CN107230181A - Realize the method and device of facial image fusion - Google Patents
Realize the method and device of facial image fusion Download PDFInfo
- Publication number
- CN107230181A CN107230181A CN201710414972.9A CN201710414972A CN107230181A CN 107230181 A CN107230181 A CN 107230181A CN 201710414972 A CN201710414972 A CN 201710414972A CN 107230181 A CN107230181 A CN 107230181A
- Authority
- CN
- China
- Prior art keywords
- facial image
- signature
- source
- image
- target
- 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.)
- Granted
Links
- 230000001815 facial effect Effects 0.000 title claims abstract description 325
- 230000004927 fusion Effects 0.000 title claims abstract description 98
- 238000000034 method Methods 0.000 title claims abstract description 38
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 claims abstract description 51
- 230000009466 transformation Effects 0.000 claims abstract description 48
- 238000013507 mapping Methods 0.000 claims abstract description 25
- 238000012545 processing Methods 0.000 claims description 10
- 230000015572 biosynthetic process Effects 0.000 claims description 8
- 238000001514 detection method Methods 0.000 claims description 6
- 239000011159 matrix material Substances 0.000 claims description 6
- 238000006243 chemical reaction Methods 0.000 claims description 4
- 238000010586 diagram Methods 0.000 description 6
- 230000008859 change Effects 0.000 description 5
- 230000000694 effects Effects 0.000 description 5
- 230000008569 process Effects 0.000 description 5
- 239000000155 melt Substances 0.000 description 3
- 230000010354 integration Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000000638 solvent extraction Methods 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 238000013519 translation Methods 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
- G06T3/02—Affine transformations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The disclosure discloses a kind of method and device for realizing facial image fusion.Methods described includes：Acquisition source facial image to be fused and target facial image are simultaneously detected, after the signature for identifying source facial image and target facial image, match the signature of source facial image target facial image；Affine transformation of the signature of calculating source facial image to the signature of target facial image；According to affine transformation, fusion source facial image and target facial image form face fusion image.From the above method, by calculating the signature of source facial image to the affine transformation of the signature of target facial image, and source facial image and target facial image are merged according to affine transformation, form face fusion image, triangulation mapping is utilized so as to instead of, the technical method of face fusion image is formed, the problem of image texture is deformed can be caused after triangle subdivision mapping by solving.
Description
Technical field
This disclosure relates to image processing field, more particularly to a kind of method and device for realizing facial image fusion.
Background technology
Face fusion technology is an important application of image processing field.Current facial image integration technology passes through three
Angle partitioning techniques, i.e., be split into multiple triangles by source images and target image, calculate triangle that the subdivisions of source images goes out and
The affine transformation for the triangle that target image subdivision goes out, the Triangular Maps for being gone out the subdivision of source images according to affine transformation to mesh
On the triangle that logo image subdivision goes out, so as to form the face fusion image of source images and target image fusion.
But in above-mentioned face fusion technology, it may appear that when triangulation maps, the affine transformation of different triangles is all
It is different situation, causes during face fusion, image texture can be caused to deform after triangle subdivision mapping
Technical problem.
The content of the invention
For asking for solving present in correlation technique image texture can be caused to occur deforming after triangle subdivision mapping
Topic, present disclose provides a kind of method and device for realizing facial image fusion.
A kind of method for realizing facial image fusion, methods described includes：
Acquisition source facial image to be fused and target facial image are simultaneously detected, identify the source facial image and described
After the signature of target facial image, the signature of target facial image described in the source facial image is matched；
The signature of the source facial image is calculated to the affine transformation of the signature of the target facial image；
According to the affine transformation, the source facial image and the target facial image are merged, face fusion figure is formed
Picture.
A kind of device for realizing facial image fusion, described device includes：
Signature module, the source facial image to be fused for obtaining and target facial image are simultaneously detected, identify institute
After the signature for stating source facial image and the target facial image, target facial image described in the source facial image is matched
Signature；
Computing module, for calculating the signature of the source facial image to the signature of the target facial image
Affine transformation；
Face fusion module, for according to the affine transformation, merging the source facial image and the target face figure
Picture, forms face fusion image.
The technical scheme provided by this disclosed embodiment can include the following benefits：
Acquisition source facial image to be fused and target facial image are simultaneously detected, identify source facial image and target face
After the signature of image, the signature of source facial image target facial image is matched；The feature mark of calculating source facial image
Remember the affine transformation of the signature of target facial image；According to affine transformation, fusion source facial image and target face figure
Picture, forms face fusion image.From the above method, by calculating the signature of source facial image to target facial image
Signature affine transformation, and source facial image and target facial image are merged according to affine transformation, form face fusion
Image, so as to instead of using triangulation mapping, forms the technical method of face fusion image, solves in triangle subdivision
The problem of image texture is deformed can be caused after mapping.
It should be appreciated that the general description of the above and detailed description hereinafter are only exemplary, this can not be limited
It is open.
Brief description of the drawings
Accompanying drawing herein is merged in specification and constitutes the part of this specification, shows the implementation for meeting the present invention
Example, and in specification together for explaining principle of the invention.
Fig. 1 is a kind of flow chart of method for realizing facial image fusion according to an exemplary embodiment；
Fig. 2 is the acquisition source facial image to be fused and target facial image of Fig. 1 correspondence embodiments and detected, is identified
After the signature of source facial image and target facial image, the signature of source facial image target facial image is matched one
The flow chart of individual embodiment；
Fig. 3 is Fig. 1 correspondence embodiments according to affine transformation, fusion source facial image and target facial image, formation people
Flow chart of the face fused images in one embodiment；
Fig. 4 is a kind of device block diagram for realizing facial image fusion according to an exemplary embodiment；
Fig. 5 is block diagram of the signature module in one embodiment of Fig. 4 correspondence embodiments；
Fig. 6 is block diagram of the face fusion module in one embodiment of Fig. 4 correspondence embodiments.
Embodiment
Here explanation will be performed to exemplary embodiment in detail, its example is illustrated in the accompanying drawings.Following description is related to
During accompanying drawing, unless otherwise indicated, the same numbers in different accompanying drawings represent same or analogous key element.Following exemplary embodiment
Described in embodiment do not represent and the consistent all embodiments of the present invention.On the contrary, they be only with it is such as appended
The example of the consistent apparatus and method of some aspects be described in detail in claims, the present invention.
Fig. 1 is a kind of flow chart of method for realizing facial image fusion according to an exemplary embodiment.Such as Fig. 1
Shown, this realizes that the method for face fusion may comprise steps of.
In step 110, source facial image to be fused and target facial image are obtained and is detected, source face figure is identified
After the signature of picture and target facial image, the signature of source facial image target facial image is matched.
Wherein, source facial image is the original facial image for carrying out facial image fusion.Target facial image is to realize people
The template facial image of face image fusion.By on the basis of target facial image, source facial image being merged, so as to realize source
The fusion of facial image and target facial image.
Image frame containing active facial image and target facial image is detected, when detecting face figure therein
During picture, i.e., source facial image and target facial image respectively as acquisition.When not detecting facial image therein, then after
It is continuous to be detected, untill obtaining source facial image and target facial image.
The signature of source facial image and target facial image is that can represent source facial image and target facial image people
The identification mark of face facial characteristics.By the source facial image and the signature of target facial image identified, fusion is realized
The face fusion function of source facial image and target facial image.
In one embodiment in the specific implementation, carrying out detection to source facial image and target facial image passes through third party
Realized in storehouse, such as dlib storehouses.
By being detected to source facial image and target facial image, source facial image and target facial image are identified
Facial characteristics mark, and the coordinate of signature is recorded, so as to obtain acquisition source facial image and target facial image
Signature.
The signature of source facial image and the signature of target facial image have the mismatch of a certain feature.Example
Such as color balance mismatch, format mismatching of facial image etc..Need the signature and target person to source facial image
The signature of face image is matched, and makes the signature of source facial image consistent with the signature of target facial image,
Facial image fusion is carried out so as to the signature of the signature according to source facial image and target facial image.
In step 130, the signature of source facial image is calculated to the affine change of the signature of target facial image
Change.
Wherein, affine transformation is the affine transformation or affine maps between two vector spaces, non-strange by one
Different linear transformation connects a translation transformation composition.
According to the signature of acquisition, the signature of source facial image is reflected to the signature of target image
Penetrate, calculate the signature of source facial image to the affine transformation of the signature of target facial image.So as to according to source people
The signature of face image realizes the fusion of facial image to the result of the affine transformation of the signature of target facial image.
In step 150, according to affine transformation, fusion source facial image and target facial image form face fusion figure
Picture.
Wherein, face fusion image is the image of obtained fusion source facial image and target facial image.The face melts
Close the facial image that image is the signature formation that the signature of target facial image is replaced with to source facial image.
According to the result of obtained affine transformation, by by the spy of the signature of source facial image and target facial image
Levy mark to be merged, source facial image and target facial image are realized, so as to generate face fusion image.
This embodiment is by calculating the signature of source facial image to the affine change of the signature of target facial image
Change, and source facial image and target facial image are merged according to affine transformation, form face fusion image and solve in triangle
It can cause the problem of image texture is deformed after subdivision mapping.
Fig. 2 is that the details to step 110 according to an exemplary embodiment is described.As shown in Fig. 2 the step
110 may include following steps.
In step 111, the source facial image and target facial image got, record source facial image and target are detected
The facial characteristics coordinate of facial image, facial characteristics coordinate is further used as the source facial image identified and target facial image
Signature.
Wherein, facial characteristics coordinate is the coordinate for the picture point that facial image upper table shows face characteristic.The facial characteristics is sat
Target is used for the position for representing the facial characteristics of facial image, so as to obtain the feature mark of source facial image and target facial image
Note.
By third party library, such as dlib storehouses, detection facial image is realized, face identification functions are realized.Pass through third party
The face identification functions in storehouse, detection source facial image and target facial image, recognize source facial image and target facial image
Facial characteristics coordinate.
The source facial image of acquisition and the facial characteristics coordinate of target facial image as the source facial image identified and
The signature of target facial image.
In step 113, the signature of source facial image is partitioned into.
Wherein, the figure that the signature of source facial image obtains face fusion image as being merged with target facial image
As, it is necessary to split from the facial image of source, consequently facilitating matching with the signature of target facial image, realization is merged
Obtain the process of face fusion image.
In step 115, the color balance of the signature for the source facial image being partitioned into is adjusted, makes source facial image
The signature of signature and target facial image matches.
Wherein, the color balance of the signature of source facial image and the signature of target facial image is different to lead
The face fusion image color disunity for causing fusion to be formed, influences the effect of face fusion.
The color balance of the signature for the source facial image being partitioned into is adjusted to the feature mark with target facial image
Note is identical, the signature of the signature and target facial image of the source facial image split is matched, so as to protect
The face fusion image color that card fusion is formed is unified.
This embodiment achieves identification source facial image and the signature of target facial image, and make source facial image
The signature of signature and target facial image matches.
In one exemplary embodiment, the signature of the calculating source facial image of Fig. 1 correspondences embodiment is to target person
The affine transformation step of the signature of face image includes：
The matrix of facial characteristics coordinate formation in signature, calculates the signature of source facial image to target
The affine transformation of the signature of facial image.
Wherein, the facial characteristics coordinate of facial image is included in signature.Facial characteristics in signature
In the matrix that coordinate is constituted, the signature of source facial image is calculated to the change of the signature of target facial image
Journey, the affine transformation of the signature of the source facial image as obtained to the signature of target facial image.
According to the result of obtained affine transformation, the signature of source facial image is obtained out to the spy of target facial image
The corresponding relation of mark is levied, so as to realize the fusion of facial image according to affine transformation.
This embodiment achieves the signature of calculating source facial image to the affine of the signature of target facial image
Conversion.
Fig. 3 is the detailed description to step 150 according to an exemplary embodiment.As shown in figure 3, the step 150
It may comprise steps of.
In step 151, according to affine transformation, the signature for calculating source facial image is mapped to target facial image
Signature on mapping point.
Wherein, the signature of source facial image is mapped to realizing on the signature of target image formula is as follows.
Wherein, TI, jThe mapping point being mapped on the face fusion image formed after the signature of target image is represented,
Si,jThe facial characteristics coordinate of signature on the facial image of expression source, n represents signature number, and k represents k-th of feature
Mark, AkRepresent the signature of k-th of source facial image to the affine transformation formation of the signature of target facial image
Matrix.
ckBy function ck=f (dk) determine, the function needs to meet two conditions：
①
②
Wherein, dkRepresent Si,jThe distance between with k-th source facial image signature.Function ckAct as provide from
Si,jTo Ti,jIntense adjustment coefficient during conversion.1. specified condition is to allow each signature point to target face figure to condition
As the influence factor summation of upper mapping point is 100%.2. specified condition allows on distance objective facial image condition
The nearest signature point of mapping point has maximum factor of influence.
In step 153, according to mapping point, by the way that the signature of source facial image is mapped into target facial image
Signature, fusion source facial image and target facial image forms face fusion image.
Wherein, according to the mapping point of acquisition, the signature of source facial image is mapped to the feature mark of target image
In note, the face fusion image obtained is formed, so as to realize the fusion of source facial image and target facial image.
The signature of source facial image is mapped to the step on the signature of target image by this, is mapped by calculating
The mapping point on face fusion image formed after to the signature of target image, realizes reflecting by signature
Penetrate, the process of facial image fusion is realized, so as to instead of original triangulation mapping techniques.Due to without using triangulation
Mapping techniques, are also solved because triangulation mapping techniques cause image texture the technical problem deformed occur.
The above method is greater than triangulation mapping techniques to the amount of calculation of mapping point, but nowadays, as operation is
The reasons such as the upgrading of system so that the calculating performance of system is enhanced, so that not influenceing the situation of face fusion performance
Under, realize the function of face fusion.
This embodiment achieves fusion source facial image and target facial image, the process of face fusion image is formed.
In one exemplary embodiment, Fig. 1 correspondences embodiment according to affine transformation, fusion source facial image and target
Facial image, is formed after face fusion image step, and this realizes that the method for facial image fusion also includes：
Fusion Edges processing is carried out to the face edge of face fusion image.
Wherein, Fusion Edges are used to make the place of source facial image and the engagement of target facial image in face fused images more
Naturally, not producing breakthrough, that is, strengthen the effect of facial image fusion.
Fusion Edges processing carries out linear superposition by the edge engaged in source facial image and target facial image,
Dimension linear decay away from itself facial image is faded out, so as to reduce synthesis vestige, makes face fused images more natural, face
The effect of image co-registration is more preferable.
This embodiment carries out Fusion Edges processing by the face edge to face fusion image, enhances facial image and melts
The effect of conjunction.
Fig. 4 is a kind of device block diagram for realizing facial image fusion according to an exemplary embodiment.The device
Perform all or part of step of any shown methods for realizing facial image fusion of Fig. 1.As shown in figure 4, the device bag
Include but be not limited to：Signature module 210, computing module 230 and face fusion module 250.
The source facial image to be fused for obtaining of signature module 210 and target facial image are simultaneously detected, are identified
After the signature of source facial image and target facial image, the signature of source facial image target facial image is matched；
Computing module 230 is used to calculate the signature of source facial image to the affine of the signature of target facial image
Conversion；
Face fusion module 250 is used to, according to affine transformation, fusion source facial image and target facial image, form face
Fused images.
Fig. 5 is block diagram of the signature module in one embodiment of Fig. 4 correspondence embodiments.As shown in figure 5, this feature mark
Note module 210 includes but is not limited to：Detection unit 211, cutting unit 213 and color adjusting unit 215.
Detection unit 211 is used to detecting the source facial image and target facial image that get, record source facial image and
The facial characteristics coordinate of target facial image, facial characteristics coordinate is further used as the source facial image identified and target face
The signature of image；
Cutting unit 213 is used for the signature for being partitioned into source facial image；
Color adjusting unit 215 is used for the color balance for adjusting the signature for the source facial image being partitioned into, and makes people from source
The signature of the signature of face image and target facial image matches.
In one exemplary embodiment, the computing module of Fig. 4 correspondences embodiment includes but is not limited to：
Affine transformation computing unit, for the matrix of the facial characteristics coordinate formation in signature, calculates people from source
Affine transformation of the signature of face image to the signature of target facial image.
Fig. 6 is block diagram of the face fusion module in one embodiment of Fig. 4 correspondence embodiments.As shown in fig. 6, the face melts
Matched moulds block 250 includes but is not limited to：Coordinate calculating unit 251 and map unit 253.
Coordinate calculating unit 251 is used for according to affine transformation, and the signature for calculating source facial image is mapped to target
Mapping point on the signature of facial image；
Map unit 253 is used for according to mapping point, by the way that the signature of source facial image is mapped into target face
The signature of image, fusion source facial image and target facial image, form face fusion image.
In one exemplary embodiment, this realizes that the device of facial image fusion can also include but is not limited to：
Fusion Edges processing module, Fusion Edges processing is carried out for the face edge to face fusion image.
The function of modules and the implementation process of effect refer to the above-mentioned side for realizing facial image fusion in said apparatus
The implementation process of correspondence step, will not be repeated here in method.
It should be appreciated that the invention is not limited in the precision architecture for being described above and being shown in the drawings, and
And various modifications and changes can be being performed without departing from the scope.The scope of the present invention is only limited by appended claim.
Claims (10)
1. a kind of method for realizing facial image fusion, it is characterised in that methods described includes：
Acquisition source facial image to be fused and target facial image are simultaneously detected, identify the source facial image and the target
After the signature of facial image, the signature of target facial image described in the source facial image is matched；
The signature of the source facial image is calculated to the affine transformation of the signature of the target facial image；
According to the affine transformation, the source facial image and the target facial image are merged, face fusion image is formed.
2. according to the method described in claim 1, it is characterised in that described to obtain source facial image and target face to be fused
Image is simultaneously detected, after the signature for identifying the source facial image and the target facial image, matches the source face
The signature step of target facial image described in image includes：
The source facial image and the target facial image got is detected, the source facial image and the target is recorded
The facial characteristics coordinate of facial image, the facial characteristics coordinate is further used as the source facial image identified and described
The signature of target facial image；
It is partitioned into the signature of the source facial image；
The color balance of the signature for the source facial image being partitioned into is adjusted, makes the signature of the source facial image
Match with the signature of the target facial image.
3. according to the method described in claim 1, it is characterised in that the signature for calculating the source facial image to institute
Stating the affine transformation step of the signature of target facial image includes：
The matrix of facial characteristics coordinate formation in the signature, the signature for calculating the source facial image is arrived
The affine transformation of the signature of the target facial image.
4. according to the method described in claim 1, it is characterised in that described according to the affine transformation, merge the source face
Image and the target facial image, forming face fusion image step includes：
According to the affine transformation, the signature for calculating the source facial image is mapped to the spy of the target facial image
Levy the mapping point on mark；
According to the mapping point, by the feature mark that the signature of source facial image is mapped to the target facial image
Note, merges the source facial image and the target facial image, forms face fusion image.
5. according to the method described in claim 1, it is characterised in that described according to the affine transformation, merge the source face
Image and the target facial image, are formed after face fusion image step, methods described also includes：
Fusion Edges processing is carried out to the face edge of the face fusion image.
6. a kind of device for realizing facial image fusion, it is characterised in that described device includes：
Signature module, the source facial image to be fused for obtaining and target facial image are simultaneously detected, identify the source
After the signature of facial image and the target facial image, the spy of target facial image described in the source facial image is matched
Levy mark；
Computing module, signature the imitating to the signature of the target facial image for calculating the source facial image
Penetrate conversion；
Face fusion module, for according to the affine transformation, merging the source facial image and the target facial image, shape
Into face fused images.
7. device according to claim 6, it is characterised in that the signature module includes：
Detection unit, for detecting the source facial image and the target facial image that get, records the source face
The facial characteristics coordinate of image and the target facial image, the facial characteristics coordinate is further used as the source identified
The signature of facial image and the target facial image；
Cutting unit, the signature for being partitioned into the source facial image；
Color adjusting unit, the color balance of the signature for adjusting the source facial image being partitioned into, makes the source
The signature of the signature of facial image and the target facial image matches.
8. the device according to right wants 6, it is characterised in that the computing module includes：
Affine transformation computing unit, for the matrix of the facial characteristics coordinate formation in the signature, calculates described
Affine transformation of the signature of source facial image to the signature of the target facial image.
9. device according to claim 6, it is characterised in that the face fusion module includes：
Coordinate calculating unit, for according to the affine transformation, the signature for calculating the source facial image to be mapped to institute
State the mapping point on the signature of target facial image；
Map unit, for according to the mapping point, by the way that the signature of source facial image is mapped into the target person
The signature of face image, merges the source facial image and the target facial image, forms face fusion image.
10. device according to claim 6, it is characterised in that described device also includes：
Fusion Edges processing module, Fusion Edges processing is carried out for the face edge to the face fusion image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710414972.9A CN107230181B (en) | 2017-06-05 | 2017-06-05 | Realize the method and device of facial image fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710414972.9A CN107230181B (en) | 2017-06-05 | 2017-06-05 | Realize the method and device of facial image fusion |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107230181A true CN107230181A (en) | 2017-10-03 |
| CN107230181B CN107230181B (en) | 2018-06-29 |
Family
ID=59934798
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710414972.9A Active CN107230181B (en) | 2017-06-05 | 2017-06-05 | Realize the method and device of facial image fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107230181B (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107977946A (en) * | 2017-12-20 | 2018-05-01 | 百度在线网络技术（北京）有限公司 | Method and apparatus for handling image |
| CN108765265A (en) * | 2018-05-21 | 2018-11-06 | 北京微播视界科技有限公司 | Image processing method, device, terminal device and storage medium |
| CN108876718A (en) * | 2017-11-23 | 2018-11-23 | 北京旷视科技有限公司 | The method, apparatus and computer storage medium of image co-registration |
| CN108876705A (en) * | 2017-11-23 | 2018-11-23 | 北京旷视科技有限公司 | Image synthetic method, device and computer storage medium |
| CN109325549A (en) * | 2018-10-25 | 2019-02-12 | 电子科技大学 | A kind of facial image fusion method |
| WO2019101113A1 (en) * | 2017-11-22 | 2019-05-31 | 腾讯科技（深圳）有限公司 | Image fusion method and device, storage medium, and terminal |
| CN110033420A (en) * | 2018-01-12 | 2019-07-19 | 北京京东金融科技控股有限公司 | A kind of method and apparatus of image co-registration |
| CN110163832A (en) * | 2019-05-24 | 2019-08-23 | 北京百度网讯科技有限公司 | Face fusion method, apparatus and terminal |
| CN110276809A (en) * | 2018-03-15 | 2019-09-24 | 深圳市紫石文化传播有限公司 | Method and apparatus for face image processing |
| CN110298826A (en) * | 2019-06-18 | 2019-10-01 | 合肥联宝信息技术有限公司 | A kind of image processing method and device |
| CN112330529A (en) * | 2020-11-03 | 2021-02-05 | 上海镱可思多媒体科技有限公司 | Dlid-based face aging method, system and terminal |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009117607A1 (en) * | 2008-03-19 | 2009-09-24 | The Trustees Of Columbia University In The City Of New York | Methods, systems, and media for automatically classifying face images |
| CN102013005A (en) * | 2009-09-07 | 2011-04-13 | 泉州市铁通电子设备有限公司 | Local dynamic threshold color balance based detecting human face detection method with polarized colored light based on |
| CN103577815A (en) * | 2013-11-29 | 2014-02-12 | 中国科学院计算技术研究所 | Face alignment method and system |
| CN105069746A (en) * | 2015-08-23 | 2015-11-18 | 杭州欣禾圣世科技有限公司 | Video real-time human face substitution method and system based on partial affine and color transfer technology |
| CN106023063A (en) * | 2016-05-09 | 2016-10-12 | 西安北升信息科技有限公司 | Video transplantation face changing method |
-
2017
- 2017-06-05 CN CN201710414972.9A patent/CN107230181B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009117607A1 (en) * | 2008-03-19 | 2009-09-24 | The Trustees Of Columbia University In The City Of New York | Methods, systems, and media for automatically classifying face images |
| CN102013005A (en) * | 2009-09-07 | 2011-04-13 | 泉州市铁通电子设备有限公司 | Local dynamic threshold color balance based detecting human face detection method with polarized colored light based on |
| CN103577815A (en) * | 2013-11-29 | 2014-02-12 | 中国科学院计算技术研究所 | Face alignment method and system |
| CN105069746A (en) * | 2015-08-23 | 2015-11-18 | 杭州欣禾圣世科技有限公司 | Video real-time human face substitution method and system based on partial affine and color transfer technology |
| CN106023063A (en) * | 2016-05-09 | 2016-10-12 | 西安北升信息科技有限公司 | Video transplantation face changing method |
Non-Patent Citations (1)
| Title |
|---|
| 仪晓斌 等: "分段仿射变换下基于泊松融合的正面人脸合成", 《计算机工程与应用》 * |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11037281B2 (en) | 2017-11-22 | 2021-06-15 | Tencent Technology (Shenzhen) Company Limited | Image fusion method and device, storage medium and terminal |
| WO2019101113A1 (en) * | 2017-11-22 | 2019-05-31 | 腾讯科技（深圳）有限公司 | Image fusion method and device, storage medium, and terminal |
| CN108876718A (en) * | 2017-11-23 | 2018-11-23 | 北京旷视科技有限公司 | The method, apparatus and computer storage medium of image co-registration |
| CN108876705A (en) * | 2017-11-23 | 2018-11-23 | 北京旷视科技有限公司 | Image synthetic method, device and computer storage medium |
| CN108876705B (en) * | 2017-11-23 | 2022-03-22 | 北京旷视科技有限公司 | Image synthesis method and device and computer storage medium |
| CN108876718B (en) * | 2017-11-23 | 2022-03-22 | 北京旷视科技有限公司 | Image fusion method and device and computer storage medium |
| CN107977946A (en) * | 2017-12-20 | 2018-05-01 | 百度在线网络技术（北京）有限公司 | Method and apparatus for handling image |
| CN110033420B (en) * | 2018-01-12 | 2023-11-07 | 京东科技控股股份有限公司 | Image fusion method and device |
| CN110033420A (en) * | 2018-01-12 | 2019-07-19 | 北京京东金融科技控股有限公司 | A kind of method and apparatus of image co-registration |
| CN110276809A (en) * | 2018-03-15 | 2019-09-24 | 深圳市紫石文化传播有限公司 | Method and apparatus for face image processing |
| CN108765265A (en) * | 2018-05-21 | 2018-11-06 | 北京微播视界科技有限公司 | Image processing method, device, terminal device and storage medium |
| CN109325549B (en) * | 2018-10-25 | 2022-03-04 | 电子科技大学 | Face image fusion method |
| CN109325549A (en) * | 2018-10-25 | 2019-02-12 | 电子科技大学 | A kind of facial image fusion method |
| CN110163832B (en) * | 2019-05-24 | 2021-06-08 | 北京百度网讯科技有限公司 | Face fusion method and device and terminal |
| CN110163832A (en) * | 2019-05-24 | 2019-08-23 | 北京百度网讯科技有限公司 | Face fusion method, apparatus and terminal |
| CN110298826A (en) * | 2019-06-18 | 2019-10-01 | 合肥联宝信息技术有限公司 | A kind of image processing method and device |
| CN112330529A (en) * | 2020-11-03 | 2021-02-05 | 上海镱可思多媒体科技有限公司 | Dlid-based face aging method, system and terminal |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107230181B (en) | 2018-06-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107230181B (en) | Realize the method and device of facial image fusion | |
| US11978243B2 (en) | System and method using augmented reality for efficient collection of training data for machine learning | |
| Hassner et al. | Effective face frontalization in unconstrained images | |
| CN103530599B (en) | The detection method and system of a kind of real human face and picture face | |
| CN104077804B (en) | A kind of method based on multi-frame video picture construction three-dimensional face model | |
| US20130162643A1 (en) | Physical Three-Dimensional Model Generation Apparatus | |
| US9305206B2 (en) | Method for enhancing depth maps | |
| CN116484029A (en) | System and method for automatically generating metadata for a media document | |
| CN104021538B (en) | Object positioning method and device | |
| US20090262989A1 (en) | Image processing apparatus and method | |
| CN110096925A (en) | Enhancement Method, acquisition methods and the device of Facial Expression Image | |
| CN110443898A (en) | A kind of AR intelligent terminal target identification system and method based on deep learning | |
| CN109887003A (en) | A kind of method and apparatus initialized for carrying out three-dimensional tracking | |
| CN102646275A (en) | Method for realizing virtual three-dimensional superposition through tracking and positioning algorithms | |
| CN103646416A (en) | Three-dimensional cartoon face texture generation method and device | |
| CN111160291B (en) | Human eye detection method based on depth information and CNN | |
| CN109934873A (en) | Mark image acquiring method, device and equipment | |
| CN110120013A (en) | A kind of cloud method and device | |
| US10755476B2 (en) | Image processing method and image processing device | |
| EP2237227A1 (en) | Video sequence processing method and system | |
| CN105374043B (en) | Visual odometry filtering background method and device | |
| Zhou et al. | Single-view view synthesis with self-rectified pseudo-stereo | |
| TW201701190A (en) | Text localization system for street view image and device thereof | |
| US20220245860A1 (en) | Annotation of two-dimensional images | |
| Jiménez et al. | Face tracking and pose estimation with automatic three-dimensional model construction |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CP01 | Change in the name or title of a patent holder | ||
| CP01 | Change in the name or title of a patent holder |
Address after: Unit G03, Room 102, 22 Guanri Road, Phase II, Xiamen Software Park, Fujian Province Patentee after: Xiamen Meishao Co., Ltd. Address before: Unit G03, Room 102, 22 Guanri Road, Phase II, Xiamen Software Park, Fujian Province Patentee before: Xiamen Mei You Information technology company limited |