CN107506559A - Star's face shaping based on human face similarity degree analysis, which is made up, recommends method and apparatus - Google Patents
Star's face shaping based on human face similarity degree analysis, which is made up, recommends method and apparatus Download PDFInfo
- Publication number
- CN107506559A CN107506559A CN201710804911.3A CN201710804911A CN107506559A CN 107506559 A CN107506559 A CN 107506559A CN 201710804911 A CN201710804911 A CN 201710804911A CN 107506559 A CN107506559 A CN 107506559A
- Authority
- CN
- China
- Prior art keywords
- face
- star
- shaping
- local
- similarity
- 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
- 238000000034 method Methods 0.000 title claims abstract description 67
- 238000007493 shaping process Methods 0.000 title claims abstract description 43
- 238000004458 analytical method Methods 0.000 title claims abstract description 28
- 230000001815 facial effect Effects 0.000 claims abstract description 44
- 239000002537 cosmetic Substances 0.000 claims abstract description 32
- 230000008569 process Effects 0.000 claims abstract description 27
- 230000000694 effects Effects 0.000 claims abstract description 18
- 210000000056 organ Anatomy 0.000 claims description 47
- 210000000887 face Anatomy 0.000 claims description 20
- 210000001508 eye Anatomy 0.000 claims description 17
- 210000004709 eyebrow Anatomy 0.000 claims description 12
- 230000000007 visual effect Effects 0.000 claims description 12
- 238000010606 normalization Methods 0.000 claims description 11
- 210000001061 forehead Anatomy 0.000 claims description 10
- 241000255789 Bombyx mori Species 0.000 claims description 6
- 230000003416 augmentation Effects 0.000 claims description 6
- 239000006071 cream Substances 0.000 claims description 6
- 210000000720 eyelash Anatomy 0.000 claims description 6
- 210000000744 eyelid Anatomy 0.000 claims description 6
- 238000005286 illumination Methods 0.000 claims description 6
- JEIPFZHSYJVQDO-UHFFFAOYSA-N iron(III) oxide Inorganic materials O=[Fe]O[Fe]=O JEIPFZHSYJVQDO-UHFFFAOYSA-N 0.000 claims description 6
- 210000001331 nose Anatomy 0.000 claims description 6
- 238000002435 rhinoplasty Methods 0.000 claims description 6
- 238000013461 design Methods 0.000 claims description 5
- 238000001303 quality assessment method Methods 0.000 claims description 5
- 238000002407 reforming Methods 0.000 claims description 5
- 238000009877 rendering Methods 0.000 claims description 4
- 235000013399 edible fruits Nutrition 0.000 claims description 3
- ONUFESLQCSAYKA-UHFFFAOYSA-N iprodione Chemical compound O=C1N(C(=O)NC(C)C)CC(=O)N1C1=CC(Cl)=CC(Cl)=C1 ONUFESLQCSAYKA-UHFFFAOYSA-N 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 238000011524 similarity measure Methods 0.000 claims description 3
- 239000000843 powder Substances 0.000 claims description 2
- 210000003128 head Anatomy 0.000 claims 1
- 230000003796 beauty Effects 0.000 abstract 1
- 238000013135 deep learning Methods 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 2
- 238000012797 qualification Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000000686 essence Substances 0.000 description 1
- 210000004209 hair Anatomy 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000002203 pretreatment Methods 0.000 description 1
- 238000010008 shearing Methods 0.000 description 1
- 238000005308 sum rule Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Geometry (AREA)
- Human Computer Interaction (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Computer Graphics (AREA)
- General Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Processing Or Creating Images (AREA)
- Image Processing (AREA)
Abstract
The present invention is provided a kind of star's face shaping and made up based on the matching of star's face and similarity analysis method recommends method and apparatus.Including:2D/3D face databases, facial image pretreatment, three-dimensional facial reconstruction, the matching of star's face, star's face shaping recommendation and vision presentation are built, star's face, which is made up, to be recommended to present with vision.The present invention carries out human face similarity degree matching in order to improve the stability of face matching, using 3D shape facilities;In order to improve user experience and interest, shaping and cosmetic process and effect to user's face carry out 3D vision presentations.Present invention can apply to online game, APP to develop, and can also be applied to shaping, beauty and cosmetic field.
Description
Technical field
It is more particularly to a kind of to be based on face phase the present invention relates to image procossing, picture search and mode identification technology
Recommend method and apparatus like star's face shaping cosmetic of degree analysis.
Background technology
In the last few years, people search for similar face and the attention rate of star's face matching increasingly improves, and some well-known applications are such as
Twins the or not, Face++ of Microsoft similar face search, Baidu's star's face, Baidu's evil spirit are schemed, find star's face, according to illumination star
The appearance of face etc., indicate that similar face search is just turning into the study hotspot of a new round.However, current star's face matching game is not
Stable, i.e., the different types of facial image of same person input often occur causes the star of matching different.In addition, current
The game of star's face allows people to feel to have not given full expression to the views.User is not desired to laugh it off after the star alike with oneself is found and just had no reply
, in fact most of user all have star powder love knot, further match interaction if can have with star's face, user can more like
Vigorously.
Therefore, the present invention will provide a kind of star's face matching process based on three-dimensional face to improve the matching of people star face
Stability.In addition, the recommendation of starization face shaping cosmetic is provided the user, and its process and step is presented in vision.By with
Upper method improves the stability and user experience of the matching of star's face, and is advantageous to the intervention of shaping, advertisement for cosmetics, to open
Hair business brings economic benefit.
The content of the invention
In order to solve problem above, the present invention proposes that a kind of star's face shaping based on human face similarity degree analysis is made up and recommended
Method and apparatus.
Star's face shaping cosmetic recommendation method based on human face similarity degree analysis, it is characterised in that comprise the following steps:
S1 builds star's 2D/3D face databases
Collect star's facial image progress size dimension normalization and handle the 2D facial images standardized, utilize 3D weights
Build technology the 2D stars face progress 3D of standardization is rebuild to obtain corresponding 3D facial images, structure comprises at least in this way
Star's 2D/3D face databases of 500.
Described 3D reconstruction techniques be using 3D deformation models method carry out 3D reconstructions, its method institute under:
Assuming that new facial image Iinput3D faces (Snew,Tnew) shape vector of existing 3D faceforms can be passed through
Linear combination with texture represents:
Wherein, SnFor the shape vector of 3D faceforms, TnFor the texture of 3D faceforms, an,bnRespectively shape
With the combination coefficient of texture.So, as long as there is combination coefficient (a, b), can just 3D reconstructions be carried out to new facial image.
3D deformation model matchings are carried out to new facial image, the method for solving combination coefficient (a, b) is:
Wherein, ImodelIt is the 2D perspective projection images of 3D faceforms, is determined by illumination model and camera model.Texture
The rest may be inferred for combination coefficient b solution.
Further, the present invention is improved the 3D face reconstruction side for proposing local 3D deformation models to 3D deformation models
Method:
First, face is divided into by the 5 big organic region such as forehead, eyebrow, eyes, nose, face according to face key point, and
3D face local organs models are built for each human face region.
Then, same division is carried out to new face, and using above-mentioned deformation model to each 2D human faces region
Carry out 3D reconstructions.
Finally, each local organs 3D rendering is carried out seamless spliced into 3D faces.
S2 facial images pre-process
The facial image of input is pre-processed, including Face datection, face key point location, face normalization, people
Face image quality assessment.
Described Face datection, it is characterised in that using traditional Adaboost methods or the face using deep learning
Detection method.
Described face key point location, it is characterised in that use traditional SDM (Supervised Descent
Method 68 key points of face) or using the method for deep learning are chosen.
Described face normalization, it is characterised in that carry out face size yardstick and unitary of illumination.
Described quality of human face image is judged, it is characterised in that the qualification for carrying out face resolution ratio and posture judges.
S3 facial images 3D is rebuild
3D reconstructions are carried out using local 3D deformation models to the 2D facial images after step S2 facial pretreatments, obtained
To corresponding 3D facial images.
S4 3D faces match and similarity analysis
Similar face search is carried out using 3D face shapes information, matches the star face most like with input face, and
Provide the similarity analysis of similar star's face.
Described similar face search, it is characterised in that carried out using the combination of shapes coefficient in 3D process of reconstruction similar
Degree matching:
First, the 3D for inputting each local organs of face is rebuild into form factor anLocal organs are carried out as characteristic vector
Match somebody with somebody, and using the cosine between two vectorial angles as the similarity measure between two local organs:
Wherein, an,anFor the characteristic vector of two matching local organs (local organs linear combination represents weights).
Then, each local organs similarity is weighted summation and draws final human face similarity degree.
The similarity analysis of described star's face, it is characterised in that provide overall similarity, each local organs similarity,
Most like local organs and least similar local organs.
S5 face shaping recommendations are presented with vision
Situation recommended user's shaping aspect is analyzed according to star's face similarity degree.Such as, recommendation will least analogous organs carry out it is whole
Shape, and show reforming process and design sketch to user using 3D form.
Further, user can also select shaping aspect according to the demand and hobby of oneself.Such as, rich forehead, brow ridge,
Double-edged eyelid, go pouch, augmentation rhinoplasty, Feng Yintang, open recessed inner eye corner, sleeping silkworm, temple filling, Feng Pingguo fleshes, rich cheek, molten lower chin line,
Molten masseter, rich lip, fill out chin, decree of dispelling line and line engraving lifting.
Further, user's face reforming process and design sketch are subjected to 3D vision presentations.
S6 faces, which are made up, to be recommended to present with vision
Recommend star's face cosmetic scheme to user, and each step effect visual in cosmetic process is showed.
Described cosmetic process step includes:Beat foundation cream, pencil one's eyebrows, eye shadow, informer, patch and play lipstick at eyelashes, brush rouge.
A kind of star's face shaping cosmetic recommendation apparatus based on human face similarity degree analysis, it is characterised in that including:
Memory cell, for storing star's 2D/3D facial images, star's face characteristic vector.Wherein characteristic vector is each office
The 3D of portion's organ rebuilds combination of shapes coefficient.
Graphics processing unit, for Face datection, crucial point location, normalization, image quality assessment.
Three-dimensional reconstruction unit, for carrying out 3D reconstructions to the 2D faces of input.
Face matching unit, for carrying out star's face similarity degree matching, find out the star face most like with user.
Shaping visual presentation unit, process and effect for 3D visual presence user's shapings.Shaping effect displaying includes:
Rich forehead, brow ridge, double-edged eyelid, pouch is removed, augmentation rhinoplasty, Feng Yintang, opens inner eye corner, crouch silkworm, temple filling, Feng Pingguo fleshes, rich cheek
It is recessed, molten lower chin line, molten masseter, rich lip, chin is filled out, decree of dispelling line, line engraving lifting.
Cosmetic visual presentation unit, the process and effect made up for showing user.Cosmetic process and effect displaying include:
Beat foundation cream, pencil one's eyebrows, eye shadow, informer, patch eyelashes, brush rouge, play lipstick.
Brief description of the drawings
A kind of star's face shapings based on human face similarity degree analysis of Fig. 1, which are made up, recommends method flow diagram.
A kind of star's face shaping cosmetic recommendation apparatus structure charts based on human face similarity degree analysis of Fig. 2.
Embodiment
To make the objects, technical solutions and advantages of the present invention clearer, clear and definite, develop simultaneously embodiment pair referring to the drawings
The present invention is further described.Exemplary, embodiment illustrates so that mobile phone version famous person matches game as an example.Should
Work as understanding, the specific embodiments described herein are merely illustrative of the present invention, is not intended to limit the present invention.
Embodiment 1:
Fig. 1 shows the flow that a kind of star's shaping cosmetic recommendation method based on human face similarity degree analysis is realized, its is detailed
State as follows:
S101 builds star's 2D/3D face databases
Collect star's facial image progress size dimension normalization and handle the 2D facial images standardized.Utilize 3D weights
Build technology the 2D stars face progress 3D of standardization is rebuild to obtain corresponding 3D facial images.
Wherein, the standard of collected star's 2D facial images is:Front and face are clear.
Wherein, described 3D facial images are rebuild to obtain by the method progress 3D of 3D deformation models, under its method institute:
Assuming that new facial image Iinput3D faces (Snew,Tnew) shape vector of existing 3D faceforms can be passed through
Linear combination with texture represents:
Wherein, SnFor the shape vector of 3D faces, TnFor the texture of 3D faces, an,bnRespectively shape and texture to
The combination coefficient of amount.So, as long as there is combination coefficient (a, b), can just 3D reconstructions be carried out to new facial image.
3D deformation model matchings are carried out to new facial image, the method for solving combination coefficient (a, b) is:
Wherein, ImodelIt is the 2D perspective projection images of 3D faceforms, is determined by illumination model and camera model.Texture
The rest may be inferred for combination coefficient b solution.
Described existing 3D faceforms, it is characterised in that the face 3D moulds of 500 objects are gathered by 3D scanners
Type.
Further, the present invention is improved the 3D face reconstruction side for proposing local 3D deformation models to 3D deformation models
Method:
First, face is divided into by the 5 big organic region such as forehead, eyebrow, eyes, nose, face according to face key point, and
3D face local organs dictionaries are built for each human face region.
Then, same division is carried out to new face, and using above-mentioned deformation model to each 2D human faces region
Carry out 3D reconstructions.
Finally, each local organs 3D rendering is carried out seamless spliced into 3D faces.
Further, this step is special as the shape of the local organs using the combination coefficient for storing each local organs 3D reconstructions
Sign vector.
S102 facial images pre-process
The pre- place of Face datection, crucial point location, normalization, image quality assessment etc. is carried out to the facial image of input
Reason.If quality of human face image judge is unqualified, user is prompted to resubmit image, only untill qualified.
Described Face datection, it is characterised in that using traditional Adaboost methods or the method using deep learning
Carry out Face datection.
Described face key point location, it is characterised in that use traditional SDM (Supervised Descent
Method 68 key points of face) or using the method for deep learning are chosen.
Described face normalization, it is characterised in that carried out according to face key point (eyebrow dirt pits, chin point, ear main points)
Size yardstick and unitary of illumination are carried out again after face shearing.
Described quality of human face image is judged, it is characterised in that the qualification for carrying out face resolution ratio and posture judges.Such as
The ratio that fruit facial image accounts for entire image is less than 1/3, and interpupillary distance is less than 40, then it is assumed that facial image resolution ratio is unqualified.Such as
Fruit face is left and right, upper and lower deflection any one be more than 30 degree, then it is assumed that facial image posture is unqualified.
S103 facial images 3D is rebuild
3D reconstructions are carried out using local 3D deformation models to the 2D facial images after S102 walks facial pretreatment, obtained
To corresponding 3D facial images and the combination of shapes coefficient of each local organs.
Described 3D method for reconstructing is the same as the 3D method for reconstructing in S101 steps.
S104 3D faces match and similarity analysis
Similar face search is carried out using 3D face shapes information, matches the star face most like with input face, and
Provide the similarity analysis of similar star's face.
Described similar face search, it is characterised in that carried out using the combination of shapes coefficient in 3D process of reconstruction similar
Degree matching:
First, the 3D for inputting each local organs of face is rebuild into form factor anLocal organs are carried out as characteristic vector
Match somebody with somebody, and using the cosine between two vectorial angles as the similarity measure between two local organs:
Wherein, an,anFor the characteristic vector of two matching local organs (local organs linear combination represents weights).
Then, each local organs similarity is weighted summation and draws final human face similarity degree, its weighted sum rule
It is then as follows:
θ=wfθf+wbθb+weθe+wnθn+wmθm
Wherein, θ is final similarity, θf,θb,θe,θn,θmRespectively forehead, eyebrow, eyes, nose, face it is similar
Degree, ws=0.2, wb=0.3, we=0.1, wn=0.2, wm=0.2 is respectively the power of forehead, eyebrow, eyes, nose and face
Value.
The similarity analysis of described star's face, it is characterised in that provide overall similarity, each local organs similarity,
Most like local organs and least similar local organs.
S105 face shaping recommendations are presented with vision
Situation recommended user's shaping aspect is analyzed according to star's face similarity degree.Such as, recommendation will least analogous organs (nose)
Shaping-Nose raising opthopedic is carried out to recommend.And show reforming process and design sketch to user using 3D form.
Further, user can also be according to the hobby of oneself, selection shaping aspect, such as rich forehead, brow ridge, double-edged eyelid,
Pouch is removed, augmentation rhinoplasty, Feng Yintang, opens inner eye corner, crouch silkworm, and temple filling, Feng Pingguo fleshes, rich cheek is recessed, molten lower chin line, molten masseter,
Rich lip, fills out chin, decree of dispelling line, line engraving lifting.
S106 faces, which are made up, to be recommended to present with vision
Recommend star's face cosmetic scheme to user, and each step effect visual in cosmetic process is showed.
Described cosmetic process and effect visual rendering step includes:Beat foundation cream, pencil one's eyebrows, eye shadow, informer, patch eyelashes, brush
Rouge, play lipstick.
Embodiment 2:
Fig. 2 shows a kind of structure chart of star's face shaping cosmetic recommendation apparatus based on human face similarity degree analysis, is described in detail such as
Under:
Memory cell, for storing star's 2D/3D facial images, star's face characteristic vector.Wherein characteristic vector is each office
The 3D of portion's organ rebuilds combination of shapes coefficient.
Graphics processing unit, for Face datection, crucial point location, normalization, image quality assessment.
Three-dimensional reconstruction unit, for carrying out 3D reconstructions to the 2D faces of input.Wherein, 3D rebuilds and uses 3D local deformations
Model method.
Face matching unit, for carrying out star's face similarity degree matching, the star face most like with user is found out, and carry out
Similarity analysis.
Shaping visual presentation unit, process and effect for 3D displaying user's micro-shapings.The displaying of micro-shaping effect includes:
Rich forehead, brow ridge, double-edged eyelid, pouch is removed, augmentation rhinoplasty, Feng Yintang, opens inner eye corner, crouch silkworm, temple filling, Feng Pingguo fleshes, rich cheek
It is recessed, molten lower chin line, molten masseter, rich lip, chin is filled out, decree of dispelling line, line engraving lifting.
Cosmetic visual presentation unit, the process and effect made up for showing user.Cosmetic process and effect displaying include:
Beat foundation cream, pencil one's eyebrows, eye shadow, informer, patch eyelashes, brush rouge, play lipstick.
Above-described embodiment is the preferable embodiment of the present invention, but embodiments of the present invention are not by above-described embodiment
Limitation, other any Spirit Essences without departing from the present invention with made under principle change, modification, replacement, combine, simplification,
Equivalent substitute mode is should be, is included within protection scope of the present invention.
Claims (7)
- A kind of 1. star's face shaping cosmetic recommendation method based on human face similarity degree analysis, it is characterised in that comprise the following steps:Build star's 2D/3D face databases:Collect star's facial image progress size dimension normalization and handle the 2D standardized Facial image, 3D is carried out to the 2D stars face of standardization using 3D reconstruction techniques and rebuilds to obtain corresponding 3D facial images, with this Mode builds star's 2D/3D face databases including at least 500;Facial image pre-processes:The facial image of user's input is pre-processed, including Face datection, face key point are determined Position, face normalization, quality of human face image are judged;Facial image 3D is rebuild:3D reconstructions are carried out using local 3D deformation models method to pretreated user's facial image, 3D facial images corresponding to obtaining;3D faces match and similarity analysis:Similar face search is carried out using 3D face shapes information, is matched and input people The most like star's face of face, and provide the similarity analysis of similar star's face;Face shaping recommendation is presented with vision:Situation recommended user's shaping aspect is analyzed according to star's face similarity degree, and utilizes 3D Form show reforming process and design sketch to user;Or select corresponding shaping side according to individual subscriber shaping requirements and hobby Case, and carry out reforming process and the 3D visual presences of effect;Face, which is made up, to be recommended to present with vision:Recommend star's face cosmetic scheme to user, and each step in cosmetic process is imitated Fruit vision shows.
- 2. a kind of star's face shaping cosmetic recommendation method based on human face similarity degree analysis according to claim 1, it is special Sign is, the method that described 3D face reconstruction techniques use local 3D deformation models:First, face is divided into by the 5 big organic region such as forehead, eyebrow, eyes, nose, face according to face key point, on backstage 3D face local organs models are built for each human face region, complete the structure of face local organs 3D model dictionaries;Then, same division is carried out to new face, and 3D weights are carried out to each 2D human faces region using 3D deformation models Build:Assuming that new face local organs image Iinput3D be expressed as (Snew,Tnew), then existing 3D faces office can be passed through The linear combination of the shape vector and texture of portion's organ model represents:Wherein, SnFor the shape vector of existing 3D faces local organs model, TnFor existing 3D faces local organs model Texture, an,bnThe respectively combination coefficient of shape and texture, as long as therefore have a combination coefficient (a, b), just can be right New face local organs image carries out 3D reconstructions;Assuming that 3D deformation model matchings are carried out to new face local organs image, then the problem of solving combination coefficient (a, b) It is converted into:Wherein, ImodelIt is the 2D perspective projection images of 3D face local organs models, is determined by illumination model and camera model, The rest may be inferred for texture-combined coefficient b solution;Finally, each local organs 3D rendering is carried out seamless spliced into 3D facial images.
- 3. a kind of star's face shaping cosmetic recommendation method based on human face similarity degree analysis according to claim 1, it is special Sign is that described similar face searching method is to carry out similarity mode using the combination of shapes coefficient in 3D process of reconstruction:First, the 3D for inputting each local organs of face is rebuild into form factor anLocal organs matching is carried out as characteristic vector, and Using the cosine between two vectorial angles as the similarity measure between two local organs:Wherein, an,anFor the characteristic vector (that is, local organs linear combination represents weights) of two matching local organs;Then, each local organs similarity is weighted summation and draws final human face similarity degree.
- 4. a kind of star's face shaping cosmetic recommendation method based on human face similarity degree analysis according to claim 1, it is special Sign is that the similarity analysis of described star's face includes:Overall similarity, each local organs similarity, most like local device Official and least similar local organs.
- 5. a kind of star's face shaping cosmetic recommendation method based on human face similarity degree analysis according to claim 1, it is special Sign is that described shaping suggested design includes:Rich forehead, brow ridge, double-edged eyelid, go pouch, augmentation rhinoplasty, Feng Yintang, open inner eye corner, Sleeping silkworm, temple filling, Feng Pingguo fleshes, rich cheek be recessed, molten lower chin line, molten masseter, rich lip, fills out chin, decree of dispelling line and line engraving and carries Rise.
- 6. a kind of star's face shaping cosmetic recommendation method based on human face similarity degree analysis according to claim 1, it is special Sign is that the displaying of described cosmetic step effect includes:Beat foundation cream, pencil one's eyebrows, eye shadow, informer, patch and beat lip at eyelashes, brush rouge Cream.
- A kind of 7. star's face shaping cosmetic recommendation apparatus based on human face similarity degree analysis, it is characterised in that including:Memory cell, for storing star's 2D/3D facial images, star's face characteristic vector, wherein characteristic vector are each local devices The 3D of official rebuilds combination of shapes coefficient;Graphics processing unit, for Face datection, crucial point location, normalization, image quality assessment;Three-dimensional reconstruction unit, for carrying out 3D reconstructions to the 2D faces of input;Face matching unit, for carrying out star's face similarity degree matching, find out the star face most like with user;Shaping visual presentation unit, for the process and effect of 3D visual presence user's shapings, shaping effect displaying includes:Rich volume Head, brow ridge, double-edged eyelid, pouch is removed, augmentation rhinoplasty, Feng Yintang, opens inner eye corner, crouch silkworm, and temple filling, Feng Pingguo fleshes, rich cheek is recessed, molten Lower chin line, molten masseter, rich lip, fills out chin, decree of dispelling line, line engraving lifting;Cosmetic visual presentation unit, the process and effect made up for showing user, cosmetic process and effect displaying include:Beat powder Bottom, pencil one's eyebrows, eye shadow, informer, patch eyelashes, brush rouge, play lipstick.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710804911.3A CN107506559B (en) | 2017-09-08 | 2017-09-08 | Star face shaping makeup recommendation method and device based on face similarity analysis |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710804911.3A CN107506559B (en) | 2017-09-08 | 2017-09-08 | Star face shaping makeup recommendation method and device based on face similarity analysis |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107506559A true CN107506559A (en) | 2017-12-22 |
CN107506559B CN107506559B (en) | 2021-03-23 |
Family
ID=60695268
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710804911.3A Expired - Fee Related CN107506559B (en) | 2017-09-08 | 2017-09-08 | Star face shaping makeup recommendation method and device based on face similarity analysis |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107506559B (en) |
Cited By (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108053365A (en) * | 2017-12-29 | 2018-05-18 | 百度在线网络技术(北京)有限公司 | For generating the method and apparatus of information |
CN108229415A (en) * | 2018-01-17 | 2018-06-29 | 广东欧珀移动通信有限公司 | Information recommendation method, device, electronic equipment and computer readable storage medium |
CN108898546A (en) * | 2018-06-15 | 2018-11-27 | 北京小米移动软件有限公司 | Face image processing process, device and equipment, readable storage medium storing program for executing |
CN109061892A (en) * | 2018-09-27 | 2018-12-21 | 广州狄卡视觉科技有限公司 | Plastic surgery medical image Model Reconstruction interacts naked-eye stereoscopic display system and method |
CN109189967A (en) * | 2018-08-24 | 2019-01-11 | 微云(武汉)科技有限公司 | A kind of lift face proposal recommending method, device and storage medium based on recognition of face |
CN110110118A (en) * | 2017-12-27 | 2019-08-09 | 广东欧珀移动通信有限公司 | Dressing recommended method, device, storage medium and mobile terminal |
CN110136270A (en) * | 2018-02-02 | 2019-08-16 | 北京京东尚科信息技术有限公司 | The method and apparatus of adornment data are held in production |
CN110473608A (en) * | 2019-06-27 | 2019-11-19 | 上海意就医疗技术中心 | Analogy method, system, device and the storage medium of weight operation on eyelid postoperative effect |
CN111062260A (en) * | 2019-11-25 | 2020-04-24 | 杭州绿度信息技术有限公司 | Automatic generation method of facial cosmetic recommendation scheme |
WO2020093634A1 (en) * | 2018-11-06 | 2020-05-14 | 平安科技(深圳)有限公司 | Face recognition-based method, device and terminal for adding image, and storage medium |
CN111966852A (en) * | 2020-06-28 | 2020-11-20 | 北京百度网讯科技有限公司 | Virtual face-lifting method and device based on human face |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101968802A (en) * | 2010-09-30 | 2011-02-09 | 百度在线网络技术(北京)有限公司 | Method and equipment for recommending content of Internet based on user browse behavior |
CN102708575A (en) * | 2012-05-17 | 2012-10-03 | 彭强 | Daily makeup design method and system based on face feature region recognition |
CN102982165A (en) * | 2012-12-10 | 2013-03-20 | 南京大学 | Large-scale human face image searching method |
US20130287294A1 (en) * | 2012-04-30 | 2013-10-31 | Cywee Group Limited | Methods for Generating Personalized 3D Models Using 2D Images and Generic 3D Models, and Related Personalized 3D Model Generating System |
CN104851123A (en) * | 2014-02-13 | 2015-08-19 | 北京师范大学 | Three-dimensional human face change simulation method |
US20160287166A1 (en) * | 2015-04-03 | 2016-10-06 | Bao Tran | Personal monitoring system |
CN106033621A (en) * | 2015-03-17 | 2016-10-19 | 阿里巴巴集团控股有限公司 | Three-dimensional modeling method and device |
CN106778468A (en) * | 2016-11-14 | 2017-05-31 | 深圳奥比中光科技有限公司 | 3D face identification methods and equipment |
-
2017
- 2017-09-08 CN CN201710804911.3A patent/CN107506559B/en not_active Expired - Fee Related
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101968802A (en) * | 2010-09-30 | 2011-02-09 | 百度在线网络技术(北京)有限公司 | Method and equipment for recommending content of Internet based on user browse behavior |
US20130287294A1 (en) * | 2012-04-30 | 2013-10-31 | Cywee Group Limited | Methods for Generating Personalized 3D Models Using 2D Images and Generic 3D Models, and Related Personalized 3D Model Generating System |
CN102708575A (en) * | 2012-05-17 | 2012-10-03 | 彭强 | Daily makeup design method and system based on face feature region recognition |
CN102982165A (en) * | 2012-12-10 | 2013-03-20 | 南京大学 | Large-scale human face image searching method |
CN104851123A (en) * | 2014-02-13 | 2015-08-19 | 北京师范大学 | Three-dimensional human face change simulation method |
CN106033621A (en) * | 2015-03-17 | 2016-10-19 | 阿里巴巴集团控股有限公司 | Three-dimensional modeling method and device |
US20160287166A1 (en) * | 2015-04-03 | 2016-10-06 | Bao Tran | Personal monitoring system |
CN106778468A (en) * | 2016-11-14 | 2017-05-31 | 深圳奥比中光科技有限公司 | 3D face identification methods and equipment |
Non-Patent Citations (2)
Title |
---|
郝宁波 等: "《基于局部形变模型三维人脸快速建模》", 《视频应用与工程》 * |
龚勋 等: "《基于特征点的三维人脸形变模型》", 《计算机图形学与计算机辅助设计》 * |
Cited By (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110110118B (en) * | 2017-12-27 | 2021-11-16 | Oppo广东移动通信有限公司 | Dressing recommendation method and device, storage medium and mobile terminal |
CN110110118A (en) * | 2017-12-27 | 2019-08-09 | 广东欧珀移动通信有限公司 | Dressing recommended method, device, storage medium and mobile terminal |
CN108053365A (en) * | 2017-12-29 | 2018-05-18 | 百度在线网络技术(北京)有限公司 | For generating the method and apparatus of information |
CN108053365B (en) * | 2017-12-29 | 2019-11-05 | 百度在线网络技术(北京)有限公司 | Method and apparatus for generating information |
CN108229415A (en) * | 2018-01-17 | 2018-06-29 | 广东欧珀移动通信有限公司 | Information recommendation method, device, electronic equipment and computer readable storage medium |
CN110136270A (en) * | 2018-02-02 | 2019-08-16 | 北京京东尚科信息技术有限公司 | The method and apparatus of adornment data are held in production |
CN108898546A (en) * | 2018-06-15 | 2018-11-27 | 北京小米移动软件有限公司 | Face image processing process, device and equipment, readable storage medium storing program for executing |
CN108898546B (en) * | 2018-06-15 | 2022-08-16 | 北京小米移动软件有限公司 | Face image processing method, device and equipment and readable storage medium |
CN109189967A (en) * | 2018-08-24 | 2019-01-11 | 微云(武汉)科技有限公司 | A kind of lift face proposal recommending method, device and storage medium based on recognition of face |
CN109061892A (en) * | 2018-09-27 | 2018-12-21 | 广州狄卡视觉科技有限公司 | Plastic surgery medical image Model Reconstruction interacts naked-eye stereoscopic display system and method |
WO2020093634A1 (en) * | 2018-11-06 | 2020-05-14 | 平安科技(深圳)有限公司 | Face recognition-based method, device and terminal for adding image, and storage medium |
CN110473608A (en) * | 2019-06-27 | 2019-11-19 | 上海意就医疗技术中心 | Analogy method, system, device and the storage medium of weight operation on eyelid postoperative effect |
CN110473608B (en) * | 2019-06-27 | 2022-04-19 | 上海合智医疗科技有限公司 | Method, system, device and storage medium for simulating postoperative effect of double-eyelid surgery |
CN111062260A (en) * | 2019-11-25 | 2020-04-24 | 杭州绿度信息技术有限公司 | Automatic generation method of facial cosmetic recommendation scheme |
CN111062260B (en) * | 2019-11-25 | 2024-03-05 | 杭州绿度信息技术有限公司 | Automatic generation method of face-beautifying recommendation scheme |
CN111966852A (en) * | 2020-06-28 | 2020-11-20 | 北京百度网讯科技有限公司 | Virtual face-lifting method and device based on human face |
CN111966852B (en) * | 2020-06-28 | 2024-04-09 | 北京百度网讯科技有限公司 | Face-based virtual face-lifting method and device |
Also Published As
Publication number | Publication date |
---|---|
CN107506559B (en) | 2021-03-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107506559A (en) | Star's face shaping based on human face similarity degree analysis, which is made up, recommends method and apparatus | |
KR102241153B1 (en) | Method, apparatus, and system generating 3d avartar from 2d image | |
US10799010B2 (en) | Makeup application assist device and makeup application assist method | |
US7634103B2 (en) | Analysis using a three-dimensional facial image | |
US20160005106A1 (en) | 3d platform for aesthetic simulation | |
CN107820591A (en) | Control method, controller, Intelligent mirror and computer-readable recording medium | |
CN103649987A (en) | Face impression analysis method, cosmetic counseling method, and face image generation method | |
US20030065255A1 (en) | Simulation of an aesthetic feature on a facial image | |
CN104205168A (en) | Makeup application assistance device, makeup application assistance method, and makeup application assistance program | |
CN108932654A (en) | A kind of virtually examination adornment guidance method and device | |
KR20210095696A (en) | Data processing method and device, and medium for generating face image | |
CN104933742B (en) | A kind of cartoon portrait automatic generation method | |
US9245369B2 (en) | Method and apparatus for generating an avatar | |
CN106446207B (en) | Makeups library banking process, personalized makeups householder method and its device | |
Sharma et al. | Comparative analysis of CycleGAN and AttentionGAN on face aging application | |
CN109376661A (en) | Method and apparatus for output information | |
JP4893968B2 (en) | How to compose face images | |
Rowland | Computer graphic control over human face and head appearance, genetic optimisation of perceptual characteristics. | |
CN113947520A (en) | Method for realizing face makeup conversion based on generation of confrontation network | |
CN114782240A (en) | Picture processing method and device | |
KR19980053565A (en) | Method and apparatus for encoding facial motion | |
CN110069716A (en) | A kind of makeups recommended method, system and computer readable storage medium | |
JP5650012B2 (en) | Facial image processing method, beauty counseling method, and facial image processing apparatus | |
CN109876457A (en) | Game role generation method, device and storage medium | |
CN110930294A (en) | Intelligent hair style recommendation and presentation method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20210323 |