device of expression prediction model, recognition method and device and electronic equipment |
CN113269700A (en) * | 2021-04-29 | 2021-08-17 | 北京达佳互联信息技术有限公司 | Video generation method and device, electronic equipment and storage medium |
CN113269700B (en) * | 2021-04-29 | 2023-12-12 | 北京达佳互联信息技术有限公司 | Video generation method, device, electronic equipment and storage medium |
CN115331292A (en) * | 2022-08-17 | 2022-11-11 | 武汉元紫东科技有限公司 | Facial image-based emotion recognition method and device and computer storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107341784A (en) | A kind of expression moving method and electronic equipment | |
US11170558B2 (en) | Automatic rigging of three dimensional characters for animation | |
US10198845B1 (en) | Methods and systems for animating facial expressions | |
CN110399825B (en) | Facial expression migration method and device, storage medium and computer equipment | |
TW202046250A (en) | Virtual object shaping method, apparatus, device, and storage medium | |
CN111260764B (en) | Method, device and storage medium for making animation | |
US20180260994A1 (en) | Expression animation generation method and apparatus for human face model | |
CN115049799B (en) | Method and device for generating 3D model and virtual image | |
CN107369174A (en) | The processing method and computing device of a kind of facial image | |
CN104616347A (en) | Expression migration method, electronic equipment and system | |
CN107457780B (en) | Method and device for controlling mechanical arm movement, storage medium and terminal equipment | |
KR102374307B1 (en) | Modification of animated characters | |
CN107704829A (en) | A kind of face key point method for tracing and application and device | |
CN111369428A (en) | Virtual head portrait generation method and device | |
WO2021115623A1 (en) | Methods for handling occlusion in augmented reality applications using memory and device tracking and related apparatus | |
WO2024098685A1 (en) | Face driving method and apparatus for virtual character, and terminal device and readable storage medium | |
CN110580733A (en) | Data processing method and device and data processing device | |
WO2020233779A1 (en) | Method and system for training generative adversarial networks with heterogeneous data | |
CN115512014A (en) | Method for training expression driving generation model, expression driving method and device | |
CN106250813A (en) | A kind of facial expression moving method and equipment | |
EP2242274A1 (en) | Image processing apparatus, image processing method, and recording medium | |
CN112712044A (en) | Face tracking method and device, electronic equipment and storage medium | |
CN115359171B (en) | Virtual image processing method and device, electronic equipment and storage medium | |
WO2023193491A1 (en) | Information processing method and apparatus, and computer device and storage medium | |
CN116524162A (en) | Three-dimensional virtual image migration method, model updating method and related equipment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20171110 |
|
WD01 | Invention patent application deemed withdrawn after publication |