CN107274493A - A kind of three-dimensional examination hair style facial reconstruction method based on mobile platform - Google Patents
A kind of three-dimensional examination hair style facial reconstruction method based on mobile platform Download PDFInfo
- Publication number
- CN107274493A CN107274493A CN201710506496.3A CN201710506496A CN107274493A CN 107274493 A CN107274493 A CN 107274493A CN 201710506496 A CN201710506496 A CN 201710506496A CN 107274493 A CN107274493 A CN 107274493A
- Authority
- CN
- China
- Prior art keywords
- face
- mrow
- hair style
- dimensional
- mfrac
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T19/00—Manipulating 3D models or images for computer graphics
- G06T19/006—Mixed reality
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T19/00—Manipulating 3D models or images for computer graphics
- G06T19/20—Editing of 3D images, e.g. changing shapes or colours, aligning objects or positioning parts
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Graphics (AREA)
- Health & Medical Sciences (AREA)
- Software Systems (AREA)
- Oral & Maxillofacial Surgery (AREA)
- General Engineering & Computer Science (AREA)
- Computer Hardware Design (AREA)
- General Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Human Computer Interaction (AREA)
- Geometry (AREA)
- Architecture (AREA)
- Processing Or Creating Images (AREA)
Abstract
Hair style facial reconstruction method is tried the invention discloses a kind of three-dimensional based on mobile platform, user face front is obtained by camera or photograph album and shines one, send server to, server is calculated by algorithm routine, after the completion of return to face relevant parameter and rebuild after face texture mapping, instantiate new human face three-dimensional model according to face relevant parameter and textures and shown.The different three-dimensional hair style of loading, accessories and displaying background etc. are selected according to user, instantiation is shown on screen, so as to realize the virtual examination hair shape effect of display.The present invention provides the user more real virtual examination hair style experience, the functions such as three-dimensional face emulation, hair style and accessories replacing and the personal information storage of user are realized on function innovation, system loads hair style and accessories in server etc. and selective, finally instantiation display, basis, the actual effect of more real embodiment difference hair style are provided for user's examination hair style.
Description
Technical field
The present invention relates to virtual reality image processing technology field, more particularly to a kind of three-dimensional based on mobile platform
Try hair style facial reconstruction method.
Background technology
With expanding economy, people focus more on self-image, are more willing to as lifting self-image makings consumption, hair style
Key effect is played to moulding for a personal image makings with apparel modeling.Different from clothes, hair style once makes a change short
It is difficult to recover in time.The problem of can not shifting to an earlier date preview for hair style, the solution for having two kinds of main flows at present:One is hair style figure
Volume and color development model, two be the examination hair style application based on two dimensional image.But all there is more obvious defect in both schemes.
Hair style atlas and hair dyeing color model can not provide the examination hair effect of specific user.Examination hair style based on two dimensional image
Using typically requiring user's photo angle correspondence hair style angle, otherwise face mask has infelicity, shape of face does not meet user.
For different user, even identical hair style is there is also difference on effect, existing method can not preferably meet consumer and change
The demand of preview before type.
The content of the invention
In view of the drawbacks described above of prior art, the technical problems to be solved by the invention are to provide a kind of based on mobile flat
The three-dimensional examination hair style facial reconstruction method of platform, is built based on mobile platform and combines server progress system optimization, examination hair
Type effect is presented in three dimensions, can carry out the rotational view of multi-angle, meet user more restyle the hair, the need of color development and accessories
Ask.With Face datection, anamorphose, image co-registration and three-dimensional model deformation technology, invented it is a kind of using single photo as according to
According to user's face three-dimensional rebuilding method, by the three dimensional user face of reconstruction be used for try hair style, can truly also original subscriber examination hair
Effect, experience and practicality are more preferably.
To achieve the above object, the invention provides a kind of three-dimensional examination hair style human face rebuilding side based on mobile platform
Method, comprises the following steps:
The three-dimensional face storehouse of shape of face grid and basic face texture mapping is set up, for facial image mesh reconstruction and texture
Textures are rebuild, and mesh reconstruction target is obtains shape of face fitting parameter, and texture mapping is rebuild and obtains the face line based on user picture
Manage textures;
Facial image is gathered, face mask position and eyebrow, eyes, nose and face outline position in image is identified
Individual features point, user's shape of face is determined according to face mask characteristic point, according to face outline position characteristic point determine extract
Pictorial element scope;
Determine after shape of face data, shape of face fitting is calculated according to shape of face fitting formula (prior art, here is omitted)
Parameter, realizes mesh reconstruction;Determine after face position, extract pictorial element, by anamorphose algorithm (prior art, this
Place is repeated no more) and graph cut processing after obtain new face texture mapping;
Grid after reconstruction and texture mapping are merged, instantiation display can obtain complete user's three-dimensional face.
It is above-mentioned to set up shape of face grid and the three-dimensional face storehouse of basic face texture mapping is specially:
The preliminary acquisition of three-dimensional face is realized by FaceShift Studio combination Kinect cameras, is used
3dsMax, Zbrush software are adjusted and corrected to the face feature position of basic grid, using Photoshop according to existing
UV expanded views repaint the basic textures material for fusion, and the grid model after finishing is combined with textures, complete basic
The foundation in shape of face three-dimensional face storehouse.
Above-mentioned three-dimensional face storehouse includes long face, oval face, square face, pyriform face, rhombus face, heart-shaped face and round face this seven kinds of bases
This shape of face grid model and a basic face texture mapping, wherein seven kinds of shape of face grid models have marked face mask pair in advance
Position feature point is answered, by these projecting characteristic points to plane XOY, reference characteristic point T is obtained after normalization1,T2,T3,...,
T7。
After above-mentioned determination shape of face data, fitting parameter is calculated according to shape of face fitting formula, mesh reconstruction is realized, specifically
For:According to the facial image of input, face mask characteristic point is identified, and contour feature point coordinates is normalized,
Obtain vector Tinput, by seven kinds of standard shape of face model projections in face shape of face storehouse to XOY plane, normalized is obtained
T1, T2, T3…T7According to formula:
Wherein,And α1,α2,...,α7For unknown number to be asked, solveObtain phase
The shape of face fitting parameter answered, parameter sends the reconstruction that mobile terminal completes grid to.
After above-mentioned determination face position, pictorial element is extracted, new face texture is obtained after being handled by image algorithm
Textures, be specially:
According to feature point detection to face position, corresponding facial image region is corroded, gained after interception corrosion
Region carries out the image co-registration after the deformation operation of rotation and scaling, deformation operation to basic textures material, through past bloom
It is the user's face textures that can be rebuild with shade.
It is above-mentioned to be corroded corresponding facial image region, be specially:
Set different corrosion number of times to reach corrosive effect for facial image different piece.
Gained region carries out the deformation operation of rotation and scaling after above-mentioned interception corrosion, is specially:
To interception corrosion after gained region rotated and scaling deformation operation, while facial characteristics control point also can root
Moved according to the coefficient of translation scaling, according to the front and rear change at a control point, inverse distance-weighting calculates and obtains remaining institute
There are the transformation matrices at control point, wherein interpolating function is:
1...N it is to represent the control point beyond four angles of picture, dis1 ... disN represents the point distance currently to be calculated
Control point 1 ... N distance;
To the reverse interpolation of blank progress in image due to stretching formation after deformation.
Image co-registration after above-mentioned deformation operation is on basic textures material, and Image Fusion is graph cut algorithm.
The beneficial effects of the invention are as follows:
The present invention provides the user more real virtual examination hair style experience, and the three-dimensional people of user is realized on function innovation
Face emulation, hair style and accessories are changed and the function such as personal information storage, and are the interactions based on mobile terminal.System will be serviced
Hair style and accessories in device etc. are loaded and selective, the file of system loads correspondence material after selection, and instantiate display.It is based on
The three-dimensional facial reconstruction algorithm of single photo provides basis, the actual effect of more real embodiment difference hair style for user's examination hair style.
The technique effect of the design of the present invention, concrete structure and generation is described further below with reference to accompanying drawing, with
It is fully understood from the purpose of the present invention, feature and effect.
Brief description of the drawings
Fig. 1 tries hair style systemic-function and structural framing figure for the three-dimensional of the present invention;
Mobile terminals and server end interaction diagrams of the Fig. 2 for the present invention;
Fig. 3 is system program overall framework figure of the invention.
Embodiment
As shown in Figure 1, 2, 3, a kind of three-dimensional examination hair style facial reconstruction method based on mobile platform, including following step
Suddenly:
The three-dimensional face storehouse of shape of face grid and basic face texture mapping is set up, for facial image mesh reconstruction and texture
Textures are rebuild, and mesh reconstruction target is obtains shape of face fitting parameter, and texture mapping is rebuild and obtains the face line based on user picture
Manage textures;
Facial image is gathered, the individual features point of face mask position and face position is identified, it is special according to face mask
Levy determination user's shape of face, the pictorial element scope for determining to extract according to face position characteristic point;
Determine after shape of face data, shape of face fitting parameter is calculated according to formula, mesh reconstruction is realized;Determine face position
Afterwards, pictorial element is extracted, new face texture mapping is obtained after being handled by image algorithm;
Grid after reconstruction and texture mapping are merged, instantiation display can obtain complete user's three-dimensional face.
It is described to set up shape of face grid and the three-dimensional face storehouse of basic face texture mapping is specially in the present embodiment:
The preliminary acquisition of three-dimensional face is realized by FaceShift Studio combination Kinect cameras, is used
3dsMax, Zbrush software carry out suitably adjusting and correcting to the face feature position of basic grid, use Photoshop foundations
Existing UV expanded views repaint the basic textures material for fusion, and the grid model after finishing is combined with textures, complete
The foundation in basic shape of face three-dimensional face storehouse.
In the present embodiment, the three-dimensional face storehouse include long face, oval face, square face, pyriform face, rhombus face, heart-shaped face and
The basic shape of face grid model of this seven kinds of round face and a basic face texture mapping, wherein seven kinds of shape of face grid models are marked in advance
Good face mask correspondence position characteristic point, by these projecting characteristic points to plane XOY, obtains reference characteristic point after normalization
T1,T2,T3,...,T7。
In the present embodiment, after the determination shape of face data, fitting parameter is calculated according to formula, mesh reconstruction is realized, had
Body is:According to formula:
Wherein,And α1,α2,...,α7For unknown number to be asked, solveObtain phase
The shape of face fitting parameter answered, parameter sends the reconstruction that mobile terminal completes grid to.
In the present embodiment, after the determination face position, pictorial element is extracted, obtains new after being handled by image algorithm
Face texture mapping, be specially:
According to feature point detection to face position, corresponding facial image region is corroded, gained after interception corrosion
Region carries out the image co-registration after the deformation operation of rotation and scaling, deformation operation to basic textures material, through past bloom
It is the user's face textures that can be rebuild with shade.
It is described to be corroded corresponding facial image region in the present embodiment, be specially:For facial image difference portion
Set up surely different corrosion number of times separately to reach relatively reasonable corrosive effect.
In the present embodiment, gained region carries out the deformation operation of rotation and scaling after the interception corrosion, is specially:
To interception corrosion after gained region rotated and scaling deformation operation, while facial characteristics control point also can root
Moved according to the coefficient of translation scaling, according to the front and rear change at a control point, inverse distance-weighting calculates and obtains remaining institute
There are the transformation matrices at control point, wherein interpolating function is:
To the reverse interpolation of blank progress in image due to stretching formation after deformation.
In the present embodiment, in image co-registration after the deformation operation to basic textures material, Image Fusion is pool
Loose blending algorithm.
The idiographic flow of the present invention is as follows:
The present invention is using resource, logic, the overall architecture design philosophy of performance separation, using MVC pattern as general procedure frame
Frame pattern, to keep flexibility and the autgmentability of system.
Resource layer mainly manages the fine arts of correlation by Unity3D resource management techniques combination fine arts resource Naming conventions
Resource, including threedimensional model, different type textures and two dimensional image etc..
The task of logical layer is to carry out logical process to difference in functionality demand in system, is divided into virtual examination hair style, consulting and obtains
Take and the big module of individual center three.Virtual Shi Fa centers allow the replacing of hair style, color development, accessories and background, touch dragging model
Multi-angle observation is carried out, supports examination hair effect sectional drawing to preserve with sharing.Popular hair style information and hair products recommending module there is provided
Trend hair style, collocation and product information etc. are seeked advice from.Individual center module is believed data such as the favorites and examination hair sectional drawing of user
Breath is managed and preserved, convenient to check again.
Presentation layer is rendered using the Unity3D figures for carrying out examination hair scene, completes whole with UGUI for IDK
The vision and interaction flow of individual system.
During operation, user face front is obtained by camera or photograph album and shines one, server is sent to, server passes through
Algorithm routine is calculated, after the completion of return to face relevant parameter and rebuild after face texture mapping, according to face relevant parameter and
Textures instantiate new human face three-dimensional model and shown.The different three-dimensional hair style of loading, accessories are selected according to user
With displaying background etc., instantiation is shown on screen, so as to realize the virtual examination hair shape effect of display, each step by step it is specific such as
Under:
(1) 3-D scanning and the foundation in basic shape of face three-dimensional face storehouse.
Women is higher to virtual examination hair style demand property, therefore three-dimensional facial reconstruction of the present invention is mainly for Asia women.Adjust
Grind analysis to find, most Asia women face masks can be by long face, oval face, square face, pyriform face, rhombus face, heart
Face and the basic shape of face fitting of this seven kinds of round face are obtained.Therefore, the three-dimensional face storehouse conduct of seven kinds of basic shapes of face comprising more than is set up
The fitting basis of mesh model reconstruction, it is more convenient.
FaceShift Studio are real-time face expression capturing tools, are mainly used to replicate mankind's face action, and turn
It is melted into threedimensional model or animation.The present invention realizes the first of three-dimensional face by FaceShift Studio combination Kinect cameras
Step is obtained.Ensure that Kinect cameras are correctly connected, selection carmine mode sensors collect the information such as depth and image.
Under training patterns, scanned person creates the expression file of oneself by imitating default nature expression, catches the table finished
Feelings file, which can be processed, generates smooth threedimensional model.
The topological structure of basic model is simple obtained by 3-D scanning, and deploys with rational UV, but textures image is present
The problems such as dislocation deformation, fuzzy uneven color.With reference to the structure and contour feature of seven kinds of different shapes of face, with 3dsMax,
Zbrush softwares carry out suitably adjusting and correcting to the face feature position of basic grid, using Photoshop according to existing UV
Expanded view repaints the basic textures material for fusion, and the grid model after finishing is combined with textures, just completes basic
The foundation in shape of face three-dimensional head portrait storehouse.In addition, basic material for seven in above-mentioned three-dimensional head portrait model library different shapes of face, it is necessary to
Some characteristic points of face mask correspondence position have been marked, have been that mesh reconstruction is prepared.
(2) three-dimensional head portrait mesh model reconstruction.
Complete threedimensional model is generally made up of grid model and texture mapping two parts, and the present invention uses grid and texture
The scheme that textures are separately rebuild, and based on individual facial photo of user.3 d human face mesh model algorithm for reconstructing
Target is to draw seven fusion coefficients, is then transported on return moved end and carries out Mesh Fitting reconstruction.
Receive after input photo, the characteristic point of face's outermost side profile in photo is detected first, and these points
Information is saved in the inside., will for the above-mentioned basic material of three-dimensional face for having marked face mask correspondence position characteristic point in advance
These projecting characteristic points obtain reference characteristic point T on plane XOY after normalization1,T2,T3,...,T7.According to formula:
Wherein,And α1,α2,...,α7For unknown number to be asked, solveIt can obtain
Corresponding shape of face fitting parameter is obtained, parameter, which sends mobile terminal to, can complete the reconstruction of grid.
(3) three-dimensional head portrait texture mapping is rebuild.
Texture mapping makes model more careful and truer in visual effect.For different user examination hair effect is presented,
Not only its shape of face is fitted, it is often more important that the texture mapping of its true appearance can be reduced by reconstructing.
The interface provided using Face datection storehouse Face++, is extracted to the key point that input photo carries out face region, obtained
Take the approximate range of eyes in photo, nose, eyebrow and face.Then the region is corroded, i.e., expanded relevant position,
Corrode the more correspondence position extensions of number of times bigger, can set different corrosion number of times for facial different piece more closes to reach
The corrosive effect of reason, then the whole characteristic area for corroding acquisition is extracted from photograph.
Above-mentioned key area is translated and scaling deformation operation, while facial characteristics control point can also be scaled according to translation
Coefficient move, according to the front and rear change at a control point, inverse distance-weighting calculates and obtains remaining all control point
Transformation matrices, wherein interpolating function are:
Needed after deformation to, because the blank that stretching is formed carries out reverse interpolation, basic thinking is above-mentioned mistake in image
The inverse process of journey.
Obtained image will be deformed, will be fused to by the way that graph cut algorithm is seamless on basic face textures material, can be compared with
The problem of good processing overlapping region is merged.G is the function of face face position source images, gradientF* pastes for background
Figure material image S function, f is that region Ω waits to find a function,For fusion border.Fusion target has two:One is to retain basic
Textures material information, two be seamless spliced.It is converted into according to Poisson image editing algorithms (Poisson Image Editing)
Optimization problem:
Its solving result solves for Poisson's equation:
The discrete solution form of above formula:
Wherein, vpq=gp-gq,|Np|=4.
(4) operational efficiency optimizes.
To realize that the Effec-tive Function of system reduces calculating and the storage pressure of mobile terminal, three-dimensional head portrait algorithm for reconstructing is built
On the server, processing operation, parameter acquiring, optimization of photo etc. will all be carried out in server end, had preferably to system
Protectiveness, while effectively improving operational efficiency.Server is as the operation platform of core algorithm, and it is special that function includes extraction face
Levy a little, the generation of the algorithms of face textures, draw some shape of face weighted datas and by the face textures of synthesis and shape of face weight number
According to sending jointly to mobile terminal.In addition, being related to the three-dimensional fine arts material money such as the larger model of data volume and textures in the present invention
Source, AssetBundle is all rationally packaged into by Unity3D and is put on the server, is dynamically loaded in system operation,
So as to discharge more memory headrooms.
(5) fine arts material optimizes.
The computing capability of mobile terminal, memory headroom and rendering capability are limited, should try one's best and subtract for the overall fluency of guarantee system
The data volume of mini Mod resource.The present invention by reducing model quantity and face number, merging and shared textures, using normal map and
Light textures etc. embody the methods such as model detail and model are optimized.
Head part's details is more, and often face and number of vertex are very high for three-dimensional face true to nature.The present invention with scan gained face
Based on ZBrush softwares carry out details engraving after, to high mould carry out again topology connect up, by total face numerical control system 2,000
Within 500, and it is more sparse to be distributed on the summit of face position more intensive remaining position, while realizing reduction face number compared with
Retain face detail well.Hair model uses similar approach, first passes through high-precision model and embodies hair level and texture, then logical
Cross the required face number that automatic or manual reduces Hair model.
Textures are another valuable sources in the present invention, and generally each model correspondence 1-2 puts up figure.Textures for mobile terminal
Resolution ratio can suitably be reduced to reduce stock number, different models are divided by rational UV realizes that textures are shared.Clothes is concavo-convex such as
Fruit needs to segment many summits and dough sheet by model performance, realizes visual concave-convex surface effect to subtract with normal map
Few data volume.Strengthen static scene lighting effect with light-maps simultaneously takes a fancy to static models by less performance consumption
Go it is truer, abundant, more have third dimension.
The function that the present invention is realized is specific as follows:
The present invention provides the user more real virtual examination hair style experience, and the three-dimensional people of user is realized on function innovation
Face emulation, hair style and accessories are changed and the function such as personal information storage, and are the interactions based on mobile terminal.
User's three-dimensional facial reconstruction is realized, user is shot after mug shot, and system generates corresponding human face three-dimensional model.
Three-dimensional facial reconstruction algorithm based on single photo provides basis, the actual effect of more real embodiment difference hair style for user's examination hair style
Really.
Realize hair style and accessories are changed, system loads hair style and accessories in server etc. and selective, after selection
The file of system loads correspondence material, and instantiate display.
The storage of users personal data is realized, selected hair style data are stored in staging server catalogue by system, and will
The convenient storage later stage checks again in its corresponding informance placement database.
Preferred embodiment of the invention described in detail above.It should be appreciated that one of ordinary skill in the art without
Need creative work just can make many modifications and variations according to the design of the present invention.Therefore, all technologies in the art
Personnel are available by logical analysis, reasoning, or a limited experiment on the basis of existing technology under this invention's idea
Technical scheme, all should be in the protection domain being defined in the patent claims.
Claims (8)
1. a kind of three-dimensional examination hair style facial reconstruction method based on mobile platform, it is characterised in that comprise the following steps:
The three-dimensional face storehouse of shape of face grid and basic face texture mapping is set up, for facial image mesh reconstruction and texture mapping
Rebuild, mesh reconstruction target is obtains shape of face fitting parameter, and texture mapping rebuilds the face texture patch obtained based on user picture
Figure;
Facial image is gathered, the phase of face mask position and eyebrow in image, eyes, nose and face outline position is identified
Characteristic point is answered, user's shape of face is determined according to face mask characteristic point, the image for determining to extract according to face outline position characteristic point
Elemental range;
Determine after shape of face data, shape of face fitting parameter is calculated according to shape of face fitting formula, mesh reconstruction is realized;Determine face position
Postpone, extract pictorial element, by obtaining new face texture mapping after anamorphose algorithm and graph cut processing;
Grid after reconstruction and texture mapping are merged, instantiation display can obtain complete user's three-dimensional face.
2. a kind of three-dimensional examination hair style facial reconstruction method based on mobile platform as claimed in claim 1, its feature exists
In described to set up shape of face grid and the three-dimensional face storehouse of basic face texture mapping is specially:
The preliminary acquisition of three-dimensional face is realized by FaceShift Studio combination Kinect cameras, with 3dsMax,
Zbrush softwares are adjusted and corrected to the face feature position of basic grid, using Photoshop according to existing UV expansion
Figure repaints the basic textures material for fusion, and the grid model after finishing is combined with textures, completes basic shape of face three
Tie up the foundation of face database.
3. a kind of three-dimensional examination hair style facial reconstruction method based on mobile platform as claimed in claim 2, its feature exists
In the three-dimensional face storehouse includes long face, oval face, square face, pyriform face, rhombus face, heart-shaped face and the basic face of this seven kinds of round face
Type grid model and a basic face texture mapping, wherein seven kinds of shape of face grid models have marked face mask correspondence position in advance
Characteristic point is put, by these projecting characteristic points to plane XOY, reference characteristic point T is obtained after normalization1,T2,T3,...,T7。
4. a kind of three-dimensional examination hair style facial reconstruction method based on mobile platform as claimed in claim 1, its feature exists
In, it is described to determine after shape of face data, fitting parameter is calculated according to shape of face fitting formula, mesh reconstruction is realized, is specially:According to
The facial image of input, identifies face mask characteristic point, and contour feature point coordinates is normalized, and obtains vector
Tinput, by seven kinds of standard shape of face model projections in face shape of face storehouse to XOY plane, normalized obtains T1, T2, T3…
T7According to formula:
<mrow>
<msub>
<mi>T</mi>
<mrow>
<mi>i</mi>
<mi>n</mi>
<mi>p</mi>
<mi>u</mi>
<mi>t</mi>
</mrow>
</msub>
<mo>=</mo>
<mover>
<mi>T</mi>
<mo>&OverBar;</mo>
</mover>
<mo>+</mo>
<msub>
<mi>&alpha;</mi>
<mn>1</mn>
</msub>
<msub>
<mi>&Delta;T</mi>
<mn>1</mn>
</msub>
<mo>+</mo>
<msub>
<mi>&alpha;</mi>
<mn>2</mn>
</msub>
<msub>
<mi>&Delta;T</mi>
<mn>2</mn>
</msub>
<mo>+</mo>
<mo>...</mo>
<mo>+</mo>
<msub>
<mi>&alpha;</mi>
<mn>7</mn>
</msub>
<msub>
<mi>&Delta;T</mi>
<mn>7</mn>
</msub>
</mrow>
Wherein,And α1,α2,...,α7For unknown number to be asked, solveObtain corresponding
Shape of face fitting parameter, parameter sends the reconstruction that mobile terminal completes grid to.
5. a kind of three-dimensional examination hair style facial reconstruction method based on mobile platform as claimed in claim 1, its feature exists
In, after the determination face position, pictorial element is extracted, new face texture mapping is obtained after being handled by image algorithm,
Specially:
According to feature point detection to face position, corresponding facial image region is corroded, gained region after interception corrosion
The image co-registration after the deformation operation of rotation and scaling, deformation operation is carried out to basic textures material, through past bloom and the moon
Shadow is the user's face textures that can be rebuild.
6. a kind of three-dimensional examination hair style facial reconstruction method based on mobile platform as claimed in claim 5, its feature exists
In, it is described to be corroded corresponding facial image region, be specially:
Set different corrosion number of times to reach corrosive effect for facial image different piece.
7. a kind of three-dimensional examination hair style facial reconstruction method based on mobile platform as claimed in claim 5, its feature exists
In gained region carries out the deformation operation of rotation and scaling after the interception corrosion, is specially:
To interception corrosion after gained region rotated and scaling deformation operation, while facial characteristics control point also can according to put down
Move the coefficient of scaling to move, according to the front and rear change at a control point, inverse distance-weighting calculates and obtains remaining all control
The transformation matrices of point are made, wherein interpolating function is:
<mrow>
<mover>
<mi>x</mi>
<mo>&OverBar;</mo>
</mover>
<mo>=</mo>
<mfrac>
<mrow>
<mfrac>
<mn>1</mn>
<mrow>
<mi>d</mi>
<mi>i</mi>
<mi>s</mi>
<mn>1</mn>
</mrow>
</mfrac>
<mo>*</mo>
<mi>d</mi>
<mi>i</mi>
<mi>s</mi>
<mn>1</mn>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<mfrac>
<mn>1</mn>
<mrow>
<mi>d</mi>
<mi>i</mi>
<mi>s</mi>
<mn>1</mn>
</mrow>
</mfrac>
<mo>+</mo>
<mn>.....</mn>
<mo>+</mo>
<mfrac>
<mn>1</mn>
<mrow>
<mi>d</mi>
<mi>i</mi>
<mi>s</mi>
<mi>N</mi>
</mrow>
</mfrac>
</mrow>
</mfrac>
<mo>+</mo>
<mn>....</mn>
<mo>+</mo>
<mfrac>
<mrow>
<mfrac>
<mi>N</mi>
<mrow>
<mi>d</mi>
<mi>i</mi>
<mi>s</mi>
<mn>1</mn>
</mrow>
</mfrac>
<mo>*</mo>
<mi>d</mi>
<mi>i</mi>
<mi>s</mi>
<mi>N</mi>
<mrow>
<mo>(</mo>
<mi>x</mi>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<mfrac>
<mn>1</mn>
<mrow>
<mi>d</mi>
<mi>i</mi>
<mi>s</mi>
<mn>1</mn>
</mrow>
</mfrac>
<mo>+</mo>
<mn>.....</mn>
<mo>+</mo>
<mfrac>
<mn>1</mn>
<mrow>
<mi>d</mi>
<mi>i</mi>
<mi>s</mi>
<mi>N</mi>
</mrow>
</mfrac>
</mrow>
</mfrac>
</mrow>
1...N it is to represent the control point beyond four angles of picture, dis1 ... disN represents the point distance controlling currently to be calculated
Point 1 ... N distance;
To the reverse interpolation of blank progress in image due to stretching formation after deformation.
8. a kind of three-dimensional examination hair style facial reconstruction method based on mobile platform as claimed in claim 5, its feature exists
In in image co-registration after the deformation operation to basic textures material, Image Fusion is graph cut algorithm.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710506496.3A CN107274493B (en) | 2017-06-28 | 2017-06-28 | Three-dimensional virtual trial type face reconstruction method based on mobile platform |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710506496.3A CN107274493B (en) | 2017-06-28 | 2017-06-28 | Three-dimensional virtual trial type face reconstruction method based on mobile platform |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107274493A true CN107274493A (en) | 2017-10-20 |
CN107274493B CN107274493B (en) | 2020-06-19 |
Family
ID=60071180
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710506496.3A Active CN107274493B (en) | 2017-06-28 | 2017-06-28 | Three-dimensional virtual trial type face reconstruction method based on mobile platform |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107274493B (en) |
Cited By (29)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108171789A (en) * | 2017-12-21 | 2018-06-15 | 迈吉客科技(北京)有限公司 | A kind of virtual image generation method and system |
CN108537861A (en) * | 2018-04-09 | 2018-09-14 | 网易(杭州)网络有限公司 | Textures generation method, device, equipment and storage medium |
CN108711180A (en) * | 2018-05-02 | 2018-10-26 | 北京市商汤科技开发有限公司 | Makeups/generation and makeups of special efficacy of changing face program file packet/special efficacy of changing face generation method and device |
CN108776983A (en) * | 2018-05-31 | 2018-11-09 | 北京市商汤科技开发有限公司 | Based on the facial reconstruction method and device, equipment, medium, product for rebuilding network |
CN109035380A (en) * | 2018-09-11 | 2018-12-18 | 北京旷视科技有限公司 | Face method of modifying, device, equipment and storage medium based on three-dimensional reconstruction |
CN109377557A (en) * | 2018-11-26 | 2019-02-22 | 中山大学 | Real-time three-dimensional facial reconstruction method based on single frames facial image |
CN109410315A (en) * | 2018-08-31 | 2019-03-01 | 南昌理工学院 | Hair styling method, device, readable storage medium storing program for executing and intelligent terminal |
CN109409274A (en) * | 2018-10-18 | 2019-03-01 | 广州云从人工智能技术有限公司 | A kind of facial image transform method being aligned based on face three-dimensional reconstruction and face |
CN109523345A (en) * | 2018-10-18 | 2019-03-26 | 河海大学常州校区 | WebGL virtual fitting system and method based on virtual reality technology |
CN109671159A (en) * | 2018-12-26 | 2019-04-23 | 贵州锦微科技信息有限公司 | The virtual try-in method of ethnic group's hairdressing based on 3D VR technology |
CN109685892A (en) * | 2018-12-31 | 2019-04-26 | 南京邮电大学盐城大数据研究院有限公司 | A kind of quick 3D face building system and construction method |
CN109857311A (en) * | 2019-02-14 | 2019-06-07 | 北京达佳互联信息技术有限公司 | Generate method, apparatus, terminal and the storage medium of human face three-dimensional model |
CN109859134A (en) * | 2019-01-30 | 2019-06-07 | 珠海天燕科技有限公司 | A kind of processing method and terminal of makeups material |
CN109886144A (en) * | 2019-01-29 | 2019-06-14 | 深圳市云之梦科技有限公司 | Virtual examination forwarding method, device, computer equipment and storage medium |
CN110009725A (en) * | 2019-03-06 | 2019-07-12 | 浙江大学 | Facial reconstruction method based on multiple RGB images |
CN110021064A (en) * | 2019-03-07 | 2019-07-16 | 李辉 | A kind of aestheticism face system and method |
CN110120053A (en) * | 2019-05-15 | 2019-08-13 | 北京市商汤科技开发有限公司 | Face's dressing processing method, device and equipment |
CN110544149A (en) * | 2019-08-06 | 2019-12-06 | 尚尚珍宝(北京)网络科技有限公司 | Virtual wearing method and device of wearable product |
CN110543826A (en) * | 2019-08-06 | 2019-12-06 | 尚尚珍宝(北京)网络科技有限公司 | Image processing method and device for virtual wearing of wearable product |
CN111179411A (en) * | 2019-11-25 | 2020-05-19 | 郭宗源 | Visual facial cosmetology plastic simulation method, system and equipment based on social platform |
CN111640182A (en) * | 2020-04-20 | 2020-09-08 | 南京征帆信息科技有限公司 | Wall surface texture drawing system and method |
CN111833236A (en) * | 2019-04-21 | 2020-10-27 | 未来市股份有限公司 | Method and device for generating three-dimensional face model simulating user |
CN111899159A (en) * | 2020-07-31 | 2020-11-06 | 北京百度网讯科技有限公司 | Method, device, apparatus and storage medium for changing hairstyle |
CN112052158A (en) * | 2020-08-05 | 2020-12-08 | 腾讯科技(成都)有限公司 | Art resource operation information acquisition method and device |
CN112116699A (en) * | 2020-08-14 | 2020-12-22 | 浙江工商大学 | Real-time real-person virtual trial sending method based on 3D face tracking |
CN112164135A (en) * | 2020-09-30 | 2021-01-01 | 山西智优利民健康管理咨询有限公司 | Virtual character image construction device and method |
CN112991523A (en) * | 2021-04-02 | 2021-06-18 | 福建天晴在线互动科技有限公司 | Efficient and automatic hair matching head type generation method and generation device thereof |
CN113724396A (en) * | 2021-09-10 | 2021-11-30 | 广州帕克西软件开发有限公司 | Virtual face-lifting method and device based on face mesh |
CN117389676A (en) * | 2023-12-13 | 2024-01-12 | 成都白泽智汇科技有限公司 | Intelligent hairstyle adaptive display method based on display interface |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102663820A (en) * | 2012-04-28 | 2012-09-12 | 清华大学 | Three-dimensional head model reconstruction method |
CN103065360A (en) * | 2013-01-16 | 2013-04-24 | 重庆绿色智能技术研究院 | Generation method and generation system of hair style effect pictures |
CN104157010A (en) * | 2014-08-29 | 2014-11-19 | 厦门幻世网络科技有限公司 | 3D human face reconstruction method and device |
CN104376594A (en) * | 2014-11-25 | 2015-02-25 | 福建天晴数码有限公司 | Three-dimensional face modeling method and device |
CN105045968A (en) * | 2015-06-30 | 2015-11-11 | 青岛理工大学 | Hair style design method and system |
CN106652025A (en) * | 2016-12-20 | 2017-05-10 | 五邑大学 | Three-dimensional face modeling method and three-dimensional face modeling printing device based on video streaming and face multi-attribute matching |
-
2017
- 2017-06-28 CN CN201710506496.3A patent/CN107274493B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102663820A (en) * | 2012-04-28 | 2012-09-12 | 清华大学 | Three-dimensional head model reconstruction method |
CN103065360A (en) * | 2013-01-16 | 2013-04-24 | 重庆绿色智能技术研究院 | Generation method and generation system of hair style effect pictures |
CN104157010A (en) * | 2014-08-29 | 2014-11-19 | 厦门幻世网络科技有限公司 | 3D human face reconstruction method and device |
CN104376594A (en) * | 2014-11-25 | 2015-02-25 | 福建天晴数码有限公司 | Three-dimensional face modeling method and device |
CN105045968A (en) * | 2015-06-30 | 2015-11-11 | 青岛理工大学 | Hair style design method and system |
CN106652025A (en) * | 2016-12-20 | 2017-05-10 | 五邑大学 | Three-dimensional face modeling method and three-dimensional face modeling printing device based on video streaming and face multi-attribute matching |
Non-Patent Citations (3)
Title |
---|
DETLEF RUPRECHT,HEINRICH MULLER: ""Image Warping with Scattered Data Interpolation"", 《IEEE COMPUTER GRAPHICS AND APPLICATIONS》 * |
王超: ""基于单视图人脸重建的移动平台发型置换引擎的研究和应用"", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
龚勋: ""基于单张二维图片的三维人脸建模"", 《中国博士学位论文全文数据库 信息科技辑》 * |
Cited By (43)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108171789A (en) * | 2017-12-21 | 2018-06-15 | 迈吉客科技(北京)有限公司 | A kind of virtual image generation method and system |
CN108171789B (en) * | 2017-12-21 | 2022-01-18 | 迈吉客科技(北京)有限公司 | Virtual image generation method and system |
CN108537861A (en) * | 2018-04-09 | 2018-09-14 | 网易(杭州)网络有限公司 | Textures generation method, device, equipment and storage medium |
CN108711180A (en) * | 2018-05-02 | 2018-10-26 | 北京市商汤科技开发有限公司 | Makeups/generation and makeups of special efficacy of changing face program file packet/special efficacy of changing face generation method and device |
CN108711180B (en) * | 2018-05-02 | 2021-08-06 | 北京市商汤科技开发有限公司 | Method and device for generating makeup and/or face-changing special effect program file package and method and device for generating makeup and/or face-changing special effect |
CN108776983A (en) * | 2018-05-31 | 2018-11-09 | 北京市商汤科技开发有限公司 | Based on the facial reconstruction method and device, equipment, medium, product for rebuilding network |
CN109410315A (en) * | 2018-08-31 | 2019-03-01 | 南昌理工学院 | Hair styling method, device, readable storage medium storing program for executing and intelligent terminal |
CN109035380B (en) * | 2018-09-11 | 2023-03-10 | 北京旷视科技有限公司 | Face modification method, device and equipment based on three-dimensional reconstruction and storage medium |
CN109035380A (en) * | 2018-09-11 | 2018-12-18 | 北京旷视科技有限公司 | Face method of modifying, device, equipment and storage medium based on three-dimensional reconstruction |
CN109409274A (en) * | 2018-10-18 | 2019-03-01 | 广州云从人工智能技术有限公司 | A kind of facial image transform method being aligned based on face three-dimensional reconstruction and face |
CN109523345A (en) * | 2018-10-18 | 2019-03-26 | 河海大学常州校区 | WebGL virtual fitting system and method based on virtual reality technology |
CN109409274B (en) * | 2018-10-18 | 2020-09-04 | 四川云从天府人工智能科技有限公司 | Face image transformation method based on face three-dimensional reconstruction and face alignment |
CN109377557B (en) * | 2018-11-26 | 2022-12-27 | 中山大学 | Real-time three-dimensional face reconstruction method based on single-frame face image |
CN109377557A (en) * | 2018-11-26 | 2019-02-22 | 中山大学 | Real-time three-dimensional facial reconstruction method based on single frames facial image |
CN109671159A (en) * | 2018-12-26 | 2019-04-23 | 贵州锦微科技信息有限公司 | The virtual try-in method of ethnic group's hairdressing based on 3D VR technology |
CN109685892A (en) * | 2018-12-31 | 2019-04-26 | 南京邮电大学盐城大数据研究院有限公司 | A kind of quick 3D face building system and construction method |
CN109886144B (en) * | 2019-01-29 | 2021-08-13 | 深圳市云之梦科技有限公司 | Virtual trial sending method and device, computer equipment and storage medium |
CN109886144A (en) * | 2019-01-29 | 2019-06-14 | 深圳市云之梦科技有限公司 | Virtual examination forwarding method, device, computer equipment and storage medium |
CN109859134A (en) * | 2019-01-30 | 2019-06-07 | 珠海天燕科技有限公司 | A kind of processing method and terminal of makeups material |
CN109857311A (en) * | 2019-02-14 | 2019-06-07 | 北京达佳互联信息技术有限公司 | Generate method, apparatus, terminal and the storage medium of human face three-dimensional model |
CN110009725A (en) * | 2019-03-06 | 2019-07-12 | 浙江大学 | Facial reconstruction method based on multiple RGB images |
CN110021064A (en) * | 2019-03-07 | 2019-07-16 | 李辉 | A kind of aestheticism face system and method |
CN111833236A (en) * | 2019-04-21 | 2020-10-27 | 未来市股份有限公司 | Method and device for generating three-dimensional face model simulating user |
CN111833236B (en) * | 2019-04-21 | 2023-07-07 | 未来市股份有限公司 | Method and device for generating three-dimensional face model for simulating user |
CN110120053A (en) * | 2019-05-15 | 2019-08-13 | 北京市商汤科技开发有限公司 | Face's dressing processing method, device and equipment |
CN110544149A (en) * | 2019-08-06 | 2019-12-06 | 尚尚珍宝(北京)网络科技有限公司 | Virtual wearing method and device of wearable product |
CN110543826A (en) * | 2019-08-06 | 2019-12-06 | 尚尚珍宝(北京)网络科技有限公司 | Image processing method and device for virtual wearing of wearable product |
CN111179411B (en) * | 2019-11-25 | 2023-03-28 | 郭宗源 | Visual facial cosmetology plastic simulation method, system and equipment based on social platform |
CN111179411A (en) * | 2019-11-25 | 2020-05-19 | 郭宗源 | Visual facial cosmetology plastic simulation method, system and equipment based on social platform |
CN111640182B (en) * | 2020-04-20 | 2023-04-07 | 南京征帆信息科技有限公司 | Wall surface texture drawing system and method |
CN111640182A (en) * | 2020-04-20 | 2020-09-08 | 南京征帆信息科技有限公司 | Wall surface texture drawing system and method |
CN111899159A (en) * | 2020-07-31 | 2020-11-06 | 北京百度网讯科技有限公司 | Method, device, apparatus and storage medium for changing hairstyle |
CN111899159B (en) * | 2020-07-31 | 2023-12-22 | 北京百度网讯科技有限公司 | Method, device, apparatus and storage medium for changing hairstyle |
CN112052158B (en) * | 2020-08-05 | 2022-09-30 | 腾讯科技(成都)有限公司 | Art resource operation information acquisition method and device |
CN112052158A (en) * | 2020-08-05 | 2020-12-08 | 腾讯科技(成都)有限公司 | Art resource operation information acquisition method and device |
CN112116699A (en) * | 2020-08-14 | 2020-12-22 | 浙江工商大学 | Real-time real-person virtual trial sending method based on 3D face tracking |
CN112116699B (en) * | 2020-08-14 | 2023-05-16 | 浙江工商大学 | Real-time real-person virtual trial sending method based on 3D face tracking |
CN112164135A (en) * | 2020-09-30 | 2021-01-01 | 山西智优利民健康管理咨询有限公司 | Virtual character image construction device and method |
CN112991523A (en) * | 2021-04-02 | 2021-06-18 | 福建天晴在线互动科技有限公司 | Efficient and automatic hair matching head type generation method and generation device thereof |
CN112991523B (en) * | 2021-04-02 | 2023-06-30 | 福建天晴在线互动科技有限公司 | Efficient and automatic hair matching head shape generation method and generation device thereof |
CN113724396A (en) * | 2021-09-10 | 2021-11-30 | 广州帕克西软件开发有限公司 | Virtual face-lifting method and device based on face mesh |
CN117389676A (en) * | 2023-12-13 | 2024-01-12 | 成都白泽智汇科技有限公司 | Intelligent hairstyle adaptive display method based on display interface |
CN117389676B (en) * | 2023-12-13 | 2024-02-13 | 成都白泽智汇科技有限公司 | Intelligent hairstyle adaptive display method based on display interface |
Also Published As
Publication number | Publication date |
---|---|
CN107274493B (en) | 2020-06-19 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107274493A (en) | A kind of three-dimensional examination hair style facial reconstruction method based on mobile platform | |
CN109377557B (en) | Real-time three-dimensional face reconstruction method based on single-frame face image | |
Jung et al. | Sketching folds: Developable surfaces from non-planar silhouettes | |
CN110335343A (en) | Based on RGBD single-view image human body three-dimensional method for reconstructing and device | |
Lu et al. | Illustrative interactive stipple rendering | |
US7953275B1 (en) | Image shader for digital image modification | |
US8019182B1 (en) | Digital image modification using pyramid vignettes | |
Wu et al. | Interactive normal reconstruction from a single image | |
CN110197462A (en) | A kind of facial image beautifies in real time and texture synthesis method | |
CN111563951B (en) | Map generation method, device, electronic equipment and storage medium | |
CN106373178A (en) | Method and apparatus for generating an artificial picture | |
Li et al. | In-home application (App) for 3D virtual garment fitting dressing room | |
CN108805090A (en) | A kind of virtual examination cosmetic method based on Plane Gridding Model | |
Hudon et al. | Deep normal estimation for automatic shading of hand-drawn characters | |
CN105913496B (en) | It is a kind of by true dress ornament rapid translating be three-dimensional dress ornament method and system | |
NZ551124A (en) | Dynamic Wrinkle mapping | |
CN106855987A (en) | Sense of reality Fashion Show method and apparatus based on model prop | |
CN109377537A (en) | Style transfer method for heavy color painting | |
Verhoeven | Computer graphics meets image fusion: The power of texture baking to simultaneously visualise 3D surface features and colour | |
WO2020104990A1 (en) | Virtually trying cloths & accessories on body model | |
CN101477705B (en) | Computer assisted character animation drawing method based on light irradiated ball model | |
KR100828935B1 (en) | Method of Image-based Virtual Draping Simulation for Digital Fashion Design | |
Li et al. | Image stylization with enhanced structure on GPU | |
Kerber et al. | Real-time generation of digital bas-reliefs | |
Li et al. | ARF-Plus: Controlling Perceptual Factors in Artistic Radiance Fields for 3D Scene Stylization |
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 |