CN107194980A - Faceform's construction method, device and electronic equipment - Google Patents
Faceform's construction method, device and electronic equipment Download PDFInfo
- Publication number
- CN107194980A CN107194980A CN201710353803.9A CN201710353803A CN107194980A CN 107194980 A CN107194980 A CN 107194980A CN 201710353803 A CN201710353803 A CN 201710353803A CN 107194980 A CN107194980 A CN 107194980A
- Authority
- CN
- China
- Prior art keywords
- image
- grid
- face
- control point
- edge
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T13/00—Animation
- G06T13/20—3D [Three Dimensional] animation
- G06T13/40—3D [Three Dimensional] animation of characters, e.g. humans, animals or virtual beings
Abstract
The present invention relates to computer graphics techniques field there is provided a kind of faceform's construction method, device and electronic equipment, methods described includes：Obtain the first facial image of face；The movement of the Edge position control point of the first facial image is constrained, by position of the non-edge control point of mobile first facial image in the first facial image, multiple second facial images are generated；Dynamic human face model is built according to the first facial image and multiple second facial images, faceform's construction method that the present invention is provided uses the movement of the Edge position control point of the first facial image of constraint, and the method that the position at the non-edge control point of mobile first facial image generates multiple second facial images, the gap produced by being acted due to faceform can be avoided, in addition, pass through first facial image, just virtual dynamic human face model can be built, face dynamic effect is generated, required equipment is simple and convenient to operate.
Description
Technical field
The present invention relates to computer graphics techniques field, in particular to a kind of faceform's construction method, device
And electronic equipment.
Background technology
Three-dimensional face expression synthesis is the computer graphics techniques applied to numerous areas, the face based on single photo
Animation refers to trigger from two-dimension human face image, synthesizes the full animation expression of the people.Human face animation has broad application prospects, main
Including：3d gaming, production of film and TV, human-computer interaction interface, long-range presentation, education etc..
At present, relatively common facial expression animation generation method has：
1. the expression animation method based on MPEG-4 standards, which defines the motion road of characteristic point under several frequently seen expression
Footpath, although simple, but defect is that the expression generated is not true enough.
2. the method based on expression cloning, obtains the three-dimensional face model with expression by scanning, the Expression Mapping is arrived
On target faceform；This kind of method can generate trickle expression, have the disadvantage that each summit is required to mapping processing, amount of calculation
It is big to lead to not processing in real time.
3. the third-largest class is the interpolation technique based on key frame, this is relatively common in production of film and TV, is pre-designed face
Three-dimensional expression, transition frames are obtained by interpolation；But it is required to redesign its key frame table for each faceform
Feelings, and need the cooperation of professional art designing personnel.
The content of the invention
It is above-mentioned to improve it is an object of the invention to provide a kind of faceform's construction method, device and electronic equipment
Problem.
To achieve these goals, the technical scheme that the embodiment of the present invention is used is as follows：
In a first aspect, the invention provides a kind of faceform's construction method, for making dynamic in individual face picture
State effect, methods described includes：Obtain the first facial image of face；Constrain the shifting of the Edge position control point of the first facial image
It is dynamic, by position of the non-edge control point of mobile first facial image in the first facial image, generate multiple second faces
Image；Dynamic human face model is built according to the first facial image and multiple second facial images.
Second aspect, the invention provides a kind of faceform's construction device, for making dynamic in individual face picture
State effect, described device includes：Image collection module, the first facial image for obtaining face；Image generation module, is used for
The movement of the Edge position control point of the first facial image is constrained, by the non-edge control point of mobile first facial image the first
Position in face image, generates multiple second facial images；Faceform builds module, for according to the first facial image and many
Individual second facial image builds dynamic human face model.
The third aspect, the invention provides a kind of electronic equipment, the electronic equipment include memory, processor and
Faceform's construction device, faceform's construction device is stored in the memory and including one or more by described
The software function module of computing device, faceform's construction device includes：Image collection module, for obtaining face
First facial image；Image generation module, the movement of the Edge position control point for constraining the first facial image, passes through mobile first
Position of the non-edge control point of facial image in the first facial image, generates multiple second facial images；Faceform's structure
Block is modeled, for building dynamic human face model according to the first facial image and multiple second facial images.
Compared with the prior art, the invention has the advantages that：A kind of faceform's construction method that the present invention is provided,
Device and electronic equipment, using the movement of the Edge position control point of the first facial image of constraint, and move the first facial image
The method that the position at non-edge control point generates multiple second facial images, so that in the non-edge control of mobile first facial image
When system point, face edge contour can be kept constant, so as to avoid the gap produced by being acted due to faceform,
In addition, by first facial image, virtual dynamic human face model just can be built, face dynamic effect is generated, with showing
There is the interpolation technique based on key frame in technology to compare, required equipment is simple and convenient to operate.
To enable the above objects, features and advantages of the present invention to become apparent, preferred embodiment cited below particularly, and coordinate
Appended accompanying drawing, is described in detail below.
Brief description of the drawings
In order to illustrate the technical solution of the embodiments of the present invention more clearly, below will be attached to what is used required in embodiment
Figure is briefly described, it will be appreciated that the following drawings illustrate only certain embodiments of the present invention, therefore is not construed as pair
The restriction of scope, for those of ordinary skill in the art, on the premise of not paying creative work, can also be according to this
A little accompanying drawings obtain other related accompanying drawings.
Fig. 1 shows the block diagram for the electronic equipment that present pre-ferred embodiments are provided.
Fig. 2 shows faceform's construction method flow chart that present pre-ferred embodiments are provided.
Fig. 3 is the sub-step flow chart of the step S102 shown in Fig. 2.
Fig. 4 is the sub-step flow chart of the sub-step S1023 shown in Fig. 3.
Fig. 5 is the sub-step flow chart of the sub-step S1024 shown in Fig. 3.
Fig. 6 is the sub-step flow chart of the step S104 shown in Fig. 2.
Fig. 7 shows the block diagram for faceform's construction device that present pre-ferred embodiments are provided.
Fig. 8 be Fig. 7 shown in faceform's construction device in image generation module block diagram.
Fig. 9 be Fig. 8 shown in image generation module in grid image generation unit block diagram.
Figure 10 be Fig. 8 shown in image generation module in image rendering unit block diagram.
Figure 11 be Fig. 7 shown in faceform's construction device in human face expression add module block diagram.
Icon：100- electronic equipments；110- memories；120- storage controls；130- processors；200- faceform's structures
Build device；201- image collection modules；202- image generation modules；2021- character controls point acquisition module；2022- grid charts
As acquiring unit；2023- grid image generation units；The sub- execution units of 20231- first；The sub- execution units of 20232- second；
The sub- execution units of 20233- the 3rd；The sub- execution units of 20234- the 4th；2024- image rendering units；Of 20241- the 5th are performed
Unit；The sub- execution units of 20242- the 6th；The sub- execution units of 20243- the 7th；203- faceforms build module；204- faces
Expression add module；2041- base image selecting units；2042- image-region acquiring units；The new Face image synthesis of 2043-
Unit；2044- human face expression generation units.
Embodiment
Below in conjunction with accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Ground is described, it is clear that described embodiment is only a part of embodiment of the invention, rather than whole embodiments.Generally exist
The component of the embodiment of the present invention described and illustrated in accompanying drawing can be arranged and designed with a variety of configurations herein.Cause
This, the detailed description of the embodiments of the invention to providing in the accompanying drawings is not intended to limit claimed invention below
Scope, but it is merely representative of the selected embodiment of the present invention.Based on embodiments of the invention, those skilled in the art are not doing
The every other embodiment obtained on the premise of going out creative work, belongs to the scope of protection of the invention.
It should be noted that：Similar label and letter represents similar terms in following accompanying drawing, therefore, once a certain Xiang Yi
It is defined in individual accompanying drawing, then it further need not be defined and explained in subsequent accompanying drawing.Meanwhile, the present invention's
In description, term " first ", " second " etc. are only used for distinguishing description, and it is not intended that indicating or implying relative importance.
Fig. 1 is refer to, Fig. 1 shows the block diagram for the electronic equipment 100 that present pre-ferred embodiments are provided.Electronics
Equipment 100 may be, but not limited to, smart mobile phone, tablet personal computer, pocket computer on knee, vehicle-mounted computer, individual digital and help
Manage (personal digital assistant, PDA), Wearable mobile terminal etc..The electronic equipment 100 includes face
Model construction device 200, memory 110, storage control 120 and processor 130.
The memory 110, storage control 120 and each element of processor 130 are directly or indirectly electrical each other
Connection, to realize the transmission or interaction of data.For example, these elements can pass through one or more communication bus or letter each other
Number line, which is realized, to be electrically connected with.Faceform's construction device 200 can be with software or firmware (firmware) including at least one
Form be stored in the memory 110 or be solidificated in the operating system (operating of the electronic equipment 100
System, OS) in software function module.The processor 130 is used to perform the executable module stored in memory 110,
Software function module or computer program that such as described faceform's construction device 200 includes.
Wherein, memory 110 may be, but not limited to, random access memory (Random Access Memory,
RAM), read-only storage (Read Only Memory, ROM), programmable read only memory (Programmable Read-Only
Memory, PROM), erasable read-only memory (Erasable Programmable Read-Only Memory, EPROM),
Electricallyerasable ROM (EEROM) (Electric Erasable Programmable Read-Only Memory, EEPROM) etc..
Wherein, memory 110 is used for storage program, and the processor 130 performs described program, this hair after execute instruction is received
Method performed by the server for the flow definition that bright any embodiment is disclosed can apply in processor 130, or by
Reason device 130 is realized.
Processor 130 can be a kind of IC chip, with signal handling capacity.Above-mentioned processor 130 can be with
It is general processor, including central processing unit (Central Processing Unit, CPU), network processing unit (Network
Processor, NP), speech processor and video processor etc.；Can also be digital signal processor, application specific integrated circuit,
Field programmable gate array or other PLDs, discrete gate or transistor logic, discrete hardware components.
It can realize or perform disclosed each method, step and the logic diagram in the embodiment of the present invention.General processor can be
Microprocessor or the processor 130 can also be any conventional processors etc..
First embodiment
Fig. 2 is refer to, Fig. 2 shows faceform's construction method flow chart that present pre-ferred embodiments are provided.Face
Model building method comprises the following steps：
Step S101, obtains the first facial image of face.
In embodiments of the present invention, the first facial image can be obtained by pointing out user voluntarily to input, such as face
When model building method starts to perform, the display screen display reminding information " upload pictures " of electronic equipment 100, user can select
The a certain local picture for being stored in electronic equipment 100 is uploaded, and the first facial image can be a positive people of face
Face picture, the positive face picture of face, which may be, but not limited to, the face picture, includes the portions such as eyebrow, eyes, nose, face
The characteristic image of position.
As a kind of embodiment, after the first facial image for getting user's upload, the method for machine learning is utilized
Face is detected, the method for machine learning may be, but not limited to, using the face template image pre-established, with the first face
Image carries out matching so as to detect comprising characteristic areas such as eyebrow, eyes, nose and faces in face, face template image,
If Face datection fails, the display screen of electronic equipment 100 can be with display reminding information " None- identified face ", now user
It can select to upload face picture again.
Step S102, constrains the movement of the Edge position control point of the first facial image, passes through the non-of mobile first facial image
Position of the Edge position control o'clock in the first facial image, generates multiple second facial images.
In embodiments of the present invention, detect after the face in the first facial image, the side of the face is detected first
Edge control point and non-edge control point, then carry out gridding, and keep Edge position control point the first to the first facial image
Position in face image is constant, passes through position of the mobile non-edge control point in the first facial image so that by non-edge control
Position of the grid vertex of system point influence in the first facial image changes, and the change in location of grid vertex causes grid top
The corresponding grid of point is deformed upon, so as to obtain the second facial image.
Fig. 3 is refer to, step S102 also includes following sub-step：
Sub-step S1021, obtains the character control point of the first facial image, and character control point includes Edge position control point and non-
Edge position control point.
In embodiments of the present invention, character control point can be used as one by carrying out contrast acquisition with face template image
Kind of embodiment, character control point has 68, including facial contour edge, eyes edge, face edge, the bridge of the nose and the wing of nose
Characteristic point, character control point includes Edge position control point and non-edge control point.
As a kind of embodiment, Edge position control point can be the character control point at facial contour edge, detect edge control
Making the method for point can be：Partial Feature control point is connected using curve to form a connected region A, detects whether to deposit
It is more than connected region A in connected region B area, if there is no such connected region B, is then in connected region A
Borderline character control point be the first facial image Edge position control point；If there is such connected region B, then examine
The area surveyed with the presence or absence of a connected region C is more than connected region B, the like, until finding out Edge position control point.
Sub-step S1022, carries out gridding to the first facial image, obtains the first face net with multiple grid vertexes
Table images.
In embodiments of the present invention, gridding can carry out two-dimensional mesh to the first facial image to format, and two-dimensional mesh is formatted
It may be, but not limited to, and the first facial image is divided into several triangles, obtain first with multiple triangular mesh
Face grid image, grid vertex can be the summit of multiple triangular mesh.
Sub-step S1023, constrained edge control point, by mobile non-edge control point in the first face grid image
Position, Control vertices move to obtain multiple second face grid images.
In embodiments of the present invention, keep position of the Edge position control o'clock in the first facial image constant, Edge position control point
After movement, the transposed matrix and Edge position control point of the movement of Edge position control point are calculated by edge the weighing factor of grid vertex
The new coordinate of the grid vertex of control point influence, the grid vertex is moved at new coordinate by original coordinate, grid top
The changes in coordinates of point causes corresponding grid to deform upon, so as to obtain the second facial image.
Fig. 4 is refer to, step S1023 also includes following sub-step：
Sub-step S10231, keeps the position of Edge position control point constant.
In embodiments of the present invention, it can make Edge position control o'clock in the first face that the position of holding Edge position control point is constant
Coordinate in grid image is constant.
Sub-step S10232, carries out displacement to non-edge control point, determines the transposed matrix at non-edge control point.
In embodiments of the present invention, the determination of transposed matrix can include two methods：
A kind of method be first by way of man-machine interaction by research staff manually to non-edge control point carry out displacement,
Then the position further according to the position before the displacement of non-edge control point and after displacement calculates transposed matrix, and computational methods are：Obtain
The position C ' after position C and displacement before the displacement of non-edge control point；According to formula C '=MC, calculate non-edge control point by
C to C ' transposed matrix M=C ' C-1。
Another method is the value for the transposed matrix M for setting non-edge control point automatically by electronic equipment 100 first, so
The position C before the displacement of non-edge control point and the transposed matrix M at non-edge control point are utilized afterwards, calculate non-edge control point displacement
Position C ' afterwards.
Sub-step S10233, when grid vertex is not in the edge of the first face grid image, controls the grid vertex
It is mobile, when grid vertex is in the edge of the first face grid image, constrain the movement of the grid vertex.
As a kind of embodiment, using the method that Edge position control point is detected in abovementioned steps S1021, grid vertex is detected
Whether the edge of first face grid image is in, when grid vertex is not in the edge of the first face grid image, according to
The weighing factor of position C ' and non-edge dominating pair of vertices grid vertex after the displacement of non-edge control point, the mobile grid vertex；
When grid vertex is in the edge of the first face grid image, keeps the position of the grid vertex constant, can so ensure
The edge pixel point of second facial image can be merged with the first facial image, it is to avoid produce gap.
Sub-step S10234, according to the transposed matrix at non-edge control point and the shadow of the non-edge dominating pair of vertices grid vertex
Weight is rung, Control vertices are moved to the second place by first position, wherein, first position is grid vertex in the first face
Coordinate on grid image, the second place is coordinate of the grid vertex on the second face grid image.
In embodiments of the present invention, for a grid, the change in location of the grid vertex of the grid can cause the grid
Deform, and the movement of grid vertex is by influenceing the non-edge control point of the grid vertex to determine, for example, for
A certain grid vertex V in one face grid image, as influence grid vertex V N number of non-edge control point C1,C2,...,CN
Position when changing, grid vertex V position can also change, the grid vertex V in the first face grid image
The grid vertex V ' changed in the second face grid image, grid vertex V ' computational methods can beWherein, MiTo influence grid vertex V i-th of non-edge control point CiTransform to non-edge control
Make point Ci' transposed matrix, w (V, Ci) it is i-th of non-edge control point CiTo grid vertex V weighing factor.
As a kind of embodiment, non-edge control point C can be by the edge to grid vertex V weighing factor w (V, C)
Control point C is determined to grid vertex V Euler's distance, that is to say, that w (V, C) and Edge position control point C is to grid vertex V's
Euler's distance is inversely proportional.
It is used as a kind of embodiment, any one non-edge control point C for influenceing grid vertex V, the non-edge control
Make point C is to the weighing factor of grid vertex：
Wherein, k is a constant coefficient, and dist (V, C) is non-edge control point C to grid
Summit V Euler's distance, dist (V, C) calculation formula is：
Wherein, (xV,yV),(xC,yC) it is respectively grid vertex V and non-
Edge position control point C coordinate information.
As a kind of embodiment, several non-edge control points C for influenceing grid vertex V1,C2,...,CN, should
Several non-edge control points C1,C2,...,CNIt is respectively to the weighing factor of grid vertex：
Wherein, k is a constant coefficient, influence grid vertex V several non-edge control points C1,C2,...,CNChoosing
Take and meet constraints
Sub-step S1024, is rendered to each second face grid image, generates multiple second facial images.
In embodiments of the present invention, the dynamic human face model for constructing, therefore, to assure that the uv coordinates in faceform
It can be mapped on the first facial image, that is to say, that it is original that faceform can upload after rectangular projection with user
The pixel of human face region is essentially coincided in picture, it is desirable, therefore, to assure that in the second facial image any one pixel pixel
Value, the pixel value of pixel that be in the first facial image corresponding with the pixel is equal, and the second face grid image is entered
It is exactly by the pixel value assignment of the pixel in the first facial image to the corresponding pixel of the second facial image that row, which is rendered,.
Fig. 5 is refer to, sub-step S1024 also includes following sub-step：
Sub-step S10241, obtains the first pixel value of each grid vertex in the first face grid image.
As a kind of embodiment, any one deformation triangular mesh in the second face grid image is chosen, and look for
To three grid vertex V of the deformation triangular mesh1′、V2′、V3' three in the first face grid image are corresponded to respectively
Grid vertex V1、V2、V3, it is assumed that three grid vertex V1、V2、V3Pixel value be respectively I (V1)、I(V2)、I(V3)。
Sub-step S10242, sets the second picture of grid vertex corresponding with the grid vertex in the second face grid image
Element value is equal to the first pixel value.
In embodiments of the present invention, it is assumed that three grid vertex V1′、V2′、V3' pixel value be respectively I (V1′)、I
(V2′)、I(V3'), then by three grid vertex V1、V2、V3Pixel value correspond be assigned to three grid vertex V1′、V2′、
V3', that is to say, that so that pixel value I (V1')=I (V1),I(V2')=I (V2),I(V3')=I (V3)。
Sub-step S10243, according to the second pixel of the grid vertex in the second face grid image, it is determined that with the net
The pixel value of other pixels of the grid of lattice vertex correspondence.
In embodiments of the present invention, three grid vertex V of deformation triangular mesh are obtained1′、V2′、V3' pixel value it
Afterwards, it is possible to use the pixel value of three grid vertexes of the deformation triangular mesh carries out linear fit and obtains the deformation triangle
Other pixel pixel values of grid.For any one pixel on the deformation triangular mesh side, approximating method can be：
Assuming that three grid vertex V of the deformation triangular mesh1′(x1,y1),V2′(x2,y2),V3′(x3,y3) corresponding pixel value point
Wei not I (x1,y1),I(x2,y2),I(x3,y3), then the three of deformation triangular mesh Grid Edge l1,l2,l3Respectively：
Then any one pixel (x on this three Grid Edgesi,yi) pixel value be respectively：
For the pixel value of any one pixel in the deformation triangular mesh, approximating method can be：Assuming that should
Any one pixel p (u in deformation triangular meshi,vi), pixel p (ui,vi) pixel value I (ui,vi) by with the picture
The pixel value fitting of the marginal point of the deformation triangle of vegetarian refreshments in the same horizontal line is obtained, and marginal point includes three grid tops
Point V1′(x1,y1),V2′(x2,y2),V3′(x3,y3) and three Grid Edges on any one pixel (xi,yi)：
Wherein, pixel (u2′,v2′),(u1′,v1') it is respectively point on two Grid Edges of the deformation triangle, and
Meet v=v2'=v1′.According to the above method, the second face grid image is traveled through, until each in the second face grid image
Individual pixel all has pixel value.
, can also be according to the pixel of the area of the deformation triangular mesh and three grid vertexes as a kind of embodiment
The weight of value, calculates any one pixel p (u in the deformation triangular meshi,vi) pixel value I (ui,vi), computational methods
It is as follows：
According to three grid vertex V1′(x1,y1),V2′(x2,y2),V3′(x3,y3) calculate the face of the deformation triangular mesh
Product：
Calculate grid vertex V1′(x1,y1),V2′(x2,y2),V3′(x3,y3) pixel value weight：
Calculate any one pixel p (u in the deformation triangular meshi,vi) pixel value I (ui,vi)：
Wherein, I (u0,v0)=I (x1,y1),I(u1,v1)=I (x2,y2),I(u2,v2)=I (x3,y3)。
Step S103, dynamic human face model is built according to the first facial image and multiple second facial images.
In embodiments of the present invention, the first facial image and multiple second facial images are merged, constructs dynamic
Faceform, the dynamic human face model can realize nodded above and below face, left and right such as is shaken the head at a variety of actions.
Step S104, on the basis of dynamic human face model, adds human face expression.
In embodiments of the present invention, after constructing dynamic human face model, for the specific region in the dynamic human face model
Addition expression, characteristic area may be, but not limited to, eyes, face, eyebrow etc., for example, eye areas addition blink expression,
In face region addition smile expression etc., dynamic human face model can be caused more lively, true to nature.
Fig. 6 is refer to, step S104 also includes following sub-step：
Sub-step S1041, arbitrarily selects face grid image based on a facial image from dynamic faceform.
In embodiments of the present invention, any basic facial image of selection one from dynamic faceform, according to sub-step
The method that S1022 is introduced, gridding, basic face grid image are carried out to the basic facial image.
Sub-step S1042, obtains the characteristic image region for needing generation to act in basic face grid image.
In embodiments of the present invention, characteristic image region can be the energy such as eyes, face, the eyebrow in basic facial image
Enough produce the image-region of human face expression.
The movement of other image-regions, passes through controlling feature image in sub-step S1043, the basic face grid image of constraint
The movement of grid vertex in region, realizes the deformation of this feature image-region, obtains new facial image.
In embodiments of the present invention, the method introduced according to sub-step S1023 and sub-step S1023, realizes this feature figure
As the deformation in region, new facial image is obtained.
Sub-step S1044, new facial image is added into dynamic human face model, is arranged with basic facial image order,
And new facial image after basic facial image to generate human face expression.
In embodiments of the present invention, if to obtain the characteristic image in blink expression, the basic facial image obtained
Region is the eyes opened, after this feature image-region deformation, is the eyes closed in obtained new facial image, by new person
Face image is added after the basic facial image into dynamic human face model, and the expression of blink just occurs in dynamic human face model,
The different kinematic matrixs having by different non-edge control points, can make various expressions and action.
In embodiments of the present invention, first, by constraining the Edge position control point in the first facial image and being in first
The grid image at the edge of face grid image so that the dynamic human face model finally built is not in gap in action；
Secondly, the position by the non-edge control point of mobile first facial image in the first facial image, generates the second face figure
Picture, the final method for building dynamic human face model so that the dynamic human face model can realize nodded above and below face, left and right is shaken the head
Etc. a variety of actions；Finally, the different kinematic matrixs having by different non-edge control points, can make various each
The expression of sample and action.
Second embodiment
Fig. 7 is refer to, Fig. 7 shows that the square frame for faceform's construction device 200 that present pre-ferred embodiments are provided shows
It is intended to.Faceform's construction device 200 includes image collection module 201, image generation module 202, faceform's structure module
203 and human face expression add module 204.
Image collection module 201, the first facial image for obtaining face.
In embodiments of the present invention, image collection module 201 can be used for performing step S101.
Image generation module 202, the movement of the Edge position control point for constraining the first facial image is the first by movement
Position of the non-edge control point of face image in the first facial image, generates multiple second facial images.
In embodiments of the present invention, image generation module 202 can be used for performing step S102.
Fig. 8 is refer to, Fig. 8 shows for the square frame of image generation module 202 in faceform's construction device 200 shown in Fig. 7
It is intended to.Image generation module 202 includes character control point acquisition module 2021, grid image acquiring unit 2022, grid image
Generation unit 2023 and image rendering unit 2024.
Character control point acquisition module 2021, the character control point for obtaining the first facial image, character control point bag
Include Edge position control point and non-edge control point.
In embodiments of the present invention, character control point acquisition module 2021 can be used for performing sub-step S1021.
Grid image acquiring unit 2022, for carrying out gridding to the first facial image, is obtained with multiple grid tops
First face grid image of point.
In embodiments of the present invention, grid image acquiring unit 2022 can be used for performing sub-step S1022.
Grid image generation unit 2023, for constrained edge control point, by mobile non-edge control point the first
Position in face grid image, Control vertices move to obtain multiple second face grid images.
In embodiments of the present invention, grid image generation unit 2023 can be used for performing sub-step S1023.
Fig. 9 is refer to, Fig. 9 shows for the square frame of grid image generation unit 2023 in the image generation module 202 shown in Fig. 8
It is intended to.Grid image generation unit 2023 includes the first sub- execution unit 20231, the second sub- execution unit 20232, the 3rd son and held
The sub- execution unit 20234 of row unit 20233 and the 4th.
First sub- execution unit 20231, for keeping the position of Edge position control point constant.
In embodiments of the present invention, the first sub- execution unit 20231 can be used for performing sub-step S10231.
Second sub- execution unit 20232, for carrying out displacement to non-edge control point, determines the displacement at non-edge control point
Matrix.
In embodiments of the present invention, the second sub- execution unit 20232 can be used for performing sub-step S10232.
When the 3rd sub- execution unit 20233, edge for being not in the first face grid image when grid vertex, control
The grid vertex is moved, and when grid vertex is in the edge of the first face grid image, constrains the movement of the grid vertex.
In embodiments of the present invention, the 3rd sub- execution unit 20233 can be used for performing sub-step S10233.
4th sub- execution unit 20234, for the transposed matrix according to non-edge control point and the non-edge dominating pair of vertices
The weighing factor of grid vertex, Control vertices are moved to the second place by first position, wherein, first position is grid top
Coordinate o'clock on the first face grid image, the second place is coordinate of the grid vertex on the second face grid image.
In embodiments of the present invention, the 4th sub- execution unit 20234 can be used for performing sub-step S10234.
Image rendering unit 2024, for being rendered to each second face grid image, generates multiple second faces
Image.
In embodiments of the present invention, image rendering unit 2024 can be used for performing sub-step S1024.
Figure 10 is refer to, Figure 10 illustrates for the square frame of image rendering unit 2024 in the image generation module 202 shown in Fig. 8
Figure.Image rendering unit 2024 includes the 5th sub- execution unit 20241, the 6th sub- of execution unit 20242 and the 7th and performs list
Member 20243.
5th sub- execution unit 20241, the first pixel for obtaining each grid vertex in the first face grid image
Value.
In embodiments of the present invention, the 5th sub- execution unit 20241 can be used for performing sub-step S10241.
6th sub- execution unit 20242, for setting grid corresponding with the grid vertex in the second face grid image
Second pixel value on summit is equal to the first pixel value.
In embodiments of the present invention, the 6th sub- execution unit 20242 can be used for performing sub-step S10242.
7th sub- execution unit 20243, for the second pixel according to the grid vertex in the second face grid image,
It is determined that the pixel value of other pixels of grid corresponding with the grid vertex.
In embodiments of the present invention, the 7th sub- execution unit 20243 can be used for performing sub-step S10243.
Faceform builds module 203, for building dynamic people according to the first facial image and multiple second facial images
Face model.
In embodiments of the present invention, faceform builds module 203 and can be used for performing step S103.
Human face expression add module 204, on the basis of dynamic human face model, adding human face expression.
In embodiments of the present invention, human face expression add module 204 can be used for performing step S104.
Figure 11 is refer to, Figure 11 is human face expression add module 204 in faceform's construction device 200 shown in Fig. 7
Block diagram.Human face expression add module 204 include base image selecting unit 2041, image-region acquiring unit 2042,
New facial image generation unit 2043 and human face expression generation unit 2044.
Base image selecting unit 2041, based on one facial image of any selection from dynamic faceform
Face grid image.
In embodiments of the present invention, base image selecting unit 2041 can be used for performing step S1041.
Image-region acquiring unit 2042, for obtaining the characteristic image for needing generation to act in basic face grid image
Region.
In embodiments of the present invention, image-region acquiring unit 2042 can be used for performing step S1042.
New facial image generation unit 2043, the movement for constraining other image-regions in basic face grid image,
By the movement of grid vertex in controlling feature image-region, the deformation of this feature image-region is realized, new facial image is obtained.
In embodiments of the present invention, new facial image generation unit 2043 can be used for performing step S1043.
Human face expression generation unit 2044, for new facial image to be added into dynamic human face model, with basic face
Image sequence is arranged, and new facial image after basic facial image to generate human face expression.
In embodiments of the present invention, human face expression generation unit 2044 can be used for performing step S1044.
In summary, the present invention is provided a kind of faceform's construction method, device and electronic equipment, in individual people
Dynamic effect is made on face picture, methods described includes：Obtain the first facial image of face；Constrain the side of the first facial image
The movement at edge control point, passes through position of the non-edge control point of mobile first facial image in the first facial image, generation
Multiple second facial images；Dynamic human face model is built according to the first facial image and multiple second facial images.The present invention is carried
Faceform's construction method of confession is relative to prior art, with advantages below：First, by constraining in the first facial image
Edge position control point and be in the first face grid image edge grid image so that the dynamic human face mould finally built
Type is not in gap in action；Secondly, by the non-edge control point of mobile first facial image in the first facial image
In position, generate the second facial image, the final method for building dynamic human face model so that the dynamic human face model can be real
Nodded above and below existing face, a variety of actions such as left and right is shaken the head；Finally, by different non-edge control points have it is different
Kinematic matrix, can make various expressions and action.
In several embodiments provided herein, it should be understood that disclosed apparatus and method, it can also pass through
Other modes are realized.Device embodiment described above is only schematical, for example, flow chart and block diagram in accompanying drawing
Show according to the device of multiple embodiments of the present invention, the architectural framework in the cards of method and computer program product,
Function and operation.At this point, each square frame in flow chart or block diagram can represent the one of a module, program segment or code
Part a, part for the module, program segment or code is used to realize holding for defined logic function comprising one or more
Row instruction.It should also be noted that in some implementations as replacement, the function of being marked in square frame can also with different from
The order marked in accompanying drawing occurs.For example, two continuous square frames can essentially be performed substantially in parallel, they are sometimes
It can perform in the opposite order, this is depending on involved function.It is also noted that every in block diagram and/or flow chart
The combination of individual square frame and block diagram and/or the square frame in flow chart, can use the special base for performing defined function or action
Realize, or can be realized with the combination of specialized hardware and computer instruction in the system of hardware.
In addition, each functional module in each embodiment of the invention can integrate to form an independent portion
Point or modules individualism, can also two or more modules be integrated to form an independent part.
If the function is realized using in the form of software function module and is used as independent production marketing or in use, can be with
It is stored in a computer read/write memory medium.Understood based on such, technical scheme is substantially in other words
The part contributed to prior art or the part of the technical scheme can be embodied in the form of software product, the meter
Calculation machine software product is stored in a storage medium, including some instructions are to cause a computer equipment (can be individual
People's computer, server, or network equipment etc.) perform all or part of step of each of the invention embodiment methods described.
And foregoing storage medium includes：USB flash disk, mobile hard disk, read-only storage (ROM, Read-Only Memory), arbitrary access are deposited
Reservoir (RAM, Random Access Memory), magnetic disc or CD etc. are various can be with the medium of store program codes.Need
Illustrate, herein, such as first and second or the like relational terms be used merely to by an entity or operation with
Another entity or operation make a distinction, and not necessarily require or imply between these entities or operation there is any this reality
The relation or order on border.Moreover, term " comprising ", "comprising" or its any other variant are intended to the bag of nonexcludability
Contain, so that process, method, article or equipment including a series of key elements are not only including those key elements, but also including
Other key elements being not expressly set out, or also include for this process, method, article or the intrinsic key element of equipment.
In the absence of more restrictions, the key element limited by sentence "including a ...", it is not excluded that including the key element
Process, method, article or equipment in also there is other identical element.
The preferred embodiments of the present invention are the foregoing is only, are not intended to limit the invention, for the skill of this area
For art personnel, the present invention can have various modifications and variations.Within the spirit and principles of the invention, that is made any repaiies
Change, equivalent substitution, improvement etc., should be included in the scope of the protection.It should be noted that：Similar label and letter exists
Similar terms is represented in following accompanying drawing, therefore, once being defined in a certain Xiang Yi accompanying drawing, is then not required in subsequent accompanying drawing
It is further defined and explained.
Claims (10)
1. a kind of faceform's construction method, it is characterised in that for making dynamic effect, the side in individual face picture
Method includes：
Obtain the first facial image of face；
The movement of the Edge position control point of first facial image is constrained, passes through the non-edge control of mobile first facial image
Position of the system point in first facial image, generates multiple second facial images；
Dynamic human face model is built according to the first facial image and multiple second facial images.
2. the method as described in claim 1, it is characterised in that the Edge position control point of constraint first facial image
It is mobile, by position of the non-edge control point of mobile first facial image in first facial image, generate many
The step of individual second facial image, including：
The character control point of first facial image is obtained, the character control point includes Edge position control point and non-edge is controlled
Point；
Gridding is carried out to first facial image, the first face grid image with multiple grid vertexes is obtained：
The Edge position control point is constrained, passes through position of the mobile non-edge control point in the first face grid image
Put, control the grid vertex motion to obtain multiple second face grid images；
Each second face grid image is rendered, multiple second facial images are generated.
3. method as claimed in claim 2, it is characterised in that the constraint Edge position control point, described non-by movement
Position of the Edge position control point in the first face grid image, controls the grid vertex motion to obtain multiple second people
The step of face grid image, including：
Keep the position of the Edge position control point constant；
Displacement is carried out to the non-edge control point, the transposed matrix at the non-edge control point is determined；
When the grid vertex is not in the edge of the first face grid image, controls the grid vertex to move, work as institute
When stating edge of the grid vertex in the first face grid image, the movement of the grid vertex is constrained；
The weighing factor of grid vertex according to the transposed matrix at the non-edge control point and the non-edge dominating pair of vertices, control
Make the grid vertex and the second place is moved to by first position, wherein, the first position is the grid vertex described
Coordinate on first face grid image, the second place is the grid vertex on the second face grid image
Coordinate.
4. method as claimed in claim 2, described that each second face grid image is rendered, multiple the are generated
The step of two facial images, including：
Obtain the first pixel value of each grid vertex in the first face grid image；
The second pixel value of grid vertex corresponding with the grid vertex in the second face grid image is set to be equal to first
Pixel value；
According to the second pixel of the grid vertex in the second face grid image, it is determined that net corresponding with the grid vertex
The pixel value of other pixels of lattice.
5. method as claimed in claim 2, it is characterised in that dynamic human face mould is built according to the multiple second facial image
After the step of type, in addition to：
On the basis of the dynamic human face model, human face expression is added.
6. method as claimed in claim 5, it is characterised in that described on the basis of the dynamic human face model, adds people
The step of face is expressed one's feelings, including：
Face grid image based on a facial image is arbitrarily selected from the dynamic human face model；
Obtain the characteristic image region for needing generation to act in the basic face grid image；
The movement of other image-regions in the basic face grid image is constrained, by controlling characteristic image region Intranet
The movement on lattice summit, realizes the deformation of this feature image-region, obtains new facial image；
The new facial image is added into the dynamic human face model, arranged with basic facial image order, and it is described new
Facial image is after the basic facial image to generate human face expression.
7. a kind of faceform's construction device, it is characterised in that for making dynamic effect, the dress in individual face picture
Put including：
Image collection module, the first facial image for obtaining face；
Image generation module, the movement of the Edge position control point for constraining first facial image passes through mobile described first
Position of the non-edge control point of facial image in first facial image, generates multiple second facial images；
Faceform builds module, for building dynamic human face model according to the first facial image and multiple second facial images.
8. device as claimed in claim 7, it is characterised in that described image generation module includes：
Character control point acquiring unit, the character control point for obtaining first facial image, the character control point bag
Include Edge position control point and non-edge control point；
Grid image acquiring unit, for carrying out gridding to first facial image, is obtained with multiple grid vertexes
First face grid image：
Grid image generation unit, for constraining the Edge position control point, by the mobile non-edge control point described the
Position in one face grid image, controls the grid vertex motion to obtain multiple second face grid images；
Image rendering unit, for being rendered to each second face grid image, generates multiple second facial images.
9. device as claimed in claim 8, it is characterised in that the grid image generation unit includes：
First sub- execution unit, for keeping the position of the Edge position control point constant；
Second sub- execution unit, for carrying out displacement to the non-edge control point, determines the displacement at the non-edge control point
Matrix；
When the 3rd sub- execution unit, edge for being not in the first face grid image when the grid vertex, control
The grid vertex is moved, and when the grid vertex is in the edge of the first face grid image, constrains the grid vertex
Movement；
4th sub- execution unit, for net described in the transposed matrix according to the non-edge control point and the non-edge dominating pair of vertices
The weighing factor on lattice summit, controls the grid vertex to be moved to the second place by first position, wherein, the first position is
Coordinate of the grid vertex on the first face grid image, the second place is the grid vertex described the
Coordinate on two face grid images.
10. a kind of electronic equipment, it is characterised in that including：
Memory；
Processor；
And faceform's construction device, faceform's construction device is stored in the memory and including one or many
The individual software function module by the computing device, faceform's construction device includes：
Image collection module, the first facial image for obtaining face；
Image generation module, the movement of the Edge position control point for constraining first facial image passes through mobile described first
Position of the non-edge control point of facial image in first facial image, generates multiple second facial images；
Faceform builds module, for building dynamic human face model according to the first facial image and multiple second facial images.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710353803.9A CN107194980A (en) | 2017-05-18 | 2017-05-18 | Faceform's construction method, device and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710353803.9A CN107194980A (en) | 2017-05-18 | 2017-05-18 | Faceform's construction method, device and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107194980A true CN107194980A (en) | 2017-09-22 |
Family
ID=59874161
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710353803.9A Pending CN107194980A (en) | 2017-05-18 | 2017-05-18 | Faceform's construction method, device and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107194980A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110675475A (en) * | 2019-08-19 | 2020-01-10 | 腾讯科技（深圳）有限公司 | Face model generation method, device, equipment and storage medium |
| CN111938655A (en) * | 2020-07-09 | 2020-11-17 | 上海交通大学 | Orbit soft tissue form evaluation method, system and equipment based on key point information |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101071513A (en) * | 2007-06-20 | 2007-11-14 | 北京中星微电子有限公司 | Method and system for realizing organ animation |
| CN101383055A (en) * | 2008-09-18 | 2009-03-11 | 北京中星微电子有限公司 | Three-dimensional human face constructing method and system |
| CN101826217A (en) * | 2010-05-07 | 2010-09-08 | 上海交通大学 | Rapid generation method for facial animation |
| CN102376100A (en) * | 2010-08-20 | 2012-03-14 | 北京盛开互动科技有限公司 | Single-photo-based human face animating method |
| CN103035022A (en) * | 2012-12-07 | 2013-04-10 | 大连大学 | Facial expression synthetic method based on feature points |
| CN104063890A (en) * | 2013-03-22 | 2014-09-24 | 中国移动通信集团福建有限公司 | Method for cartooning human face and system thereof |
| CN104318606A (en) * | 2014-09-12 | 2015-01-28 | 上海明穆电子科技有限公司 | 3D face type change method and 3D face type change system |
| CN104658024A (en) * | 2013-11-21 | 2015-05-27 | 大连佑嘉软件科技有限公司 | Human face expression synthesis method based on characteristic point |
| CN105139007A (en) * | 2015-09-30 | 2015-12-09 | 小米科技有限责任公司 | Positioning method and apparatus of face feature point |
| US20150356346A1 (en) * | 2013-01-15 | 2015-12-10 | Nec Corporation | Feature point position detecting appararus, feature point position detecting method and feature point position detecting program |
| CN105374055A (en) * | 2014-08-20 | 2016-03-02 | 腾讯科技（深圳）有限公司 | Image processing method and device |
| WO2016033085A1 (en) * | 2014-08-25 | 2016-03-03 | Possibility Place, Llc | Method of making a personalized animatable mesh |
-
2017
- 2017-05-18 CN CN201710353803.9A patent/CN107194980A/en active Pending
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101071513A (en) * | 2007-06-20 | 2007-11-14 | 北京中星微电子有限公司 | Method and system for realizing organ animation |
| CN101383055A (en) * | 2008-09-18 | 2009-03-11 | 北京中星微电子有限公司 | Three-dimensional human face constructing method and system |
| CN101826217A (en) * | 2010-05-07 | 2010-09-08 | 上海交通大学 | Rapid generation method for facial animation |
| CN102376100A (en) * | 2010-08-20 | 2012-03-14 | 北京盛开互动科技有限公司 | Single-photo-based human face animating method |
| CN103035022A (en) * | 2012-12-07 | 2013-04-10 | 大连大学 | Facial expression synthetic method based on feature points |
| US20150356346A1 (en) * | 2013-01-15 | 2015-12-10 | Nec Corporation | Feature point position detecting appararus, feature point position detecting method and feature point position detecting program |
| CN104063890A (en) * | 2013-03-22 | 2014-09-24 | 中国移动通信集团福建有限公司 | Method for cartooning human face and system thereof |
| CN104658024A (en) * | 2013-11-21 | 2015-05-27 | 大连佑嘉软件科技有限公司 | Human face expression synthesis method based on characteristic point |
| CN105374055A (en) * | 2014-08-20 | 2016-03-02 | 腾讯科技（深圳）有限公司 | Image processing method and device |
| WO2016033085A1 (en) * | 2014-08-25 | 2016-03-03 | Possibility Place, Llc | Method of making a personalized animatable mesh |
| CN104318606A (en) * | 2014-09-12 | 2015-01-28 | 上海明穆电子科技有限公司 | 3D face type change method and 3D face type change system |
| CN105139007A (en) * | 2015-09-30 | 2015-12-09 | 小米科技有限责任公司 | Positioning method and apparatus of face feature point |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110675475A (en) * | 2019-08-19 | 2020-01-10 | 腾讯科技（深圳）有限公司 | Face model generation method, device, equipment and storage medium |
| CN110675475B (en) * | 2019-08-19 | 2024-02-20 | 腾讯科技（深圳）有限公司 | Face model generation method, device, equipment and storage medium |
| CN111938655A (en) * | 2020-07-09 | 2020-11-17 | 上海交通大学 | Orbit soft tissue form evaluation method, system and equipment based on key point information |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109325437B (en) | Image processing method, device and system | |
| US10679046B1 (en) | Machine learning systems and methods of estimating body shape from images | |
| CN111325851B (en) | Image processing method and device, electronic equipment and computer readable storage medium | |
| CN107316020A (en) | Face replacement method, device and electronic equipment | |
| CN108701352A (en) | Amending image using the identification based on three dimensional object model and enhancing | |
| CN105184249A (en) | Method and device for processing face image | |
| CN112819947A (en) | Three-dimensional face reconstruction method and device, electronic equipment and storage medium | |
| CN108197547A (en) | Face pose estimation, device, terminal and storage medium | |
| JP6207210B2 (en) | Information processing apparatus and method | |
| WO2020134436A1 (en) | Method for generating animated expression and electronic device | |
| US10964083B1 (en) | Facial animation models | |
| US10909744B1 (en) | Simulating garment with wrinkles based on physics based cloth simulator and machine learning model | |
| CN111047509A (en) | Image special effect processing method and device and terminal | |
| CN109410119A (en) | Mask image distortion method and its system | |
| US11741678B2 (en) | Virtual object construction method, apparatus and storage medium | |
| CN109215131A (en) | The driving method and device of conjecture face | |
| KR20230085931A (en) | Method and system for extracting color from face images | |
| CN116097316A (en) | Object recognition neural network for modeless central prediction | |
| CN107194980A (en) | Faceform's construction method, device and electronic equipment | |
| CN113822965A (en) | Image rendering processing method, device and equipment and computer storage medium | |
| CN109376618A (en) | Image processing method, device and electronic equipment | |
| EP2260403B1 (en) | Mesh transfer | |
| CN111275610A (en) | Method and system for processing face aging image | |
| CN115393487B (en) | Virtual character model processing method and device, electronic equipment and storage medium | |
| CN113076918A (en) | Video-based facial expression cloning method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170922 |
|
| RJ01 | Rejection of invention patent application after publication |