CN107680071A - A kind of face and the method and system of body fusion treatment - Google Patents
A kind of face and the method and system of body fusion treatment Download PDFInfo
- Publication number
- CN107680071A CN107680071A CN201710994338.7A CN201710994338A CN107680071A CN 107680071 A CN107680071 A CN 107680071A CN 201710994338 A CN201710994338 A CN 201710994338A CN 107680071 A CN107680071 A CN 107680071A
- Authority
- CN
- China
- Prior art keywords
- face
- color
- area
- fusion
- image
- 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
- 230000004927 fusion Effects 0.000 title claims abstract description 62
- 238000000034 method Methods 0.000 title claims abstract description 41
- 230000007704 transition Effects 0.000 claims abstract description 15
- 238000012546 transfer Methods 0.000 claims abstract description 10
- 238000012545 processing Methods 0.000 claims abstract description 9
- 230000001815 facial effect Effects 0.000 claims abstract description 8
- 230000008569 process Effects 0.000 claims abstract description 8
- 230000010354 integration Effects 0.000 claims description 18
- 210000003128 head Anatomy 0.000 claims description 13
- 238000005516 engineering process Methods 0.000 claims description 9
- 230000000694 effects Effects 0.000 claims description 7
- 230000011218 segmentation Effects 0.000 claims description 6
- 230000008859 change Effects 0.000 claims description 5
- 238000013527 convolutional neural network Methods 0.000 claims description 4
- 210000004209 hair Anatomy 0.000 claims description 4
- 210000000746 body region Anatomy 0.000 claims description 3
- 238000001514 detection method Methods 0.000 claims description 3
- 210000004709 eyebrow Anatomy 0.000 claims description 3
- 230000005012 migration Effects 0.000 claims description 3
- 238000013508 migration Methods 0.000 claims description 3
- 239000007787 solid Substances 0.000 claims description 2
- 239000011800 void material Substances 0.000 claims description 2
- 238000005267 amalgamation Methods 0.000 abstract description 2
- 230000008901 benefit Effects 0.000 description 4
- 238000000605 extraction Methods 0.000 description 4
- 230000008921 facial expression Effects 0.000 description 3
- 239000003086 colorant Substances 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005286 illumination Methods 0.000 description 2
- 230000007935 neutral effect Effects 0.000 description 2
- 235000008331 Pinus X rigitaeda Nutrition 0.000 description 1
- 235000011613 Pinus brutia Nutrition 0.000 description 1
- 241000018646 Pinus brutia Species 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000002844 melting Methods 0.000 description 1
- 230000008018 melting Effects 0.000 description 1
- RMUPNSINHUOUPL-UHFFFAOYSA-N n-carbamimidoyl-n-cyanoformamide Chemical compound NC(=N)N(C=O)C#N RMUPNSINHUOUPL-UHFFFAOYSA-N 0.000 description 1
- 230000035807 sensation Effects 0.000 description 1
- 238000001308 synthesis method Methods 0.000 description 1
- 238000003786 synthesis reaction Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G06V40/161—Detection; Localisation; Normalisation
- G06V40/162—Detection; Localisation; Normalisation using pixel segmentation or colour matching
-
- 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
- G06V40/161—Detection; Localisation; Normalisation
- G06V40/165—Detection; Localisation; Normalisation using facial parts and geometric relationships
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30196—Human being; Person
- G06T2207/30201—Face
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Multimedia (AREA)
- Geometry (AREA)
- Image Processing (AREA)
Abstract
The present invention provides a kind of face and the method for body fusion treatment, including, people's head region acquisition process；Mould body color transfer will be marked to face complexion color；Fusion background area is set；Carry out graph cut.The present invention proposes a kind of face and body method for amalgamation processing and system, on the one hand, because the otherness of different people face complexion is larger, and the intended body color merged is fixed, it is unnatural to be merged when solving the problems, such as the facial image of fusion and too big intended body color (brightness) difference, the method migrated using the mark mould body colour of skin to face complexion, the purpose that can reach color nature transition is pre-processed before ensureing fusion.On the other hand restricted boundary condition is passed through so that face chin and number of people region transfers pixel and human body neck neck intersecting area are merged, and remove the influence of face chin transition pixel.
Description
Technical field
The present invention relates to area of computer graphics, more particularly to a kind of face and the method and system of body fusion treatment.
Background technology
With the development of human society, clothes it is alternative more and more.In allegro life, people, which have, not to be had to
The demand that can of changing one's clothes tries on a dress, so virtual fitting product arises at the historic moment.Virtual fitting technology can substantially be divided into two kinds：
2D and 3D, model and clothes based on 3D virtual fitting technologies are that 3D makes, and advantage is that clothes and template build can exist
The upper seamless connections of 3D, and there are 360 degree of visual effects, shortcoming is that every clothes 3D scannings take time and effort.It is virtual based on 2D
For fitting technology by the way of 2D image mosaics, mark mould uses the form of 2D pictures by the way of 3D projects to 2D or directly.
Advantage is that the upper linear velocity of novel clothes is fast, and shortcoming is the effect without 360 degree of displayings.Virtual fitting technology based on 2D passes through
The image with the number of people is shot, face and hair zones image information is obtained, is spliced on preprepared mark mould body.Due to
The face area of different people shooting is influenceed by the colour of skin, illumination and other factors, it is impossible to after the number of people is spliced on body by guarantee
Colour of skin energy harmony and natural is excessive.So how to enter face area natural fusion in body image, whole human body is set to seem certainly
It is so a difficult point for being currently based on 2D virtual fittings.
On solving the problems, such as fusion, most widely used present image graphics process field is to use graph cut, that is, is solved
The method of Poisson's equation.But graph cut at combination of edge color difference it is too big when, can cause fusion after distortion the problem of.
Prior art proposes a kind of improved repeatedly Poisson image interfusion method (application publication number CN105096287 A
2015-11-25).The purpose of this method is that to solve the problems, such as that larger object region merges with Background unnatural.This method
Target image and background image are changed to HSV color spaces, Poisson image co-registration first is carried out to each passage, then extracted
Object region edge gradient information simultaneously calculates target internal subregion, to target internal sub-image area and initial fusion
Image carries out graph cut and operates to obtain final result again.The advantages of this method is the internally figure of image subsection and initial fusion
As doing graph cut again, can a given layer all on optimize internal subpicture picture confluent colours difference it is big the problem of, shortcoming is
Heterochromia can only be slowed down, but can still have the problem of heterochromia.
In addition, prior art proposes the human face countenance synthesis method (Shen based on the extraction of quick expression information and graph cut
Please the A 2016-10-26 of publication number CN 106056650), this method mainly solves prior art can not fast and effective extraction expression
Detailed information, and the problem of its information can not be synthesized to target person on the face.First, obtained correspondingly according to face feature point data
Expression template, then according to destination object expression shape template by neutral expression's image of destination object and source object it is non-in
Property facial expression image is deformed under the non-neutral expression shape of destination object.After the facial expression image block deformation of frequency domain extraction source
Expression detailed information, the source object expression details of extraction is filtered using the method for graph cut, to filtered expression
Detailed information, it is synthesized to using the method for graph cut in the deformation expression of destination object, obtains final composite result.
The advantages of this method is to ensure that grain details do not have a huge impact to fusion, source facial expression is first obtained on frequency domain
Details and removal, finally retain structural information merged.Shortcoming is not account for solving shade or intense light irradiation environment following table
The difference of feelings Fusion of Color.
Both the above prior art solves the problems, such as the fusion of different task using graph cut, be all based on by source object without
Stitch it is comprehensive be integrated into target image, do not solve the problems, such as only to intersecting area merge, people's face and body it is uncoordinated only
The problem of chin area colour of skin merges, and if only leaned on the colour of skin of chin area to intended body, chin and people can be caused again
The problem of disharmony in the other regions of face.
The content of the invention
The present invention solves the problems, such as that the face chin colour of skin and mark mould body neck neck region colour of skin transition are unnatural：Due to not
Face complexion and photoenvironment difference of taking pictures with people, cause different face complexions can not fix the body colour of skin with mark mould natural
Transition；Because the number of people image taken out has transition pixel of the number of people to background, if transition pixel is directly adhered directly onto mark
It can seem on mould body very lofty, so the processing to number of people image transition pixel is very necessary；If simply using graph cut,
Only face chin area is merged with mark mould body neck area, the change of chin color can be with the other field colors of face
Have a long way to go, cause unnatural effect.
In view of the above-mentioned problems, a kind of face of the present invention and the method for body fusion treatment, including,
People's head region acquisition process；
Mould body color transfer will be marked to face complexion color；
Fusion background area is set；
Carry out graph cut.
Further, described pending human body shooting image obtains the number of people with hair by number of people cutting techniques
Area image.
Further, described number of people cutting techniques, realized based on number of people segmentation convolutional neural networks, the volume of number of people segmentation
Product neural network sample, based on mark number of people area image.
Further, it is described to realize that the colour of skin migrates by the way of pixel color passage average is alignd with variance, including：
Obtained by face key point and remove the correct face complexion area of face region searching；
Using Face datection and face critical point detection technology, according to face outline key point, human face region is obtained.
Further, described human face region includes, the non-area of skin color such as eyes, eyebrow, face, to take underthe nose to arrange
Except the region of face, that is, remove average of the chin area of face as face skin area A, respectively calculating RGB triple channels
AMeanR, AMeanG, AMeanB and variance AVarR, AVarG, AVarB；
It is area of skin color to mark mould body region, is designated as B, equally calculates the RGB threeways of mark mould body area of skin color respectively
The average BMeanR, BMeanG, BMeanB and variance BVarR, BVarG, BVarB in road；
Migrate area of skin color of human body B each pixel P as follows：
PR ,=AVarR/BVarR* (PR-BMeanR)+AMeanR
PG ,=AVarG/BVarG* (PG-BMeanG)+AMeanG
PB ,=AVarB/BVarB* (PB-BMeanB)+AMeanB
Wherein PR,Pixel value corresponding to representing area of skin color of human body some pixel of B P new R passages, that is, correspond to the colour of skin
Pixel value after migration.PR represents the original R passages pixel values of area of skin color of human body some pixel of B P, by 3 passages
Each pixel value of area of skin color of human body is migrated respectively so that the color of area of skin color of human body is close proximity to face complexion area.
Further, the principle of described graph cut, ensure the constant premise of border color, minimize in integration region
The change of Grad, i.e. below equation：
Wherein, f is integration region, and Ω is border,Merged after representing fusion
The gradient in region, v represent the gradient of integration region before fusion, f*For background area；According to formula, graph cut must is fulfilled for melting
Border color and background color after conjunction are consistent, and are merged front and rear integration region gradient and kept constant as far as possible, corresponding circle of sensation
It is to ensure that the minutia of fused images is constant that domain gradient, which keeps constant, and the color of integration region and background area are on border
Solid colour is to ensure that transition is natural.
Further, described graph cut, including, foreground image, background image, the Mask of foreground area to be fused
The position of background area where mask and Mask masks, wherein foreground image are facial image to be fused, background image
For the fusion background image of making, Mask masks image is entirely stingy tribal chief's face image, the position of background area where Mask masks
Put, as make the position that fusion background places facial image.
Further, described graph cut is to meet the Mask of Poisson's equation condition in tri- path computations of RGB respectively
The pixel value in mask region, by graph cut, the edge transition of head image is scratched at face edge, and pixel fusion is entered to mark under mould body
Bar region, meanwhile, other regions beyond face chin are without because fusion produces cross-color effect.
The present invention provides a kind of face and the system of body fusion treatment, including,
People's head region acquisition process module；
Mould body color transfer will be marked to face complexion color module；
Fusion background area module is set；
Carry out graph cut module.
The present invention provides a kind of face and the product of body fusion treatment, including suitable for virtual reality, virtual fitting, void
Intend social, U.S. figure, image restoring etc..
Beneficial effect
The present invention proposes a kind of face and body method for amalgamation processing and system, on the one hand, due to different people face complexion
Otherness it is larger, and merge intended body color fix, for solve fusion facial image and intended body color it is (bright
Degree) difference it is too big when merge the problem of unnatural, using the method that migrate to face complexion of the mark mould body colour of skin, ensure before merging
Pretreatment can reach the purpose of color nature transition.On the other hand restricted boundary condition is passed through so that face chin and the number of people
Region transfers pixel and human body neck neck intersecting area are merged, and remove the influence of face chin transition pixel.
Brief description of the drawings
The original sticking effect schematic diagrames of Fig. 1
Fig. 2 merges background image schematic diagram
Fig. 3 fusion results figures
Original sticking effect refer to directly by the number of people image after stingy head pastes mark mould body on, it can be seen that the number of people with
Chin linking is unnatural.
It is that step 3 makes background area image to merge background image, it is therefore an objective to sets boundary condition for fusion.
It is smooth naturally many that facial image face chin after fusion is substantially transitioned into mark mould body.
Embodiment
The embodiment of the present invention provides a kind of face and the method for body fusion treatment, including,
People's head region acquisition process；
Mould body color transfer will be marked to face complexion color；
Fusion background area is set；
Carry out graph cut.
Preferred embodiment, pending human body shooting image is carried by number of people cutting techniques, acquisition in the embodiment of the present invention
The number of people area image of hair.
Preferred embodiment, number of people cutting techniques in the present embodiment, realized based on number of people segmentation convolutional neural networks, preferably
Embodiment, realize that the colour of skin migrates in the present embodiment by the way of pixel color passage average is alignd with variance, including：
Obtained by face key point and remove the correct face complexion area of face region searching；
Using Face datection and face critical point detection technology, according to face outline key point, human face region is obtained.
Preferred embodiment, human face region includes in the present embodiment, the non-area of skin color such as eyes, eyebrow, face, to take nose
Lower section excludes the region of face, that is, removes the chin area of face as face skin area A, calculate RGB triple channels respectively
Average AMeanR, AMeanG, AMeanB and variance AVarR, AVarG, AVarB；
It is area of skin color to mark mould body region, is designated as B, equally calculates the RGB threeways of mark mould body area of skin color respectively
The average BMeanR, BMeanG, BMeanB and variance BVarR, BVarG, BVarB in road；
Migrate area of skin color of human body B each pixel P as follows：
PR ,=AVarR/BVarR* (PR-BMeanR)+AMeanR
PG ,=AVarG/BVarG* (PG-BMeanG)+AMeanG
PB ,=AVarB/BVarB* (PB-BMeanB)+AMeanB
Wherein PR, pixel value corresponding to expression area of skin color of human body some pixel of B P new R passages, that is, correspond to the colour of skin
Pixel value after migration.PR represents the original R passages pixel values of area of skin color of human body some pixel of B P, by 3 passages
Each pixel value of area of skin color of human body is migrated respectively so that the color of area of skin color of human body is close proximity to face complexion area.
Preferred embodiment, the principle of graph cut in the present embodiment, ensure the constant premise of border color, minimize fusion
The change of region manhole ladder angle value, i.e. below equation：
Wherein, f is integration region, and Ω is border, and ▽ f represent the gradient of integration region after fusion, and v represents to melt
The gradient of integration region, f before conjunction*For background area；According to formula, graph cut must is fulfilled for the border after fusion
Color and background color keep is consistent, and merges front and rear integration region gradient and keep constant as far as possible, and integration region gradient is kept
Constant is to ensure that the minutia of fused images is constant, and the color of integration region and back of the body preferred embodiment, is moored in the present embodiment
Pine fusion, including, foreground image, background image, the Mask masks of foreground area to be fused and background where Mask masks
The position in region, wherein foreground image are facial image to be fused, and background image is the fusion background image made, Mask
Mask image is entirely stingy tribal chief's face image, the position of background area where Mask masks, as makes fusion background and places people
The position of face image：Because people's head region to be fused and mark mould body are two separated images, so must be first by them
An image is synthesized into, the image of synthesis is as shown in Fig. 2 fusion background image in accompanying drawing.According to mark mould body and number of people size
Meet fixed proportion, by number of people image scaling to mark mould body sizes size, and by number of people image according to the number of people in neck neck experience
Position, mark mould body neck neck region is put into advance, and merge background image in mark mould body neck neck region and number of people image weight
Close region, i.e. neck neck and chin overlapping region, display mark mould body neck neck region；The purpose for making fusion background area is limitation
Boundary condition is merged, the number of people for removing chin area merges border color with keeping constant before, the border of number of people chin area
Color is on the basis of marking mould body color.Setting this reason for merging boundary condition is：According to graph cut principle, border face
Color keeps constant, and fusion inner gradient change keeps constant benchmark as far as possible, so the border color of face chin area is changed into
Mark mould body neck neck field color, and to keep gradient keep it is constant in the case of, only face chin integral color to mark mould
Body color is leaned on；Mark mould body color due to having done step 2 is moved to face complexion color transfer so finishing the colour of skin
The border color and face chin border color gap of mark mould body after shifting are little, on the one hand in order to increase in uneven illumination ring
The face complexion Shandong nation property taken pictures under border, on the other hand in order to solve the problems, such as that the transition pixel of stingy tribal chief's head region is unnatural,
So also need to do graph cut processing as the border color of face chin using the color for marking mould body neck neck region.
Preferred embodiment, graph cut is to meet Poisson's equation condition in tri- path computations of RGB respectively in the present embodiment
Mask masks region pixel value, by graph cut, the edge transition of head image is scratched at face edge, and pixel fusion is entered to mark mould
Body chin area, meanwhile, other regions beyond face chin are without because fusion produces cross-color effect.
The embodiment of the present invention provides a kind of face and the system of body fusion treatment, including,
People's head region acquisition process module；
Mould body color transfer will be marked to face complexion color module；
Fusion background area module is set；
Carry out graph cut module.
The embodiment of the present invention provides a kind of face and the product of body fusion treatment, including suitable for virtual reality, virtual
Fitting, virtual social, U.S. figure, image restoring etc..
Claims (10)
1. a kind of face and the method for body fusion treatment, it is characterised in that including,
People's head region acquisition process；
Mould body color transfer will be marked to face complexion color；
Fusion background area is set；
Carry out graph cut.
2. a kind of face as claimed in claim 1 and the method for body fusion treatment, it is characterised in that described pending people
Body shooting image obtains the number of people area image with hair by number of people cutting techniques.
3. the method that a kind of face as claimed in claim 1 integrates processing with body, it is characterised in that described number of people segmentation
Technology, realized based on number of people segmentation convolutional neural networks, the convolutional neural networks sample of number of people segmentation, based on mark people's head region
Image.
4. the method that a kind of face as claimed in claim 1 integrates processing with body, described use pixel color passage are equal
The mode that value is alignd with variance realizes that the colour of skin migrates, including：
Obtained by face key point and remove the correct face complexion area of face region searching；
Using Face datection and face critical point detection technology, according to face outline key point, human face region is obtained.
5. the method that a kind of face as claimed in claim 4 integrates processing with body, described human face region include, eyes,
The non-area of skin color such as eyebrow, face, to take underthe nose to exclude the region of face, that is, the chin area of face is removed as face
Skin area A, the average AMeanR, AMeanG, AMeanB and variance AVarR, AVarG, AVarB of RGB triple channels are calculated respectively；
It is area of skin color to mark mould body region, is designated as B, equally calculates the RGB triple channels of mark mould body area of skin color respectively
Average BMeanR, BMeanG, BMeanB and variance BVarR, BVarG, BVarB；
Migrate area of skin color of human body B each pixel P as follows：
PR '=AVarR/BVarR* (PR-BMeanR)+AMeanR
PG '=AVarG/BVarG* (PG-BMeanG)+AMeanG
PB '=AVarB/BVarB* (PB-BMeanB)+AMeanB
Pixel value corresponding to wherein PR ' expression area of skin color of human body some pixel of B P new R passages, that is, correspond to colour of skin migration
Pixel value afterwards.PR represents the original R passages pixel values of area of skin color of human body some pixel of B P, by distinguishing on 3 passages
Migrate each pixel value of area of skin color of human body so that the color of area of skin color of human body is close proximity to face complexion area.
6. the method that a kind of face as claimed in claim 1 integrates processing with body, it is characterised in that described graph cut
Principle, ensure the constant premise of border color, minimize the change of integration region manhole ladder angle value, i.e. below equation：
<mrow>
<munder>
<mrow>
<mi>m</mi>
<mi>i</mi>
<mi>n</mi>
</mrow>
<mi>f</mi>
</munder>
<mo>&Integral;</mo>
<msub>
<mo>&Integral;</mo>
<mi>&Omega;</mi>
</msub>
<mo>|</mo>
<mo>&dtri;</mo>
<mi>f</mi>
<mo>-</mo>
<mi>v</mi>
<msup>
<mo>|</mo>
<mn>2</mn>
</msup>
<mi>w</mi>
<mi>i</mi>
<mi>t</mi>
<mi>h</mi>
<mi>f</mi>
<msub>
<mo>|</mo>
<mrow>
<mo>&part;</mo>
<mi>&Omega;</mi>
</mrow>
</msub>
<mo>=</mo>
<msup>
<mi>f</mi>
<mo>*</mo>
</msup>
<msub>
<mo>|</mo>
<mrow>
<mo>&part;</mo>
<mi>&Omega;</mi>
</mrow>
</msub>
</mrow>
Wherein, f is integration region, and Ω is border, and ▽ f represent the gradient of integration region after fusion, and v represents integration region before fusion
Gradient, f*For background area；According to formula, graph cut must is fulfilled for the border color after fusion and background color keeps one
Cause, and merge front and rear integration region gradient and keep constant as far as possible, it is to ensure fused images that integration region gradient, which keeps constant,
Minutia is constant, and the color of integration region is to ensure that transition is natural with background area solid colour on border.
7. the method that a kind of face as claimed in claim 1 integrates processing with body, described graph cut, including, prospect
Image, background image, the Mask masks of foreground area to be fused and the position of background area where Mask masks, wherein before
Scape image is facial image to be fused, and background image is the fusion background image made, and Mask masks image is whole stingy
Tribal chief's face image, the position of background area where Mask masks, as make the position that fusion background places facial image.
8. a kind of method of people's face and body fusion treatment as claimed in claim 7, described graph cut is respectively in RGB
Three path computations meet the pixel value in the Mask masks region of Poisson's equation condition, and by graph cut, head is scratched at face edge
The edge transition of image, pixel fusion enter to mark mould body chin area, meanwhile, other regions beyond face chin without because
Fusion produces cross-color effect.
9. a kind of face and the system of body fusion treatment, it is characterised in that including,
People's head region acquisition process module；
Mould body color transfer will be marked to face complexion color module；
Fusion background area module is set；
Carry out graph cut module.
10. a kind of face and the product of body fusion treatment, it is characterised in that including suitable for virtual reality, virtual fitting, void
Intend social, beautiful figure, image restoring etc., it is characterised in that the face and the product of body fusion treatment are claim 1 to 8
A kind of face and the method and system of body fusion treatment described in middle any one.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710994338.7A CN107680071B (en) | 2017-10-23 | 2017-10-23 | Method and system for fusion processing of human face and human body |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710994338.7A CN107680071B (en) | 2017-10-23 | 2017-10-23 | Method and system for fusion processing of human face and human body |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107680071A true CN107680071A (en) | 2018-02-09 |
| CN107680071B CN107680071B (en) | 2020-08-07 |
Family
ID=61141438
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710994338.7A Active CN107680071B (en) | 2017-10-23 | 2017-10-23 | Method and system for fusion processing of human face and human body |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107680071B (en) |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108596839A (en) * | 2018-03-22 | 2018-09-28 | 中山大学 | A kind of human-face cartoon generation method and its device based on deep learning |
| CN108764143A (en) * | 2018-05-29 | 2018-11-06 | 北京字节跳动网络技术有限公司 | Image processing method, device, computer equipment and storage medium |
| CN108932735A (en) * | 2018-07-10 | 2018-12-04 | 广州众聚智能科技有限公司 | A method of generating deep learning sample |
| CN109376618A (en) * | 2018-09-30 | 2019-02-22 | 北京旷视科技有限公司 | Image processing method, device and electronic equipment |
| CN109615593A (en) * | 2018-11-29 | 2019-04-12 | 北京市商汤科技开发有限公司 | Image processing method and device, electronic equipment and storage medium |
| CN109784301A (en) * | 2019-01-28 | 2019-05-21 | 广州酷狗计算机科技有限公司 | Image processing method, device, computer equipment and storage medium |
| CN109949207A (en) * | 2019-01-31 | 2019-06-28 | 深圳市云之梦科技有限公司 | Virtual objects synthetic method, device, computer equipment and storage medium |
| CN110069125A (en) * | 2018-09-21 | 2019-07-30 | 北京微播视界科技有限公司 | The control method and device of virtual objects |
| CN110084744A (en) * | 2019-03-06 | 2019-08-02 | 深圳市云之梦科技有限公司 | Image processing method, device, computer equipment and storage medium |
| CN110348496A (en) * | 2019-06-27 | 2019-10-18 | 广州久邦世纪科技有限公司 | A kind of method and system of facial image fusion |
| CN110490029A (en) * | 2018-05-15 | 2019-11-22 | 瑞昱半导体股份有限公司 | The image treatment method of differentiation processing can be done to face data |
| CN111612897A (en) * | 2020-06-05 | 2020-09-01 | 腾讯科技（深圳）有限公司 | Three-dimensional model fusion method, device and equipment and readable storage medium |
| CN111654622A (en) * | 2020-05-28 | 2020-09-11 | 维沃移动通信有限公司 | Shooting focusing method and device, electronic equipment and storage medium |
| CN112949360A (en) * | 2019-12-11 | 2021-06-11 | 广州市久邦数码科技有限公司 | Video face changing method and device |
| CN112990134A (en) * | 2021-04-29 | 2021-06-18 | 北京世纪好未来教育科技有限公司 | Image simulation method and device, electronic equipment and storage medium |
| CN113409329A (en) * | 2021-06-03 | 2021-09-17 | Oppo广东移动通信有限公司 | Image processing method, image processing apparatus, terminal, and readable storage medium |
| CN113781292A (en) * | 2021-08-23 | 2021-12-10 | 北京达佳互联信息技术有限公司 | Image processing method and device, electronic device and storage medium |
| CN113808003A (en) * | 2020-06-17 | 2021-12-17 | 北京达佳互联信息技术有限公司 | Training method of image processing model, image processing method and device |
| CN113870404A (en) * | 2021-09-23 | 2021-12-31 | 聚好看科技股份有限公司 | Skin rendering method and device of 3D model |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103607554A (en) * | 2013-10-21 | 2014-02-26 | 无锡易视腾科技有限公司 | Fully-automatic face seamless synthesis-based video synthesis method |
| 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 |
-
2017
- 2017-10-23 CN CN201710994338.7A patent/CN107680071B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103607554A (en) * | 2013-10-21 | 2014-02-26 | 无锡易视腾科技有限公司 | Fully-automatic face seamless synthesis-based video synthesis method |
| 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 |
Non-Patent Citations (2)
| Title |
|---|
| PATRICK P´EREZ ET AL.: "Poisson image editing", 《ACM TRANSACTIONS ON GRAPHICS》 * |
| 周漾 等: "面泊松融合结合色彩变换的无缝纹理辐射处理", 《中国图象圈形学报》 * |
Cited By (30)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108596839A (en) * | 2018-03-22 | 2018-09-28 | 中山大学 | A kind of human-face cartoon generation method and its device based on deep learning |
| CN110490029B (en) * | 2018-05-15 | 2022-04-15 | 瑞昱半导体股份有限公司 | Image processing method capable of performing differentiation processing on face data |
| CN110490029A (en) * | 2018-05-15 | 2019-11-22 | 瑞昱半导体股份有限公司 | The image treatment method of differentiation processing can be done to face data |
| CN108764143A (en) * | 2018-05-29 | 2018-11-06 | 北京字节跳动网络技术有限公司 | Image processing method, device, computer equipment and storage medium |
| CN108764143B (en) * | 2018-05-29 | 2020-11-24 | 北京字节跳动网络技术有限公司 | Image processing method, image processing device, computer equipment and storage medium |
| CN108932735A (en) * | 2018-07-10 | 2018-12-04 | 广州众聚智能科技有限公司 | A method of generating deep learning sample |
| CN110069125A (en) * | 2018-09-21 | 2019-07-30 | 北京微播视界科技有限公司 | The control method and device of virtual objects |
| CN110069125B (en) * | 2018-09-21 | 2023-12-22 | 北京微播视界科技有限公司 | Virtual object control method and device |
| CN109376618A (en) * | 2018-09-30 | 2019-02-22 | 北京旷视科技有限公司 | Image processing method, device and electronic equipment |
| CN109615593A (en) * | 2018-11-29 | 2019-04-12 | 北京市商汤科技开发有限公司 | Image processing method and device, electronic equipment and storage medium |
| CN109784301A (en) * | 2019-01-28 | 2019-05-21 | 广州酷狗计算机科技有限公司 | Image processing method, device, computer equipment and storage medium |
| CN109949207A (en) * | 2019-01-31 | 2019-06-28 | 深圳市云之梦科技有限公司 | Virtual objects synthetic method, device, computer equipment and storage medium |
| CN109949207B (en) * | 2019-01-31 | 2023-01-10 | 深圳市云之梦科技有限公司 | Virtual object synthesis method and device, computer equipment and storage medium |
| CN110084744A (en) * | 2019-03-06 | 2019-08-02 | 深圳市云之梦科技有限公司 | Image processing method, device, computer equipment and storage medium |
| CN110084744B (en) * | 2019-03-06 | 2022-11-08 | 深圳市云之梦科技有限公司 | Image processing method, image processing device, computer equipment and storage medium |
| CN110348496A (en) * | 2019-06-27 | 2019-10-18 | 广州久邦世纪科技有限公司 | A kind of method and system of facial image fusion |
| CN110348496B (en) * | 2019-06-27 | 2023-11-14 | 广州久邦世纪科技有限公司 | Face image fusion method and system |
| CN112949360A (en) * | 2019-12-11 | 2021-06-11 | 广州市久邦数码科技有限公司 | Video face changing method and device |
| CN111654622A (en) * | 2020-05-28 | 2020-09-11 | 维沃移动通信有限公司 | Shooting focusing method and device, electronic equipment and storage medium |
| CN111612897B (en) * | 2020-06-05 | 2023-11-10 | 腾讯科技（深圳）有限公司 | Fusion method, device and equipment of three-dimensional model and readable storage medium |
| CN111612897A (en) * | 2020-06-05 | 2020-09-01 | 腾讯科技（深圳）有限公司 | Three-dimensional model fusion method, device and equipment and readable storage medium |
| CN113808003A (en) * | 2020-06-17 | 2021-12-17 | 北京达佳互联信息技术有限公司 | Training method of image processing model, image processing method and device |
| CN113808003B (en) * | 2020-06-17 | 2024-02-09 | 北京达佳互联信息技术有限公司 | Training method of image processing model, image processing method and device |
| CN112990134B (en) * | 2021-04-29 | 2021-08-20 | 北京世纪好未来教育科技有限公司 | Image simulation method and device, electronic equipment and storage medium |
| CN112990134A (en) * | 2021-04-29 | 2021-06-18 | 北京世纪好未来教育科技有限公司 | Image simulation method and device, electronic equipment and storage medium |
| CN113409329A (en) * | 2021-06-03 | 2021-09-17 | Oppo广东移动通信有限公司 | Image processing method, image processing apparatus, terminal, and readable storage medium |
| CN113409329B (en) * | 2021-06-03 | 2023-11-14 | Oppo广东移动通信有限公司 | Image processing method, image processing device, terminal and readable storage medium |
| CN113781292A (en) * | 2021-08-23 | 2021-12-10 | 北京达佳互联信息技术有限公司 | Image processing method and device, electronic device and storage medium |
| CN113870404A (en) * | 2021-09-23 | 2021-12-31 | 聚好看科技股份有限公司 | Skin rendering method and device of 3D model |
| CN113870404B (en) * | 2021-09-23 | 2024-05-07 | 聚好看科技股份有限公司 | Skin rendering method of 3D model and display equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107680071B (en) | 2020-08-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107680071A (en) | A kind of face and the method and system of body fusion treatment | |
| CN102663766B (en) | Non-photorealistic based art illustration effect drawing method | |
| CN103914863B (en) | A kind of coloured image abstract method for drafting | |
| RU2019101747A (en) | DIGITAL MAKEUP MIRROR SYSTEM AND METHOD FOR ITS IMPLEMENTATION | |
| EP1030267B1 (en) | Method of correcting face image, makeup simulation method, makeup method, makeup supporting device and foundation transfer film | |
| US8325205B2 (en) | Methods and files for delivering imagery with embedded data | |
| CN103473780B (en) | The method of portrait background figure a kind of | |
| CN107993216A (en) | A kind of image interfusion method and its equipment, storage medium, terminal | |
| CN103942794B (en) | A kind of image based on confidence level is collaborative scratches drawing method | |
| CN102831584B (en) | Data-driven object image restoring system and method | |
| CN103258343B (en) | A kind of eyes image disposal route based on picture editting | |
| CN109829930A (en) | Face image processing process, device, computer equipment and readable storage medium storing program for executing | |
| CN102903135A (en) | Method and apparatus for realistic simulation of wrinkle aging and de-aging | |
| JPH11224329A (en) | Photo booth for forming digitally processed image | |
| JP2001057630A (en) | Image processing unit and image processing method | |
| CN104063888B (en) | A kind of wave spectrum artistic style method for drafting based on feeling of unreality | |
| Ye et al. | Hybrid scheme of image’s regional colorization using mask r-cnn and Poisson editing | |
| CN106940792A (en) | The human face expression sequence truncation method of distinguished point based motion | |
| Seo et al. | Interactive painterly rendering with artistic error correction | |
| JP2007144194A (en) | Method for face image modification, method for makeup simulation, method for makeup, support equipment for makeup and cosmetic foundation transcription film | |
| CN109345470A (en) | Facial image fusion method and system | |
| Wang et al. | Facial image composition based on active appearance model | |
| CN102355555B (en) | Video processing method and system | |
| Doyle et al. | Painted stained glass | |
| KR101340936B1 (en) | Pop art portraiture |
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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20231127 Address after: Gao Lou Zhen Hong Di Cun, Rui'an City, Wenzhou City, Zhejiang Province, 325200 Patentee after: Wang Conghai Address before: 10 / F, Yihua financial technology building, 2388 Houhai Avenue, high tech park, Yuehai street, Nanshan District, Shenzhen, Guangdong 518000 Patentee before: SHENZHEN CLOUDREAM INFORMATION TECHNOLOGY CO.,LTD. |