CN106934759A - The front method and electronic equipment of a kind of human face characteristic point - Google Patents
The front method and electronic equipment of a kind of human face characteristic point Download PDFInfo
- Publication number
- CN106934759A CN106934759A CN201511025139.2A CN201511025139A CN106934759A CN 106934759 A CN106934759 A CN 106934759A CN 201511025139 A CN201511025139 A CN 201511025139A CN 106934759 A CN106934759 A CN 106934759A
- Authority
- CN
- China
- Prior art keywords
- characteristic point
- dimensional coordinate
- face
- matrix
- dimensional
- 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
- 238000000034 method Methods 0.000 title claims abstract description 49
- 239000011159 matrix material Substances 0.000 claims abstract description 150
- 230000001143 conditioned effect Effects 0.000 claims abstract description 42
- 238000013519 translation Methods 0.000 claims abstract description 38
- 230000003190 augmentative effect Effects 0.000 claims description 3
- 230000008921 facial expression Effects 0.000 abstract description 9
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
Classifications
-
- G06T3/18—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformation in the plane of the image
- G06T3/40—Scaling the whole image or part thereof
- G06T3/4084—Transform-based scaling, e.g. FFT domain scaling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformation in the plane of the image
- G06T3/60—Rotation of a whole image or part thereof
Abstract
The invention discloses the front method and electronic equipment of a kind of human face characteristic point, methods described includes：Obtain the two-dimensional coordinate of the characteristic point of face；Obtain the three-dimensional coordinate of the characteristic point of default three-dimensional face model；According to the two-dimensional coordinate of the characteristic point, the three-dimensional coordinate of the characteristic point of three-dimensional face model, preset formula and pre-conditioned, acquisition spin matrix, scaled matrix and translation matrix are preset；Two-dimensional coordinate and the spin matrix according to the characteristic point, scaled matrix and translation matrix obtain the two-dimensional coordinate of the characteristic point behind the front of the face, the method is by adding a series of default conditions, so that the characteristic point in front is influenceed smaller by human face expression, with precision higher.
Description
Technical field
The present invention relates to field of computer technology, the front method and electricity of more particularly to a kind of human face characteristic point
Sub- equipment.
Background technology
The method in existing human face characteristic point front does not consider symmetry of the face in each dimension generally,
So when face has expression the feature point tolerance in front can be caused larger.
The content of the invention
In order to solve the above problems, the present invention provides the front method and electronic equipment of a kind of human face characteristic point.
The technical scheme is as follows：
First aspect, there is provided a kind of front method of human face characteristic point, methods described includes：
Obtain the two-dimensional coordinate of the characteristic point of face；
Obtain the three-dimensional coordinate of the characteristic point of default three-dimensional face model；
According to the two-dimensional coordinate of the characteristic point, the three-dimensional coordinate of the characteristic point of three-dimensional face model is preset, in advance
If formula and pre-conditioned, acquisition spin matrix, scaled matrix and translation matrix；
Two-dimensional coordinate and the spin matrix according to the characteristic point, scaled matrix and translation matrix obtain institute
State the two-dimensional coordinate of the characteristic point behind the front of face.
With reference in a first aspect, in the first possible implementation, described according to the two of the characteristic point
Dimension coordinate, presets the three-dimensional coordinate of the characteristic point of three-dimensional face model, preset formula and pre-conditioned, acquisition
Before spin matrix, scaled matrix and translation matrix, also include：
The two-dimensional coordinate of the characteristic point is expanded to the three-dimensional coordinate of characteristic point by way of zero padding.
With reference in a first aspect, in second possible implementation, the three-dimensional face model is front
Three-dimensional face model, the three-dimensional coordinate of the characteristic point of the three-dimensional face model is the characteristic point in front
Three-dimensional coordinate.
With reference in a first aspect, in the third possible implementation, it is described it is pre-conditioned including：
The spin matrix is orthogonal matrix；
The axial symmetry error of the face is less than the first preset value；
The horizontal error of the face is less than the second preset value；And,
At least one in the extreme value of the difference in height in the face between individual features point.
Second aspect, there is provided a kind of electronic equipment, the electronic equipment includes：
The two-dimensional coordinate acquisition module of characteristic point, the two-dimensional coordinate of the characteristic point for obtaining face；
The three-dimensional coordinate acquisition module of the characteristic point of default three-dimensional face model, for obtaining default three-dimensional face
The three-dimensional coordinate of the characteristic point of model；
Matrix acquisition module, for the two-dimensional coordinate according to the characteristic point, the spy for presetting three-dimensional face model
Levy three-dimensional coordinate a little, preset formula and pre-conditioned, acquisition spin matrix, scaled matrix and translation matrix；
Front module, for the two-dimensional coordinate according to the characteristic point and the spin matrix, scaled matrix
With the two-dimensional coordinate of the characteristic point behind the front that translation matrix obtains the face.
With reference to second aspect, in the first possible implementation, the equipment also includes being augmented module,
Three-dimensional coordinate for the two-dimensional coordinate of the characteristic point to be expanded to characteristic point by way of zero padding.
With reference to second aspect, in second possible implementation, the three-dimensional face model is front
Three-dimensional face model, the three-dimensional coordinate of the characteristic point of the three-dimensional face model is the characteristic point in front
Three-dimensional coordinate.
With reference to second aspect, in the third possible implementation, it is described it is pre-conditioned including：
The spin matrix is orthogonal matrix；
The axial symmetry error of the face is less than the first preset value；
The horizontal error of the face is less than the second preset value；And,
At least one in the extreme value of the difference in height in the face between individual features point.
The third aspect, there is provided a kind of electronic equipment, the equipment include memory, and with the storage
The processor of device connection, wherein, the memory is used to store batch processing code, and the processor is called
The program code that the memory is stored is used to perform following operation：
Obtain the two-dimensional coordinate of the characteristic point of face；
Obtain the three-dimensional coordinate of the characteristic point of default three-dimensional face model；
According to the two-dimensional coordinate of the characteristic point, the three-dimensional coordinate of the characteristic point of three-dimensional face model is preset, in advance
If formula and pre-conditioned, acquisition spin matrix, scaled matrix and translation matrix；
Two-dimensional coordinate and the spin matrix according to the characteristic point, scaled matrix and translation matrix obtain institute
State the two-dimensional coordinate of the characteristic point behind the front of face.
With reference to the third aspect, in the first possible implementation, the processor calls the memory
The program code for being stored is used to perform following operation：
The two-dimensional coordinate of the characteristic point is expanded to the three-dimensional coordinate of characteristic point by way of zero padding.
With reference to the third aspect, in second possible implementation, the three-dimensional face model is front
Three-dimensional face model, the three-dimensional coordinate of the characteristic point of the three-dimensional face model is the characteristic point in front
Three-dimensional coordinate.
With reference to the third aspect, in the third possible implementation, it is described it is pre-conditioned including：
The spin matrix is orthogonal matrix；
The axial symmetry error of the face is less than the first preset value；
The horizontal error of the face is less than the second preset value；And,
At least one in the extreme value of the difference in height in the face between individual features point.
The front method and electronic equipment of a kind of human face characteristic point are the embodiment of the invention provides, by adding
A series of default conditions so that the characteristic point in front is influenceed smaller by human face expression, with higher
Precision.
Brief description of the drawings
Technical scheme in order to illustrate more clearly the embodiments of the present invention, institute in being described to embodiment below
The accompanying drawing for needing 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 is a kind of flow chart of the front method of human face characteristic point provided in an embodiment of the present invention；
Fig. 2 is a kind of flow chart of the front method of human face characteristic point provided in an embodiment of the present invention；
Fig. 3 is a kind of flow chart of the front method of human face characteristic point provided in an embodiment of the present invention；
Fig. 4 is a kind of flow chart of the front method of human face characteristic point provided in an embodiment of the present invention；
Fig. 5 is a kind of flow chart of the front method of human face characteristic point provided in an embodiment of the present invention；
Fig. 6 is the structural representation of a kind of electronic equipment provided in an embodiment of the present invention；
Fig. 7 is the structural representation of a kind of electronic equipment provided in an embodiment of the present invention；
Fig. 8 is face two-dimensional coordinate system schematic diagram provided in an embodiment of the present invention.
Specific 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.
The front method and electronic equipment of a kind of human face characteristic point are the embodiment of the invention provides, by adding
A series of default conditions so that the characteristic point in front is influenceed smaller by human face expression, with higher
Precision.
Embodiment one
The embodiment of the invention provides a kind of front method of human face characteristic point, shown in reference picture 1, method
Flow includes：
101st, the two-dimensional coordinate of the characteristic point of face is obtained.
Specifically, the process can be：
The characteristic point of face is obtained according to active appearance models (Active Appearance Model)；
Obtain the corresponding two-dimensional coordinate of characteristic point of the face.
The feature of face can also be optionally obtained by active shape model (Active Shape Model)
Point；
Obtain the corresponding two-dimensional coordinate of characteristic point of the face.
102nd, the three-dimensional coordinate of the characteristic point of default three-dimensional face model is obtained.
Specifically, the process can be：
The characteristic point of the default three-dimensional face model is calculated according to training and recurrence；
Obtain the corresponding three-dimensional coordinate of characteristic point of the default three-dimensional face model of this coordinate.
103rd, according to the two-dimensional coordinate of the characteristic point, the three-dimensional coordinate of the characteristic point of three-dimensional face model is preset,
Preset formula and pre-conditioned, acquisition spin matrix, scaled matrix and translation matrix.
Specifically, the process can be：
The two-dimensional coordinate of this feature point is extended to the three-dimensional coordinate of characteristic point by way of zero padding；
According to the coordinate of the three-dimensional feature point after the extension, the seat of the three-dimensional feature point of three-dimensional face model is preset
Mark, preset formula and pre-conditioned acquisition spin matrix, scaled matrix and translation matrix.
104th, according to the characteristic point two-dimensional coordinate and the spin matrix, scaled matrix and translation matrix are obtained
Take the two-dimensional coordinate of the characteristic point behind the front of the face.
Specifically, the process can be：
According to the spin matrix for obtaining, scaled matrix and translation matrix are adjusted to the two-dimensional coordinate of this feature point
It is whole, obtain the two-dimensional coordinate of the characteristic point behind front.
A kind of front method of human face characteristic point is the embodiment of the invention provides, it is a series of default by adding
Condition so that the characteristic point in front is influenceed smaller by human face expression, with precision higher.
Embodiment two
The embodiment of the invention provides a kind of front method of human face characteristic point, shown in reference picture 2, method
Flow includes：
201st, the two-dimensional coordinate of the characteristic point of face is obtained.
The step is identical with step 101, and here is omitted.
202nd, the three-dimensional coordinate of the characteristic point of default three-dimensional face model is obtained.
The step is identical with step 102, and here is omitted.
203rd, according to the two-dimensional coordinate of the characteristic point, the three-dimensional coordinate of the characteristic point of three-dimensional face model is preset,
Preset formula and pre-conditioned, acquisition spin matrix, scaled matrix and translation matrix.
Specifically, the process can be：
The two-dimensional coordinate of this feature point is extended to the three-dimensional coordinate of this feature point by way of zero padding；
According to the three-dimensional coordinate of the characteristic point after the extension, the three-dimensional seat of the characteristic point of three-dimensional face model is preset
Mark, preset formula and pre-conditioned acquisition 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 three-dimensional coordinate matrix of the characteristic point of model, X3It is the three-dimensional after the two-dimensional coordinate matrix-expand of human face characteristic point
Coordinates matrix；
It is pre-conditioned including：
RRT=I,
Wherein, R represents spin matrix；
According to the preset formula and pre-conditioned can obtain formula：
s.t.RRT=I
According to formula (1) and given M, X3Can be in the hope of corresponding R, S, T.
204th, according to the characteristic point two-dimensional coordinate and the spin matrix, scaled matrix and translation matrix,
Obtain the two-dimensional coordinate of the characteristic point behind the front of the face.
Specifically, the process can be：
According to the R for trying to achieve, S, T and X can be in the hope of the two-dimensional coordinate matrixes of the human face characteristic point behind front
X', computing formula is：
X'=SRX+T
Wherein, X is the two-dimensional coordinate matrix of the characteristic point of the face.
A kind of front method of human face characteristic point is the embodiment of the invention provides, by add rotation related
It is pre-conditioned so that the characteristic point in front is influenceed smaller by face rotation, with precision higher.
Embodiment three
The embodiment of the invention provides a kind of front method of human face characteristic point, shown in reference picture 3, method
Flow includes：
301st, the two-dimensional coordinate of the characteristic point of face is obtained.
The step is identical with step 101, and here is omitted.
302nd, the three-dimensional coordinate of the characteristic point of default three-dimensional face model is obtained.
The step is identical with step 102, and here is omitted.
303rd, according to the two-dimensional coordinate of the characteristic point, the three-dimensional coordinate of the characteristic point of three-dimensional face model is preset,
Preset formula and pre-conditioned, acquisition spin matrix, scaled matrix and translation matrix.
Specifically, the process can be：
The two-dimensional coordinate of this feature point is extended to the three-dimensional coordinate of this feature point by way of zero padding；
According to the coordinate of the three-dimensional feature point after the extension, the seat of the three-dimensional feature point of three-dimensional face model is preset
Mark, preset formula and pre-conditioned acquisition 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, X3It is the three-dimensional after the two-dimensional coordinate matrix-expand of human face characteristic point
Coordinates matrix；
It is pre-conditioned including：
Pre-conditioned one：
RRT=I,
Wherein, R represents spin matrix；
Pre-conditioned two：
||Σ(xi+x′i)||2＜ ε1,
Wherein, the coordinate system shown in reference picture 8, xi∈ X, x 'i∈ X, X are the two dimension of the characteristic point of the face
Coordinates matrix, xiWith x 'iIt is the value of the abscissa of axisymmetric 2 points of left and right on the face, ε1It is default first
Threshold value.
According to the preset formula and pre-conditioned can obtain formula：
s.t.RRT=I, (2)
||Σ(xi+x′i)||2＜ ε1
According to formula (2) and given M, X3Can be in the hope of corresponding R, S, T.
304th, according to the characteristic point two-dimensional coordinate and the spin matrix, scaled matrix and translation matrix are obtained
Take the two-dimensional coordinate of the characteristic point behind the front of the face.
Specifically, the process can be：
According to the R for trying to achieve, S, T and X can be in the hope of the two-dimensional coordinate matrixes of the human face characteristic point behind front
X', computing formula is：
X'=SRX+T,
Wherein, X is the two-dimensional coordinate matrix of the characteristic point of the face.
A kind of front method of human face characteristic point is the embodiment of the invention provides, by add rotation related
It is pre-conditioned related to left and right axial symmetry pre-conditioned so that shadow of the characteristic point in front by human face expression
Sound is smaller, with precision higher.
Example IV
The embodiment of the invention provides a kind of front method of human face characteristic point, shown in reference picture 4, method
Flow includes：
401st, the two-dimensional coordinate of the characteristic point of face is obtained.
The step is identical with step 101, and here is omitted.
402nd, the three-dimensional coordinate of the characteristic point of default three-dimensional face model is obtained.
The step is identical with step 102, and here is omitted.
403rd, according to the two-dimensional coordinate of the characteristic point, the three-dimensional coordinate of the characteristic point of three-dimensional face model is preset,
Preset formula and pre-conditioned, acquisition spin matrix, scaled matrix and translation matrix.
Specifically, the process can be：
The two-dimensional coordinate of this feature point is extended to the three-dimensional coordinate of this feature point by way of zero padding；
According to the coordinate of the three-dimensional feature point after the extension, the seat of the three-dimensional feature point of three-dimensional face model is preset
Mark, preset formula and pre-conditioned acquisition 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, X3It is the three-dimensional after the two-dimensional coordinate matrix-expand of human face characteristic point
Coordinates matrix；
It is pre-conditioned including：
Pre-conditioned one：
RRT=I,
Wherein, R represents spin matrix；
Pre-conditioned two：
||Σ(xi+x′i)||2＜ ε1,
Wherein, the coordinate system shown in reference picture 8, xi∈ X, x 'i∈ X, X are the two dimension of the characteristic point of the face
Coordinates matrix, xiWith x 'iIt is the value of the abscissa of axisymmetric 2 points of left and right on the face, ε1It is default first
Threshold value；
Pre-conditioned three：
||Σ(yi-y′i)||2＜ ε2,
Wherein, the coordinate system shown in reference picture 8, yi∈ X, y 'i∈ X, X are the two dimension of the characteristic point of the face
Coordinates matrix, yiWith y 'iIt is the value of the ordinate of axisymmetric 2 points of left and right on the face, ε2It is default second
Threshold value.
According to the preset formula and pre-conditioned can obtain formula：
s.t.RRT=I
||Σ(xi+x′i)||2＜ ε1, (3)
||Σ(yi-y′i)||2＜ ε2
According to formula (3) and given M, X3Can be in the hope of corresponding R, S, T.
404th, according to the characteristic point two-dimensional coordinate and the spin matrix, scaled matrix and translation matrix are obtained
Take the two-dimensional coordinate of the characteristic point behind the front of the face.
Specifically, the process can be：
According to the R for trying to achieve, S, T and X can be in the hope of the two-dimensional coordinate matrixes of the human face characteristic point behind front
X', computing formula is：
X'=SRX+T,
Wherein, X is the two-dimensional coordinate matrix of the characteristic point of the face.
A kind of front method of human face characteristic point is the embodiment of the invention provides, by add rotation related
It is pre-conditioned related to left and right axial symmetry pre-conditioned so that shadow of the characteristic point in front by human face expression
Sound is smaller, with precision higher.
Embodiment five
The embodiment of the invention provides a kind of front method of human face characteristic point, shown in reference picture 5, method
Flow includes：
501st, the two-dimensional coordinate of the characteristic point of face is obtained.
The step is identical with step 101, and here is omitted.
502nd, the three-dimensional coordinate of the characteristic point of default three-dimensional face model is obtained.
The step is identical with step 102, and here is omitted.
503rd, according to the two-dimensional coordinate of the characteristic point, the three-dimensional coordinate of the characteristic point of three-dimensional face model is preset,
Preset formula and pre-conditioned, acquisition spin matrix, scaled matrix and translation matrix.
Specifically, the process can be：
The two-dimensional coordinate of this feature point is extended to the three-dimensional coordinate of this feature point by way of zero padding；
According to the coordinate of the three-dimensional feature point after the extension, the seat of the three-dimensional feature point of three-dimensional face model is preset
Mark, preset formula and pre-conditioned acquisition 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, X3It is the three-dimensional after the two-dimensional coordinate matrix-expand of human face characteristic point
Coordinates matrix；
It is pre-conditioned including：
Pre-conditioned one：
RRT=I,
Wherein, R represents spin matrix；
Pre-conditioned two：
||Σ(xi+x′i)||2＜ ε1,
Wherein, the coordinate system shown in reference picture 8, xi∈ X, x 'i∈ X, X are the two dimension of the characteristic point of the face
Coordinates matrix, xiWith x 'iIt is the value of the abscissa of axisymmetric 2 points of left and right on the face, ε1It is default first
Threshold value；
Pre-conditioned three：
||Σ(yi-y′i)||2＜ ε2,
Wherein, the coordinate system shown in reference picture 8, yi∈ X, y 'i∈ X, X are the two dimension of the characteristic point of the face
Coordinates matrix, yiWith y 'iIt is the value of the ordinate of axisymmetric 2 points of left and right on the face, ε2It is default second
Threshold value；
Pre-conditioned four：
||ye-yn|-y | ＜ ε3,
Wherein, the coordinate system shown in reference picture 8, yeFor in the face the corresponding characteristic point of left eye interior angle it is vertical
Sit target value, ynIt is the value of the ordinate of corresponding characteristic point on the outside of the left nose wing in the face, y is default institute
State the maximum of ordinate difference between two characteristic points.
According to the preset formula and pre-conditioned can obtain formula：
s.t.RRT=I
||Σ(xi+x′i)||2＜ ε1, (4)
||Σ(yi-y′i)||2＜ ε2
||ye-yn|-y | ＜ ε3
According to formula (4) and given M, X3Can be in the hope of corresponding R, S, T.
504th, according to the characteristic point two-dimensional coordinate and the spin matrix, scaled matrix and translation matrix are obtained
Take the two-dimensional coordinate of the characteristic point behind the front of the face.
Specifically, the process can be：
According to the R for trying to achieve, S, T and X can be in the hope of the two-dimensional coordinate matrixes of the human face characteristic point behind front
X', computing formula is：
X'=SRX+T,
Wherein, X is the two-dimensional coordinate matrix of the characteristic point of the face.
A kind of front method of human face characteristic point is the embodiment of the invention provides, by add rotation related
It is pre-conditioned, the related pre-conditioned and extremum conditions of left and right axial symmetry so that the characteristic point in front is received
The influence of human face expression is smaller, with precision higher.
Embodiment six
A kind of electronic equipment is the embodiment of the invention provides, shown in reference picture 6, the electronic equipment includes：
The two-dimensional coordinate acquisition module 601 of characteristic point, the two-dimensional coordinate of the characteristic point for obtaining face；
The three-dimensional coordinate acquisition module 602 of the characteristic point of default three-dimensional face model, for obtaining default three-dimensional people
The three-dimensional coordinate of the characteristic point of face model；
Matrix acquisition module 603, for the two-dimensional coordinate according to the characteristic point, presets three-dimensional face model
The three-dimensional coordinate of characteristic point, preset formula and pre-conditioned, acquisition spin matrix, scaled matrix and translation square
Battle array；
Front module 604, for the two-dimensional coordinate according to the characteristic point and the spin matrix, scales square
Battle array and translation matrix obtain the two-dimensional coordinate of the characteristic point behind the front of the face.
Optionally, the equipment also include be augmented module 605, for by way of zero padding by the characteristic point
Two-dimensional coordinate expand to the three-dimensional coordinate of characteristic point.
Optionally, the three-dimensional face model is the three-dimensional face model in front, the three-dimensional face model
Characteristic point three-dimensional coordinate be front characteristic point three-dimensional coordinate.
Optionally, it is described it is pre-conditioned including：
The spin matrix is orthogonal matrix；
The axial symmetry error of the face is less than the first preset value；
The horizontal error of the face is less than the second preset value；And,
At least one in the extreme value of the difference in height in the face between individual features point.
A kind of electronic equipment is the embodiment of the invention provides, by adding a series of default conditions so that just
The characteristic point in face is influenceed smaller by human face expression, with precision higher.
Embodiment seven
A kind of electronic equipment is the embodiment of the invention provides, shown in reference picture 7, the equipment includes memory
701, and the processor 702 being connected with memory 701, wherein, memory 701 is used to store one group of journey
Sequence code, processor 702 calls the program code that memory 701 is stored for performing following operation：
Obtain the two-dimensional coordinate of the characteristic point of face；
Obtain the three-dimensional coordinate of the characteristic point of default three-dimensional face model；
According to the two-dimensional coordinate of the characteristic point, the three-dimensional coordinate of the characteristic point of three-dimensional face model is preset, in advance
If formula and pre-conditioned, acquisition spin matrix, scaled matrix and translation matrix；
Two-dimensional coordinate and the spin matrix according to the characteristic point, scaled matrix and translation matrix obtain institute
State the two-dimensional coordinate of the characteristic point behind the front of face.
Optionally, processor 702 calls the program code that memory 701 is stored for performing following operation：
The two-dimensional coordinate of the characteristic point is expanded to the three-dimensional coordinate of characteristic point by way of zero padding.
Optionally, the three-dimensional coordinate of the characteristic point of the three-dimensional face model is the three-dimensional of the characteristic point in front
Coordinate.
Optionally, it is described it is pre-conditioned including：
The spin matrix is orthogonal matrix；
The axial symmetry error of the face is less than the first preset value；
The horizontal error of the face is less than the second preset value；And,
At least one in the extreme value of the difference in height in the face between individual features point.
A kind of electronic equipment is the embodiment of the invention provides, by adding a series of default conditions so that just
The characteristic point in face is influenceed smaller by human face expression, with precision higher.
The above is only presently preferred embodiments of the present invention, any formal limitation not made to the present invention,
Although the present invention is disclosed as above with preferred embodiment, but is not limited to the present invention, this area is common
Technical staff without departing from the scope of the present invention, when the technology contents using the disclosure above are made
It is a little change or be modified to the Equivalent embodiments of equivalent variations, as long as be without departing from technical solution of the present invention content,
Any simple modification, equivalent variations and the modification made to above example according to technical spirit of the invention,
Still fall within the range of technical solution of the present invention.
Claims (8)
1. a kind of front method of human face characteristic point, it is characterised in that methods described includes：
Obtain the two-dimensional coordinate of the characteristic point of face；
Obtain the three-dimensional coordinate of the characteristic point of default three-dimensional face model；
According to the two-dimensional coordinate of the characteristic point, the three-dimensional coordinate of the characteristic point of three-dimensional face model is preset, in advance
If formula and pre-conditioned, acquisition spin matrix, scaled matrix and translation matrix；
Two-dimensional coordinate and the spin matrix according to the characteristic point, scaled matrix and translation matrix obtain institute
State the two-dimensional coordinate of the characteristic point behind the front of face.
2. method according to claim 1, it is characterised in that described according to the two of the characteristic point
Dimension coordinate, presets the three-dimensional coordinate of the characteristic point of three-dimensional face model, preset formula and pre-conditioned, acquisition
Before spin matrix, scaled matrix and translation matrix, also include：
The two-dimensional coordinate of the characteristic point is expanded to the three-dimensional coordinate of characteristic point by way of zero padding.
3. method according to claim 1, it is characterised in that the three-dimensional face model is front
Three-dimensional face model, the three-dimensional coordinate of the characteristic point of the three-dimensional face model is the characteristic point in front
Three-dimensional coordinate.
4. method according to claim 1, it is characterised in that it is described it is pre-conditioned including：
The spin matrix is orthogonal matrix；
The axial symmetry error of the face is less than the first preset value；
The horizontal error of the face is less than the second preset value；And,
At least one in the extreme value of the difference in height in the face between individual features point.
5. a kind of electronic equipment, it is characterised in that the electronic equipment includes：
The two-dimensional coordinate acquisition module of characteristic point, the two-dimensional coordinate of the characteristic point for obtaining face；
The three-dimensional coordinate acquisition module of the characteristic point of default three-dimensional face model, for obtaining default three-dimensional face
The three-dimensional coordinate of the characteristic point of model；
Matrix acquisition module, for the two-dimensional coordinate according to the characteristic point, the spy for presetting three-dimensional face model
Levy three-dimensional coordinate a little, preset formula and pre-conditioned, acquisition spin matrix, scaled matrix and translation matrix；
Front module, for the two-dimensional coordinate according to the characteristic point and the spin matrix, scaled matrix
With the two-dimensional coordinate of the characteristic point behind the front that translation matrix obtains the face.
6. equipment according to claim 5, it is characterised in that the equipment also includes being augmented module,
Three-dimensional coordinate for the two-dimensional coordinate of the characteristic point to be expanded to characteristic point by way of zero padding.
7. equipment according to claim 5, it is characterised in that the three-dimensional face model is front
Three-dimensional face model, the three-dimensional coordinate of the characteristic point of the three-dimensional face model is the three of the characteristic point in front
Dimension coordinate.
8. equipment according to claim 5, it is characterised in that it is described it is pre-conditioned including：
The spin matrix is orthogonal matrix；
The axial symmetry error of the face is less than the first preset value；
The horizontal error of the face is less than the second preset value；And,
At least one in the extreme value of the difference in height in the face between individual features point.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511025139.2A CN106934759A (en) | 2015-12-30 | 2015-12-30 | The front method and electronic equipment of a kind of human face characteristic point |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201511025139.2A CN106934759A (en) | 2015-12-30 | 2015-12-30 | The front method and electronic equipment of a kind of human face characteristic point |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106934759A true CN106934759A (en) | 2017-07-07 |
Family
ID=59441971
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201511025139.2A Pending CN106934759A (en) | 2015-12-30 | 2015-12-30 | The front method and electronic equipment of a kind of human face characteristic point |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106934759A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111160232A (en) * | 2019-12-25 | 2020-05-15 | 上海骏聿数码科技有限公司 | Front face reconstruction method, device and system |
| CN111179174A (en) * | 2019-12-27 | 2020-05-19 | 成都品果科技有限公司 | Image stretching method and device based on face recognition points |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102262724A (en) * | 2010-05-31 | 2011-11-30 | 汉王科技股份有限公司 | Object image characteristic points positioning method and object image characteristic points positioning system |
| CN102999942A (en) * | 2012-12-13 | 2013-03-27 | 清华大学 | Three-dimensional face reconstruction method |
| CN103093215A (en) * | 2013-02-01 | 2013-05-08 | 北京天诚盛业科技有限公司 | Eye location method and device |
| US20130129159A1 (en) * | 2011-11-22 | 2013-05-23 | Ronald Huijgens | Face recognition method and apparatus |
| US20130127827A1 (en) * | 2011-11-22 | 2013-05-23 | Derek Shiell | Multiview Face Content Creation |
| CN104978764A (en) * | 2014-04-10 | 2015-10-14 | 华为技术有限公司 | Three-dimensional face mesh model processing method and three-dimensional face mesh model processing equipment |
-
2015
- 2015-12-30 CN CN201511025139.2A patent/CN106934759A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102262724A (en) * | 2010-05-31 | 2011-11-30 | 汉王科技股份有限公司 | Object image characteristic points positioning method and object image characteristic points positioning system |
| US20130129159A1 (en) * | 2011-11-22 | 2013-05-23 | Ronald Huijgens | Face recognition method and apparatus |
| US20130127827A1 (en) * | 2011-11-22 | 2013-05-23 | Derek Shiell | Multiview Face Content Creation |
| CN102999942A (en) * | 2012-12-13 | 2013-03-27 | 清华大学 | Three-dimensional face reconstruction method |
| CN103093215A (en) * | 2013-02-01 | 2013-05-08 | 北京天诚盛业科技有限公司 | Eye location method and device |
| CN104978764A (en) * | 2014-04-10 | 2015-10-14 | 华为技术有限公司 | Three-dimensional face mesh model processing method and three-dimensional face mesh model processing equipment |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111160232A (en) * | 2019-12-25 | 2020-05-15 | 上海骏聿数码科技有限公司 | Front face reconstruction method, device and system |
| CN111179174A (en) * | 2019-12-27 | 2020-05-19 | 成都品果科技有限公司 | Image stretching method and device based on face recognition points |
| CN111179174B (en) * | 2019-12-27 | 2023-11-03 | 成都品果科技有限公司 | Image stretching method and device based on face recognition points |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11715224B2 (en) | Three-dimensional object reconstruction method and apparatus | |
| WO2020181900A1 (en) | Image processing method and apparatus, image device and storage medium | |
| US20210216878A1 (en) | Deep learning-based coregistration | |
| US11151780B2 (en) | Lighting estimation using an input image and depth map | |
| WO2022116423A1 (en) | Object posture estimation method and apparatus, and electronic device and computer storage medium | |
| CN111880657A (en) | Virtual object control method and device, electronic equipment and storage medium | |
| US11398044B2 (en) | Method for face modeling and related products | |
| CN109063584B (en) | Facial feature point positioning method, device, equipment and medium based on cascade regression | |
| CN110246163A (en) | Image processing method and its device, equipment, computer storage medium | |
| US20210406579A1 (en) | Model training method, identification method, device, storage medium and program product | |
| CN106407179A (en) | Chinese character pattern similarity calculation method and system thereof | |
| WO2018107934A1 (en) | Data processing method and apparatus, and electronic device | |
| KR102500715B1 (en) | Electronic apparatus and controlling method thereof | |
| CN113635311B (en) | Method and system for out-of-hand calibration of eye for fixing calibration plate | |
| CN106934759A (en) | The front method and electronic equipment of a kind of human face characteristic point | |
| CN106934351A (en) | Gesture identification method, device and electronic equipment | |
| US20200273198A1 (en) | Method and apparatus for determining position of pupil | |
| CN107170010A (en) | System calibration method, device and three-dimensional reconstruction system | |
| CN111754391A (en) | Face correcting method, face correcting equipment and computer readable storage medium | |
| CN107203962A (en) | The method and electronic equipment of a kind of pseudo- 3D rendering of utilization 2D picture makings | |
| CN107203961A (en) | A kind of method and electronic equipment of migration of expressing one's feelings | |
| US20220198743A1 (en) | Method for generating location information, related apparatus and computer program product | |
| CN113205090B (en) | Picture correction method, device, electronic equipment and computer readable storage medium | |
| Li et al. | Head pose classification based on line portrait | |
| CN107168520A (en) | Method for tracing based on monocular cam, VR equipment and VR helmets |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170707 |