CN110415164A - Facial metamorphosis processing method, storage medium, electronic equipment and system - Google Patents
Facial metamorphosis processing method, storage medium, electronic equipment and system Download PDFInfo
- Publication number
- CN110415164A CN110415164A CN201810395695.6A CN201810395695A CN110415164A CN 110415164 A CN110415164 A CN 110415164A CN 201810395695 A CN201810395695 A CN 201810395695A CN 110415164 A CN110415164 A CN 110415164A
- Authority
- CN
- China
- Prior art keywords
- point
- value
- vec1
- vec
- abscissa
- 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
- 230000001815 facial effect Effects 0.000 title claims abstract description 29
- 230000029052 metamorphosis Effects 0.000 title claims abstract description 28
- 238000003672 processing method Methods 0.000 title claims abstract description 19
- 238000000034 method Methods 0.000 claims abstract description 27
- 230000008569 process Effects 0.000 claims abstract description 23
- 238000004364 calculation method Methods 0.000 claims description 27
- 230000003321 amplification Effects 0.000 claims description 7
- 238000003199 nucleic acid amplification method Methods 0.000 claims description 7
- 238000004590 computer program Methods 0.000 claims description 6
- 230000006870 function Effects 0.000 description 5
- 230000000694 effects Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 208000002874 Acne Vulgaris Diseases 0.000 description 1
- 206010000496 acne Diseases 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 210000004709 eyebrow Anatomy 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/18—Image warping, e.g. rearranging pixels individually
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of Facial metamorphosis processing method, storage medium, electronic equipment and systems, are related to field of video applications, and this method includes based on the key point in face key point technological orientation face picture;Chosen in the key point oriented deformation target point A (x, y) and target point deformation after where coordinate points B (x, y), calculate the vector Vec (x, y) between point A (x, y) and point B (x, y);Calculate the impact factor W of the point D (x, y) of point A (x, y) to point A (x, y) nearby;Calculate the offset vector Vec1 (x, y) of point D (x, y);According to place coordinate points E (x, y) after offset vector calculating point D (x, y) deformation process;Pixel at point E (x, y) is replaced with into the pixel at point D (x, y).The present invention freely can carry out deformation process to human face region.
Description
Technical field
The present invention relates to field of video applications, and in particular to a kind of Facial metamorphosis processing method, storage medium, electronic equipment
And system.
Background technique
With the development of internet technology, more and more people like enriching the business life of oneself by viewing live streaming
It is living, and be broadcast live in the present age more by the welcome of young man.
Now, during live streaming, substantially at standard configuration, U.S. face is referred to changing by algorithm and be regarded main broadcaster for U.S. face and U.S. type
The colour of skin of face in frequency or picture eliminates acne print etc. on face, and U.S. type refers to increasing people in video or picture by algorithm
The eyes of object or the shape of face for reducing personage, but existing U.S. face and U.S. type processing mode be only capable of it is automatically right according to preset scheme
The face of personage is handled, and flexibility is lower when in use by main broadcaster, causes to experience poor.
Summary of the invention
In view of the deficiencies in the prior art, the purpose of the present invention is to provide a kind of Facial metamorphosis processing methods, certainly
By carrying out deformation process to human face region.
To achieve the above objectives, the technical solution adopted by the present invention is that, comprising:
Based on the key point in face key point technological orientation face picture;
Chosen in the key point oriented deformation target point A (x, y) and target point deformation after where coordinate points B (x,
Y), the vector Vec (x, y) between point A (x, y) and point B (x, y) is calculated;
Calculate the impact factor W of the point D (x, y) of point A (x, y) to point A (x, y) nearby, calculation are as follows:
W=β/L2
Wherein, L2 be the distance between point D (x, y) and point A (x, y), β be impact factor coefficient, value range for [0.5,
5.0];
Calculate the offset vector Vec1 (x, y) of point D (x, y):
Vec1.x=Vec.x*W
Vec1.y=Vec.y*W
Wherein, Vec1.x is the abscissa value of offset vector, and Vec1.y is the ordinate value of offset vector;
According to place coordinate points E (x, y) after offset vector calculating point D (x, y) deformation process:
E.x=D.x-Vec1.x
E.y=D.y-Vec1.y
Wherein, E.x is the abscissa value of point E (x, y), and the ordinate value of E.y point E (x, y), D.x is the cross of point D (x, y)
Coordinate value, the ordinate value of D.y point D (x, y);
Pixel at point E (x, y) is replaced with into the pixel at point D (x, y).
Based on the above technical solution, for vector Vec (x, y), specific formula for calculation are as follows:
Vec.x=B.x-A.x
Vec.y=B.y-A.y
Wherein, Vec.x is the abscissa value of vector Vec (x, y), and Vec.y is the ordinate value of vector Vec (x, y), A.x
For the abscissa value of point A (x, y), A.y is the ordinate value of point A (x, y), and B.x is the abscissa value of point B (x, y), and B.y is point B
The ordinate value of (x, y).
Based on the above technical solution, for the distance between point D (x, y) and point A (x, y) L2, calculation formula are as follows:
L2=sqr ((A.x-D.x) * (A.x-D.x)+(A.y-D.y) * (A.y-D.y))
Wherein, sqr is evolution function, and A.x is the abscissa value of point A (x, y), and A.y is the ordinate value of point A (x, y).
Based on the above technical solution, for the impact factor W being calculated, value range is [0,1.0], and is worked as
When the value for the impact factor W being calculated is greater than 1.0, impact factor W value 1.0, when the value for the impact factor W being calculated is small
When 0, impact factor W value 0.
Based on the above technical solution,
Carrying out deformation process to the face picture of video or picture includes the processing of amplification face and the processing of thin face;
When L1 is greater than L, handled for the processing of amplification face when L1 is less than L for thin face, wherein L is point A (x, y) and point
The distance between C (x, y), L1 are the distance between point B (x, y) and point C, and point C (x, y) is that the center key point of face nose is sat
Mark, L=sqr ((A.x-C.x) * (A.x-C.x)+(A.y-C.y) * (A.y-C.y)), L1=sqr ((B.x-C.x) * (B.x-
C.x)+(B.y-C.y) * (B.y-C.y)), sqr is evolution function, and A.x is the abscissa value of point A (x, y), and A.y is point A (x, y)
Ordinate value, B.x be point B (x, y) abscissa value, B.y be point B (x, y) ordinate value, C.x for point C (x, y) cross
Coordinate value, C.y are the ordinate value of point C (x, y).
The present invention also provides a kind of storage medium, computer program, the computer program are stored on the storage medium
It is performed the steps of when being executed by processor
Based on the key point in face key point technological orientation face picture;
Chosen in the key point oriented deformation target point A (x, y) and target point deformation after where coordinate points B (x,
Y), the vector Vec (x, y) between point A (x, y) and point B (x, y) is calculated;
Calculate the impact factor W of the point D (x, y) of point A (x, y) to point A (x, y) nearby, calculation are as follows:
W=β/L2
Wherein, L2 be the distance between point D (x, y) and point A (x, y), β be impact factor coefficient, value range for [0.5,
5.0];
Calculate the offset vector Vec1 (x, y) of point D (x, y):
Vec1.x=Vec.x*W
Vec1.y=Vec.y*W
Wherein, Vec1.x is the abscissa value of offset vector, and Vec1.y is the ordinate value of offset vector;
According to place coordinate points E (x, y) after offset vector calculating point D (x, y) deformation process:
E.x=D.x-Vec1.x
E.y=D.y-Vec1.y
Wherein, E.x is the abscissa value of point E (x, y), and the ordinate value of E.y point E (x, y), D.x is the cross of point D (x, y)
Coordinate value, the ordinate value of D.y point D (x, y);
Pixel at point E (x, y) is replaced with into the pixel at point D (x, y).
The present invention also provides a kind of electronic equipment, the electronic equipment includes:
First unit, the first unit are used for based on the key point in face key point technological orientation face picture;
Second unit, the second unit are used to choose the target point A (x, y) and mesh of deformation in the key point oriented
Coordinate points B (x, y) where after punctuate deformation, calculates the vector Vec (x, y) between point A (x, y) and point B (x, y);
Third unit, the third unit be used to calculate the influence of the point D (x, y) of point A (x, y) nearby to point A (x, y) because
Sub- W, calculation are as follows:
W=β/L2
Wherein, L2 be the distance between point D (x, y) and point A (x, y), β be impact factor coefficient, value range for [0.5,
5.0];
Unit the 4th, Unit the 4th are used to calculate the offset vector Vec1 (x, y) of point D (x, y):
Vec1.x=Vec.x*W
Vec1.y=Vec.y*W
Wherein, Vec1.x is the abscissa value of offset vector, and Vec1.y is the ordinate value of offset vector;
Unit the 5th, Unit the 5th are used to calculate place coordinate points after point D (x, y) deformation process according to offset vector
E (x, y):
E.x=D.x-Vec1.x
E.y=D.y-Vec1.y
Wherein, E.x is the abscissa value of point E (x, y), and the ordinate value of E.y point E (x, y), D.x is the cross of point D (x, y)
Coordinate value, the ordinate value of D.y point D (x, y);
Unit the 6th, Unit the 6th are used to replacing with the pixel at point E (x, y) into the pixel at point D (x, y).
The present invention also provides a kind of Facial metamorphosis processing systems, comprising:
Key point locating module is used for based on the key point in face key point technological orientation face picture;
Vector calculation module, the target point A (x, y) and target point for being used to choose deformation in the key point oriented become
Coordinate points B (x, y) where after shape, calculates the vector Vec (x, y) between point A (x, y) and point B (x, y);
Impact factor computing module is used to calculate the impact factor of the point D (x, y) of point A (x, y) nearby to point A (x, y)
W, calculation are as follows:
W=β/L2
Wherein, L2 be the distance between point D (x, y) and point A (x, y), β be impact factor coefficient, value range for [0.5,
5.0];
Offset vector computing module is used to calculate the offset vector Vec1 (x, y) of point D (x, y):
Vec1.x=Vec.x*W
Vec1.y=Vec.y*W
Wherein, Vec1.x is the abscissa value of offset vector, and Vec1.y is the ordinate value of offset vector;
Deform execution module, be used to be calculated according to offset vector place coordinate points E after point D (x, y) deformation process (x,
Y), the pixel at point E (x, y) is then replaced with to the pixel at point D (x, y), in which:
E.x=D.x-Vec1.x
E.y=D.y-Vec1.y
E.x is the abscissa value of point E (x, y), and the ordinate value of E.y point E (x, y), D.x is the abscissa of point D (x, y)
Value, the ordinate value of D.y point D (x, y).
Based on the above technical solution, for vector Vec (x, y), specific formula for calculation are as follows:
Vec.x=B.x-A.x
Vec.y=B.y-A.y
Wherein, Vec.x is the abscissa value of vector Vec (x, y), and Vec.y is the ordinate value of vector Vec (x, y), A.x
For the abscissa value of point A (x, y), A.y is the ordinate value of point A (x, y), and B.x is the abscissa value of point B (x, y), and B.y is point B
The ordinate value of (x, y).
Based on the above technical solution, for the distance between point D (x, y) and point A (x, y) L2, calculation formula are as follows:
L2=sqr ((A.x-D.x) * (A.x-D.x)+(A.y-D.y) * (A.y-D.y))
Wherein, sqr is evolution function, and A.x is the abscissa value of point A (x, y), and A.y is the ordinate value of point A (x, y).
Compared with the prior art, the advantages of the present invention are as follows: selection target point, the face that artificial unrestricted choice need to deform
Position, while the location of the point coordinate that target point attachment after deformation is calculated based on vector sum offset vector, complete people
Face deformation, and freely deformation extent can be selected, strong flexibility, freedom degree is high, increases the interest of live content.
Detailed description of the invention
Fig. 1 is a kind of flow chart of Facial metamorphosis processing method in the embodiment of the present invention;
Fig. 2 is the structural schematic diagram of a kind of electronic equipment in the embodiment of the present invention.
Specific embodiment
Invention is further described in detail with reference to the accompanying drawings and embodiments.
Shown in Figure 1, the embodiment of the present invention provides a kind of Facial metamorphosis processing method, for in video or picture
Face picture carries out deformation process, and especially in live streaming field, main broadcaster uses the Facial metamorphosis processing method of the embodiment of the present invention,
Deformation process is carried out to itself face, arbitrarily stretches face, the positions such as zoom eyes, mouth, nose, uglify face or
Face is carried out exaggeration processing by person, thus to increase the interest of live content.The Facial metamorphosis processing side of the embodiment of the present invention
Method, specifically includes the following steps:
S1: based on the key point in face key point technological orientation face picture.It such as needs to become face in video
When shape processing, then using the key point of face in face key point technological orientation video, if you need to become to the face in picture
When shape processing, then using the key point of face in face key point technological orientation picture.Gone out by face key point technological orientation
Key point in face picture, be subsequent Amoebida punctuate, target point deformation recoil punctuate, face picture cheek key point,
The selection of the key points such as face picture nose key point or target point provides reference.Face key point technology can be nowadays to answer
With the crucial 68 point location technologies of more mature face, face key point technology can orient in real time cheek, eyes, eyebrow,
The position of the key points such as mouth, chin, nose.
S2: chosen in the key point oriented deformation target point A (x, y) and target point deformation after where coordinate points B
(x, y) calculates the vector Vec (x, y) between point A (x, y) and point B (x, y).I.e. firstly the need of the position artificially deformed to needs
It is chosen, and chooses locating point position after deformation, such as need to reduce cheek, then needed to choose in face picture at cheek
Point a, then some point b, this time point a again between selected point a and nose are exactly the target point chosen, and point b is that target point becomes
Coordinate points where after shape, if desired amplification stretches cheek, then needs to choose the point c in face picture at cheek, then choose again
Some point d, this time point c of the side point c and separate nose are exactly the target point chosen, and point d is place coordinate after target point deformation
Point, accordingly if you need to carry out stretching enhanced processing to eyes, then the target point chosen is the coordinate points at eyes.By choosing mesh
Coordinate points where after punctuate and target point deformation, are to be draw textured processing or contracting to some position of face with determination
Small deformation processing.
For vector Vec (x, y), specific formula for calculation are as follows:
Vec.x=B.x-A.x
Vec.y=B.y-A.y
Wherein, Vec.x is the abscissa value of vector Vec (x, y), and Vec.y is the ordinate value of vector Vec (x, y), A.x
For the abscissa value of point A (x, y), A.y is the ordinate value of point A (x, y), and B.x is the abscissa value of point B (x, y), and B.y is point B
The ordinate value of (x, y).
Carrying out deformation process to the face picture of video or image includes the processing of amplification face and the processing of thin face;When L1 is greater than
When L, for amplification face processing, when L1 be less than L when, for thin face handle, wherein L between point A (x, y) and point C (x, y) away from
From L1 is the distance between point B (x, y) and point C, and point C (x, y) is the center key point coordinate of face nose, L=sqr
((A.x-C.x) * (A.x-C.x)+(A.y-C.y) * (A.y-C.y)), L1=sqr ((B.x-C.x) * (B.x-C.x)+(B.y-
C.y) * (B.y-C.y)), sqr is evolution function, and C.x is the abscissa value of point C (x, y), and C.y is the ordinate of point C (x, y)
Value.By place coordinate points after selection target point and target point deformation, so that stretching or diminution processing flexibly are carried out to face,
Strong applicability, flexibility ratio are high.
S3: the impact factor W of the point D (x, y) of point A (x, y) to point A (x, y) nearby, calculation are calculated are as follows:
W=β/L2
Wherein, L2 be the distance between point D (x, y) and point A (x, y), β be impact factor coefficient, value range for [0.5,
5.0].For the distance between point D (x, y) and point A (x, y) L2, calculation formula are as follows: L2=sqr ((A.x-D.x) * (A.x-
D.x)+(A.y-D.y)*(A.y-D.y))。
Impact factor W is not only related with impact factor coefficient, the distance dependent between putting also and near target point and target point,
Impact factor W is directly proportional to impact factor factor beta, is inversely proportional with distance L2, and impact factor is bigger, then influence of the point A to point D be just
Bigger, impact factor is smaller, then with regard to smaller, L2 is bigger for influence of the point A to point D, and with regard to smaller, L2 is smaller for influence of the point A to point D,
Influence of the point A to point D is bigger.In the application, according to the actual situation, the value range of β is [0.5,5.0], is avoided because of β too
Small, face cheek deformation effect is unobvious, and β is too big, and face facial deformations effect is excessively exaggerated, and guarantees at the deformation effect of face
In in an acceptable zone of reasonableness.
The value of impact factor factor beta is affected to final impact factor W, if the value range of β is excessive, can make
The value of impact factor W is crossed the border, therefore for impact factor W, value range is [0,1.0], guarantees that Facial metamorphosis is in reasonable
In range, and when the value for the impact factor W being calculated is greater than 1.0, impact factor W value 1.0, when the influence being calculated
When the value of factor W is less than 0, impact factor W value 0, impact factor W calculated result then takes and is actually calculated in [0,1.0]
Value.
S4: the offset vector Vec1 (x, y) of point D (x, y) is calculated:
Vec1.x=Vec.x*W
Vec1.y=Vec.y*W
Wherein, Vec1.x is the abscissa value of offset vector, and Vec1.y is the ordinate value of offset vector;
S5: according to place coordinate points E (x, y) after offset vector calculating point D (x, y) deformation process:
E.x=D.x-Vec1.x
E.y=D.y-Vec1.y
Wherein, E.x is the abscissa value of point E (x, y), and the ordinate value of E.y point E (x, y), D.x is the cross of point D (x, y)
Coordinate value, the ordinate value of D.y point D (x, y).So far coordinate the location of has just been obtained after the point deformation near target point,
Calculating coordinate locating after all point deformation near target point, can be attached to target point after choosing target point in practical operation
Close point carries out deformation process together, point locating after deforming is calculated, therefore deformation is near selected target point
One piece of region carry out deformation process, for the point near target point, by artificially carrying out Definition of Division, with determine deformation region
Size.
S6: the pixel at point E (x, y) is replaced with into the pixel at point D (x, y), that is, by the picture at point D (x, y)
Element is moved at point E (x, y).According to the method described above, point locating after all point deformation is calculated near target point, and will
Locating point pixel replaces with the corresponding pixel that target point is nearby put before after deformation, completes deformation, shows final people
Face deformation effect.
For above-mentioned Facial metamorphosis processing method, total Facial metamorphosis formula is D=S*I, wherein S is the image of input,
I is the algorithm of the Facial metamorphosis processing method of the embodiment of the present invention, and D is Face Changing treated result images.
The Facial metamorphosis processing method of the embodiment of the present invention, selection target point, the face that artificial unrestricted choice need to deform
Position, while the location of the point coordinate that target point attachment after deformation is calculated based on vector sum offset vector, complete people
Face deformation, and freely deformation extent can be selected, strong flexibility, freedom degree is high, increases the interest of live content.
In addition, corresponding above-mentioned Facial metamorphosis processing method stores on storage medium the present invention also provides a kind of storage medium
There is computer program, is performed the steps of when computer program is executed by processor
Based on the key point in face key point technological orientation face picture;
Chosen in the key point oriented deformation target point A (x, y) and target point deformation after where coordinate points B (x,
Y), the vector Vec (x, y) between point A (x, y) and point B (x, y) is calculated;
Calculate the impact factor W of the point D (x, y) of point A (x, y) to point A (x, y) nearby, calculation are as follows:
W=β/L2
Wherein, L2 be the distance between point D (x, y) and point A (x, y), β be impact factor coefficient, value range for [0.5,
5.0];
Calculate the offset vector Vec1 (x, y) of point D (x, y):
Vec1.x=Vec.x*W
Vec1.y=Vec.y*W
Wherein, Vec1.x is the abscissa value of offset vector, and Vec1.y is the ordinate value of offset vector;
According to place coordinate points E (x, y) after offset vector calculating point D (x, y) deformation process:
E.x=D.x-Vec1.x
E.y=D.y-Vec1.y
Wherein, E.x is the abscissa value of point E (x, y), and the ordinate value of E.y point E (x, y), D.x is the cross of point D (x, y)
Coordinate value, the ordinate value of D.y point D (x, y);
Pixel at point E (x, y) is replaced with into the pixel at point D (x, y).
It should be noted that the storage medium include USB flash disk, mobile hard disk, ROM (Read-Only Memory, it is read-only to deposit
Reservoir), RAM (Random Access Memory, random access memory), magnetic or disk etc. is various can store program
The medium of code.
Shown in Figure 2, corresponding above-mentioned Facial metamorphosis processing method, the present invention also provides a kind of electronic equipment, comprising:
First unit, the first unit are used for based on the key point in face key point technological orientation face picture;
Second unit, the second unit are used to choose the target point A (x, y) and mesh of deformation in the key point oriented
Coordinate points B (x, y) where after punctuate deformation, calculates the vector Vec (x, y) between point A (x, y) and point B (x, y);
Third unit, the third unit be used to calculate the influence of the point D (x, y) of point A (x, y) nearby to point A (x, y) because
Sub- W, calculation are as follows:
W=β/L2
Wherein, L2 be the distance between point D (x, y) and point A (x, y), β be impact factor coefficient, value range for [0.5,
5.0];
Unit the 4th, Unit the 4th are used to calculate the offset vector Vec1 (x, y) of point D (x, y):
Vec1.x=Vec.x*W
Vec1.y=Vec.y*W
Wherein, Vec1.x is the abscissa value of offset vector, and Vec1.y is the ordinate value of offset vector;
Unit the 5th, Unit the 5th are used to calculate place coordinate points after point D (x, y) deformation process according to offset vector
E (x, y):
E.x=D.x-Vec1.x
E.y=D.y-Vec1.y
Wherein, E.x is the abscissa value of point E (x, y), and the ordinate value of E.y point E (x, y), D.x is the cross of point D (x, y)
Coordinate value, the ordinate value of D.y point D (x, y);
Unit the 6th, Unit the 6th are used to replacing with the pixel at point E (x, y) into the pixel at point D (x, y).
The invention also discloses a kind of Facial metamorphosis processing systems based on above-mentioned Facial metamorphosis processing method, including positioning
Key point locating module, vector calculation module, impact factor computing module, offset vector computing module and deformation execution module.
Key point locating module is used for based on the key point in face key point technological orientation face picture;Vector calculates mould
Block be used for chosen in the key point oriented deformation target point A (x, y) and target point deformation after place coordinate points B (x, y),
Calculate the vector Vec (x, y) between point A (x, y) and point B (x, y);Impact factor computing module is for calculating point A (x, y) to point
The impact factor W of the point D (x, y) of A (x, y) nearby, calculation are as follows:
W=β/L2
Wherein, L2 be the distance between point D (x, y) and point A (x, y), β be impact factor coefficient, value range for [0.5,
5.0]。
Offset vector computing module is used to calculate the offset vector Vec1 (x, y) of point D (x, y):
Vec1.x=Vec.x*W
Vec1.y=Vec.y*W
Wherein, Vec1.x is the abscissa value of offset vector, and Vec1.y is the ordinate value of offset vector;
It deforms execution module and is used to calculate place coordinate points E (x, y) after point D (x, y) deformation process according to offset vector, so
Pixel at point E (x, y) is replaced with to the pixel at point D (x, y) afterwards, in which:
E.x=D.x-Vec1.x
E.y=D.y-Vec1.y
E.x is the abscissa value of point E (x, y), and the ordinate value of E.y point E (x, y), D.x is the abscissa of point D (x, y)
Value, the ordinate value of D.y point D (x, y).
For vector Vec (x, y), specific formula for calculation are as follows:
Vec.x=B.x-A.x
Vec.y=B.y-A.y
Wherein, Vec.x is the abscissa value of vector Vec (x, y), and Vec.y is the ordinate value of vector Vec (x, y), A.x
For the abscissa value of point A (x, y), A.y is the ordinate value of point A (x, y), and B.x is the abscissa value of point B (x, y), and B.y is point B
The ordinate value of (x, y).
For the distance between point D (x, y) and point A (x, y) L2, calculation formula are as follows:
L2=sqr ((A.x-D.x) * (A.x-D.x)+(A.y-D.y) * (A.y-D.y))
Wherein, sqr is evolution function, and A.x is the abscissa value of point A (x, y), and A.y is the ordinate value of point A (x, y).
The Facial metamorphosis processing system of the embodiment of the present invention, selection target point, the face that artificial unrestricted choice need to deform
Position, while the location of the point coordinate that target point attachment after deformation is calculated based on vector sum offset vector, complete people
Face deformation, and freely deformation extent can be selected, strong flexibility, freedom degree is high, increases the interest of live content.
The present invention is not limited to the above-described embodiments, for those skilled in the art, is not departing from
Under the premise of the principle of the invention, several improvements and modifications can also be made, these improvements and modifications are also considered as protection of the invention
Within the scope of.The content being not described in detail in this specification belongs to the prior art well known to professional and technical personnel in the field.
Claims (10)
1. a kind of Facial metamorphosis processing method, which is characterized in that deformation process is carried out for the face picture to video or picture,
The following steps are included:
Based on the key point in face key point technological orientation face picture;
Chosen in the key point oriented deformation target point A (x, y) and target point deformation after where coordinate points B (x, y), meter
Calculate the vector Vec (x, y) between point A (x, y) and point B (x, y);
Calculate the impact factor W of the point D (x, y) of point A (x, y) to point A (x, y) nearby, calculation are as follows:
W=β/L2
Wherein, L2 be the distance between point D (x, y) and point A (x, y), β be impact factor coefficient, value range for [0.5,
5.0];
Calculate the offset vector Vec1 (x, y) of point D (x, y):
Vec1.x=Vec.x*W
Vec1.y=Vec.y*W
Wherein, Vec1.x is the abscissa value of offset vector, and Vec1.y is the ordinate value of offset vector;
According to place coordinate points E (x, y) after offset vector calculating point D (x, y) deformation process:
E.x=D.x-Vec1.x
E.y=D.y-Vec1.y
Wherein, E.x is the abscissa value of point E (x, y), and the ordinate value of E.y point E (x, y), D.x is the abscissa of point D (x, y)
Value, the ordinate value of D.y point D (x, y);
Pixel at point E (x, y) is replaced with into the pixel at point D (x, y).
2. a kind of Facial metamorphosis processing method as described in claim 1, it is characterised in that: for vector Vec (x, y), specifically
Calculation formula are as follows:
Vec.x=B.x-A.x
Vec.y=B.y-A.y
Wherein, Vec.x is the abscissa value of vector Vec (x, y), and Vec.y is the ordinate value of vector Vec (x, y), and A.x is point A
The abscissa value of (x, y), A.y are the ordinate value of point A (x, y), and B.x is the abscissa value of point B (x, y), and B.y is point B (x, y)
Ordinate value.
3. a kind of Facial metamorphosis processing method as described in claim 1, it is characterised in that: for point D (x, y) and point A (x, y)
The distance between L2, calculation formula are as follows:
L2=sqr ((A.x-D.x) * (A.x-D.x)+(A.y-D.y) * (A.y-D.y))
Wherein, sqr is evolution function, and A.x is the abscissa value of point A (x, y), and A.y is the ordinate value of point A (x, y).
4. a kind of Facial metamorphosis processing method as described in claim 1, it is characterised in that: for the impact factor being calculated
W, value range are [0,1.0], and when the value for the impact factor W being calculated is greater than 1.0, impact factor W value 1.0, when
When the value for the impact factor W being calculated is less than 0, impact factor W value 0.
5. a kind of Facial metamorphosis processing method as described in claim 1, it is characterised in that:
Carrying out deformation process to the face picture of video or picture includes the processing of amplification face and the processing of thin face;
When L1 is greater than L, handled for the processing of amplification face when L1 is less than L for thin face, wherein L be point A (x, y) and point C (x,
The distance between y), L1 is the distance between point B (x, y) and point C, and point C (x, y) is the center key point coordinate of face nose, L
=sqr ((A.x-C.x) * (A.x-C.x)+(A.y-C.y) * (A.y-C.y)), L1=sqr ((B.x-C.x) * (B.x-C.x)+
(B.y-C.y) * (B.y-C.y)), sqr is evolution function, and A.x is the abscissa value of point A (x, y), and A.y is the vertical of point A (x, y)
Coordinate value, B.x are the abscissa value of point B (x, y), and B.y is the ordinate value of point B (x, y), and C.x is the abscissa of point C (x, y)
Value, C.y are the ordinate value of point C (x, y).
6. a kind of storage medium, computer program is stored on the storage medium, it is characterised in that: the computer program is located
Reason device performs the steps of when executing
Based on the key point in face key point technological orientation face picture;
Chosen in the key point oriented deformation target point A (x, y) and target point deformation after where coordinate points B (x, y), meter
Calculate the vector Vec (x, y) between point A (x, y) and point B (x, y);
Calculate the impact factor W of the point D (x, y) of point A (x, y) to point A (x, y) nearby, calculation are as follows:
W=β/L2
Wherein, L2 be the distance between point D (x, y) and point A (x, y), β be impact factor coefficient, value range for [0.5,
5.0];
Calculate the offset vector Vec1 (x, y) of point D (x, y):
Vec1.x=Vec.x*W
Vec1.y=Vec.y*W
Wherein, Vec1.x is the abscissa value of offset vector, and Vec1.y is the ordinate value of offset vector;
According to place coordinate points E (x, y) after offset vector calculating point D (x, y) deformation process:
E.x=D.x-Vec1.x
E.y=D.y-Vec1.y
Wherein, E.x is the abscissa value of point E (x, y), and the ordinate value of E.y point E (x, y), D.x is the abscissa of point D (x, y)
Value, the ordinate value of D.y point D (x, y);
Pixel at point E (x, y) is replaced with into the pixel at point D (x, y).
7. a kind of electronic equipment, which is characterized in that the electronic equipment includes:
First unit, the first unit are used for based on the key point in face key point technological orientation face picture;
Second unit, the second unit are used to choose the target point A (x, y) and target point of deformation in the key point oriented
Coordinate points B (x, y) where after deformation, calculates the vector Vec (x, y) between point A (x, y) and point B (x, y);
Third unit, the third unit are used to calculate the impact factor W of the point D (x, y) of point A (x, y) nearby to point A (x, y),
Calculation are as follows:
W=β/L2
Wherein, L2 be the distance between point D (x, y) and point A (x, y), β be impact factor coefficient, value range for [0.5,
5.0];
Unit the 4th, Unit the 4th are used to calculate the offset vector Vec1 (x, y) of point D (x, y):
Vec1.x=Vec.x*W
Vec1.y=Vec.y*W
Wherein, Vec1.x is the abscissa value of offset vector, and Vec1.y is the ordinate value of offset vector;
Unit the 5th, Unit the 5th be used to be calculated according to offset vector place coordinate points E after point D (x, y) deformation process (x,
Y):
E.x=D.x-Vec1.x
E.y=D.y-Vec1.y
Wherein, E.x is the abscissa value of point E (x, y), and the ordinate value of E.y point E (x, y), D.x is the abscissa of point D (x, y)
Value, the ordinate value of D.y point D (x, y);
Unit the 6th, Unit the 6th are used to replacing with the pixel at point E (x, y) into the pixel at point D (x, y).
8. a kind of Facial metamorphosis processing system characterized by comprising
Key point locating module is used for based on the key point in face key point technological orientation face picture;
Vector calculation module, after being used to choose target point A (x, y) and the target point deformation of deformation in the key point oriented
Place coordinate points B (x, y) calculates the vector Vec (x, y) between point A (x, y) and point B (x, y);
Impact factor computing module is used to calculate the impact factor W of the point D (x, y) of point A (x, y) nearby to point A (x, y), meter
Calculation mode are as follows:
W=β/L2
Wherein, L2 be the distance between point D (x, y) and point A (x, y), β be impact factor coefficient, value range for [0.5,
5.0];
Offset vector computing module is used to calculate the offset vector Vec1 (x, y) of point D (x, y):
Vec1.x=Vec.x*W
Vec1.y=Vec.y*W
Wherein, Vec1.x is the abscissa value of offset vector, and Vec1.y is the ordinate value of offset vector;
Execution module is deformed, is used to calculate place coordinate points E (x, y) after point D (x, y) deformation process according to offset vector, so
Pixel at point E (x, y) is replaced with to the pixel at point D (x, y) afterwards, in which:
E.x=D.x-Vec1.x
E.y=D.y-Vec1.y
E.x is the abscissa value of point E (x, y), and the ordinate value of E.y point E (x, y), D.x is the abscissa value of point D (x, y), D.y
The ordinate value of point D (x, y).
9. a kind of Facial metamorphosis processing system as claimed in claim 8, it is characterised in that: for vector Vec (x, y), specifically
Calculation formula are as follows:
Vec.x=B.x-A.x
Vec.y=B.y-A.y
Wherein, Vec.x is the abscissa value of vector Vec (x, y), and Vec.y is the ordinate value of vector Vec (x, y), and A.x is point A
The abscissa value of (x, y), A.y are the ordinate value of point A (x, y), and B.x is the abscissa value of point B (x, y), and B.y is point B (x, y)
Ordinate value.
10. a kind of Facial metamorphosis processing system as claimed in claim 8, it is characterised in that: for point D (x, y) and point A (x,
The distance between y) L2, calculation formula are as follows:
L2=sqr ((A.x-D.x) * (A.x-D.x)+(A.y-D.y) * (A.y-D.y))
Wherein, sqr is evolution function, and A.x is the abscissa value of point A (x, y), and A.y is the ordinate value of point A (x, y).
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810395695.6A CN110415164A (en) | 2018-04-27 | 2018-04-27 | Facial metamorphosis processing method, storage medium, electronic equipment and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810395695.6A CN110415164A (en) | 2018-04-27 | 2018-04-27 | Facial metamorphosis processing method, storage medium, electronic equipment and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110415164A true CN110415164A (en) | 2019-11-05 |
Family
ID=68346889
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810395695.6A Pending CN110415164A (en) | 2018-04-27 | 2018-04-27 | Facial metamorphosis processing method, storage medium, electronic equipment and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110415164A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112509005A (en) * | 2020-12-10 | 2021-03-16 | 北京达佳互联信息技术有限公司 | Image processing method, image processing device, electronic equipment and storage medium |
CN114296622A (en) * | 2020-09-23 | 2022-04-08 | 北京达佳互联信息技术有限公司 | Image processing method, image processing device, electronic equipment and storage medium |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2011233073A (en) * | 2010-04-30 | 2011-11-17 | Omron Corp | Image deformation device, electronic equipment, image deformation method and image deformation program |
CN105096353A (en) * | 2014-05-05 | 2015-11-25 | 腾讯科技(深圳)有限公司 | Image processing method and device |
CN105956997A (en) * | 2016-04-27 | 2016-09-21 | 腾讯科技(深圳)有限公司 | Image deformation treatment method and device |
CN106373095A (en) * | 2016-08-29 | 2017-02-01 | 广东欧珀移动通信有限公司 | Image processing method and terminal |
-
2018
- 2018-04-27 CN CN201810395695.6A patent/CN110415164A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2011233073A (en) * | 2010-04-30 | 2011-11-17 | Omron Corp | Image deformation device, electronic equipment, image deformation method and image deformation program |
CN105096353A (en) * | 2014-05-05 | 2015-11-25 | 腾讯科技(深圳)有限公司 | Image processing method and device |
CN105956997A (en) * | 2016-04-27 | 2016-09-21 | 腾讯科技(深圳)有限公司 | Image deformation treatment method and device |
CN106373095A (en) * | 2016-08-29 | 2017-02-01 | 广东欧珀移动通信有限公司 | Image processing method and terminal |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN114296622A (en) * | 2020-09-23 | 2022-04-08 | 北京达佳互联信息技术有限公司 | Image processing method, image processing device, electronic equipment and storage medium |
CN114296622B (en) * | 2020-09-23 | 2023-08-08 | 北京达佳互联信息技术有限公司 | Image processing method, device, electronic equipment and storage medium |
CN112509005A (en) * | 2020-12-10 | 2021-03-16 | 北京达佳互联信息技术有限公司 | Image processing method, image processing device, electronic equipment and storage medium |
WO2022121577A1 (en) * | 2020-12-10 | 2022-06-16 | 北京达佳互联信息技术有限公司 | Image processing method and apparatus |
CN112509005B (en) * | 2020-12-10 | 2023-01-20 | 北京达佳互联信息技术有限公司 | Image processing method, image processing device, electronic equipment and storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108596024B (en) | Portrait generation method based on face structure information | |
CN110555796B (en) | Image adjusting method, device, storage medium and equipment | |
CN106919906B (en) | Image interaction method and interaction device | |
CN110418146A (en) | Method of changing face, storage medium, electronic equipment and system applied to live scene | |
CN102567716B (en) | Face synthetic system and implementation method | |
CN105447823A (en) | Image processing method and electronic device | |
CN110415164A (en) | Facial metamorphosis processing method, storage medium, electronic equipment and system | |
CN111985268A (en) | Method and device for driving animation by human face | |
CN109302628A (en) | A kind of face processing method based on live streaming, device, equipment and storage medium | |
CN107610239A (en) | The virtual try-in method and device of a kind of types of facial makeup in Beijing operas | |
CN111833257A (en) | Video dynamic face changing method and device, computer equipment and storage medium | |
CN110910512B (en) | Virtual object self-adaptive adjustment method, device, computer equipment and storage medium | |
CN111652795A (en) | Face shape adjusting method, face shape adjusting device, live broadcast method, live broadcast device, electronic equipment and storage medium | |
CN110910308A (en) | Image processing method, apparatus, device and medium | |
CN110415168B (en) | Face local scaling processing method, storage medium, electronic device and system | |
CN112507766B (en) | Face image extraction method, storage medium and terminal equipment | |
WO2023103813A1 (en) | Image processing method and apparatus, device, storage medium, and program product | |
CN108171671B (en) | Beautifying method and device for enlarging eye | |
CN114140828B (en) | Real-time lightweight 2D human body posture estimation method | |
CN111652023B (en) | Mouth-type adjustment and live broadcast method and device, electronic equipment and storage medium | |
CN111652792A (en) | Image local processing method, image live broadcasting method, image local processing device, image live broadcasting equipment and storage medium | |
CN112785490B (en) | Image processing method and device and electronic equipment | |
CN110020982B (en) | Cheek automatic beautifying method, storage medium, electronic device and system | |
CN113379623A (en) | Image processing method, image processing device, electronic equipment and storage medium | |
CN108833771A (en) | A kind of portrait patterned lines generation method, system and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20191105 |
|
RJ01 | Rejection of invention patent application after publication |