CN107590708A - A kind of method and apparatus for generating the specific bodily form model of user - Google Patents
A kind of method and apparatus for generating the specific bodily form model of user Download PDFInfo
- Publication number
- CN107590708A CN107590708A CN201610534953.5A CN201610534953A CN107590708A CN 107590708 A CN107590708 A CN 107590708A CN 201610534953 A CN201610534953 A CN 201610534953A CN 107590708 A CN107590708 A CN 107590708A
- Authority
- CN
- China
- Prior art keywords
- model
- user
- bodily form
- human body
- section
- 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
Landscapes
- Processing Or Creating Images (AREA)
Abstract
The invention discloses a kind of method and apparatus for generating the specific bodily form model of user, belong to virtual fitting field.Methods described includes：The bodily form characteristic information inputted according to user, determines basic manikin corresponding to the bodily form characteristic information in the bodily form Sample Storehouse prestored；The actual human body image uploaded to the user is carried out calculating processing generation sketch figure picture, and human bodily form's data of the user are obtained from the sketch figure picture；According to human bodily form's data, deformation adjustment is carried out to the basic manikin, generates the specific bodily form model of user.Using the present invention, the matching degree of the specific bodily form model of user and the true bodily form can be improved.
Description
Technical field
It is more particularly to a kind of to generate the specific bodily form model method of user and device the present invention relates to virtual fitting field.
Background technology
With the continuous development of network technology, shopping online has turned into a kind of shopping indispensable in people's daily life
Mode, increasing people are keen in online purchase commodity, such as clothes, food.
Virtual fitting platform is generally provided with general human body model, and general human body model is according to standard by technical staff
The dummy model that the Body Profile data of human body is established on computers.User is in online purchase clothes, it is desirable to provide the height of oneself,
The bodily form characteristic information such as body weight, measurements of the chest, waist and hips, then virtual fitting platform can be according to these bodily form characteristic informations to general human body mould
Type is adjusted, so as to obtain generally conforming to the specific bodily form model of user of user's bodily form.
During the present invention is realized, inventor has found that prior art at least has problems with：
Because the bodily form of people is influenceed by many factors such as age, region, heredity, simply according to height, body weight, measurements of the chest, waist and hips
The specific bodily form model of user for adjusting to obtain Deng bodily form characteristic information can not accurately embody the true bodily form of different people, so,
The matching degree of the specific bodily form model of user and the true bodily form is poor.
The content of the invention
In order to solve problem of the prior art, the embodiments of the invention provide a kind of side for generating the specific bodily form model of user
Method and device.The technical scheme is as follows：
First aspect, there is provided a kind of method for generating the specific bodily form model of user, methods described include：
The bodily form characteristic information inputted according to user, the bodily form feature letter is determined in the bodily form Sample Storehouse prestored
Basic manikin corresponding to breath；
The actual human body image uploaded to the user carries out calculating processing generation sketch figure picture, and from the sketch figure picture
The middle human bodily form's data for obtaining the user；
According to human bodily form's data, deformation adjustment is carried out to the basic manikin, generates the specific bodily form of user
Model.
Second aspect, there is provided a kind of device for generating the specific bodily form model of user, described device include：
Determining module, for the bodily form characteristic information inputted according to user, determined in the bodily form Sample Storehouse prestored
Basic manikin corresponding to the bodily form characteristic information；
Acquisition module, the actual human body image for being uploaded to the user carry out calculating processing generation sketch figure picture, and
Human bodily form's data of the user are obtained from the sketch figure picture；
Generation module, for according to human bodily form's data, carrying out deformation adjustment to the basic manikin, generating
The specific bodily form model of user.
The beneficial effect that technical scheme provided in an embodiment of the present invention is brought is：
It is true in the bodily form Sample Storehouse prestored according to the bodily form characteristic information of user's input in the embodiment of the present invention
Determine basic manikin corresponding to bodily form characteristic information, the actual human body image uploaded to user carries out calculating processing generation outline
Image, and human bodily form's data of user are obtained from sketch figure picture, according to human bodily form's data, basic manikin is carried out
Deformation adjustment, generates the specific bodily form model of user.So, server can be real according to the actual human body image pair of user and user
The bodily form more similar basic manikin in border carries out deformation process, so as to generation the specific bodily form model of user and user it is true
The matching degree of entity shape is higher.
Brief description of the drawings
Technical scheme in order to illustrate the embodiments of the present invention more clearly, make required in being described below to embodiment
Accompanying drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the present invention, for
For those of ordinary skill in the art, on the premise of not paying creative work, other can also be obtained according to these accompanying drawings
Accompanying drawing.
Fig. 1 is a kind of method flow diagram for generating the specific bodily form model of user provided in an embodiment of the present invention；
Fig. 2 is a kind of schematic diagram of actual human body image provided in an embodiment of the present invention；
Fig. 3 is a kind of principle schematic for determining actual human body size provided in an embodiment of the present invention；
Fig. 4 is a kind of rough schematic for choosing user people's body section provided in an embodiment of the present invention；
Fig. 5 is a kind of schematic diagram of grid provided in an embodiment of the present invention；
Fig. 6 is a kind of principle schematic of plastic deformation provided in an embodiment of the present invention；
Fig. 7 is a kind of apparatus structure schematic diagram for generating the specific bodily form model of user provided in an embodiment of the present invention；
Fig. 8 is a kind of structural representation of server provided in an embodiment of the present invention.
Embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with accompanying drawing to embodiment party of the present invention
Formula is described in further detail.
The embodiments of the invention provide a kind of method for generating the specific bodily form model of user, the executive agent of this method is clothes
Business device.Wherein, server can be virtual fitting platform corresponding to server, possess three-dimensional modeling and image processing function.Clothes
Processor, memory, screen etc. can be provided with business device.Processor can be used for the process for generating the specific bodily form model of user
Handled, memory can be used for the number for needing to use or storing during being stored in the generation specific bodily form model of user
According to screen is displayed for image or model.In the present embodiment, so that server is computer as an example, the detailed of scheme is carried out
Illustrate, other situations are similar therewith, and the present embodiment is not repeated.
Below in conjunction with embodiment, the handling process shown in Fig. 1 is described in detail, content can be as
Under：
Step 101, the bodily form characteristic information inputted according to user, determine that the bodily form is special in the bodily form Sample Storehouse prestored
Basic manikin corresponding to reference breath.
In force, it is contemplated that many factors such as all ages and classes, sex, region and hobby have larger to user's bodily form
Influence.Such as northerner's bodily form is more tall and big, southerner's bodily form is more modest, and teenager's body is more tall and straight, old human body
Inclined rickets, likes that the leg muscle of running is more strong, and the upper body muscle ratio for liking boxing is more flourishing etc..So it can utilize
The crowd that three-dimensional scanning device has characteristic feature to all ages and classes, hobby, region, sex etc. carries out three-dimensional bodily form scanning, so
Afterwards using these scanning three-dimensional modeling datas build a bodily form Sample Storehouse, i.e., can be established in the bodily form Sample Storehouse has age,
Hobby, region, sex etc. and the corresponding relation of basic manikin.These will by the basic manikin of 3-D scanning generation
Optimised processing, make its topological structure more specification, so that basic manikin is in three-dimensional animation and physical simulation simulation neck
Application in domain.So, when the specific bodily form model of desired generation user, user, which can first input, meets the bodily form feature of itself
Age of information, i.e. user, hobby, region etc., then server can be existed according to the bodily form characteristic information that above-mentioned user inputs
Basic manikin corresponding to being searched in bodily form Sample Storehouse.It is noted that each basic manikin can include presetting
Number surface vertices, each surface vertices are provided with corresponding summit sequence number
Step 102, the actual human body image uploaded to user carries out calculating processing generation sketch figure picture, and from sketch figure picture
The middle human bodily form's data for obtaining user.
In force, user can also upload the actual human body image of oneself, that is, upload after bodily form characteristic information is inputted
The photo of the user human bodily form of shooting.When shooting photo, in order to reduce the error of subsequent treatment, the camera lens of camera should be with
The center of human body keeps camera lens to look squarely in same level and vertical plane, no pitching perspective, meanwhile, user can wear
Underclothes, are easy to actual human body image accurately to reflect the actual bodily form of user.In addition, user should be according to default appearance
Gesture is shot, as shown in Fig. 2 when shooting front is shone, both arms are flattened, the posture of two leg differential standing, when being shone in shooting side,
Arms down, cling to the body, legs joined, so can effectively reduce the error caused by various block.It is noted that
One can also be set in the picture of shooting has standard-sized object of reference, so can be according to reference in actual human body image
The picture size and standard size of thing determine the proportionate relationship of actual human body image and the actual bodily form of user.
Afterwards, the actual human body image that server can upload to user carries out calculating processing generation sketch figure picture, in life
In silhouette after image, server can determine the people of user according to the outline data and aforementioned proportion relation of human body in sketch figure picture
Body Body Profile data.Specifically, the width and height of known actual human body image：Wi, Hi (unit：Pixel), photoreceptor using mm as
The width and height of unit：Ws, Hs (unit：Mm), then projection zoom factor M=(Wi from actual human body image to photoreceptor2
+Hi2)/(Ws2+Hs2)；Afterwards, can be according to size a " of the object of reference in digital photograph (i.e. actual human body image) and reality
Size b " of the human body in digital photograph, projection size a=a " × M and reality of the object of reference on photoreceptor are calculated respectively
Projection size b=b " × M of the human body on photoreceptor；The actual size A and actual persons of focal length f, object of reference further according to camera
Offset distance s between body and object of reference, human body actual size B=b (Af/a-s)/f can be calculated, can refer to Fig. 3.
The exemplary method for giving two kinds of feasible generation sketch figure pictures herein：1. subtract the back of the body with actual human body image
Scape image, specifically, user can first shoot and upload one with seat in the plane background image, herein, refer to camera place with seat in the plane
Locus, shooting angle etc. keep identical, do not include user in background image, remainder all with actual human body image
It is identical, so, after server gets actual human body image and background image, actual human body image can be entered with background image
Row contrast, wherein the region differed is exactly human body region, so as to which actual human body image is converted into bitmap images,
The pixel value of the pixel in the same area in actual human body image and background image is set to 0, the picture in different zones
The pixel value of vegetarian refreshments is set to 1, so as to generate sketch figure picture.2. the lookup pattern recognition method of existing contour line can be used true
The contour line of human body in actual human body image is determined, so as to which actual human body image is converted into bitmap images, i.e., by contour line
Within region in the pixel value of pixel be set to 1, the pixel value of the pixel in region outside contour line is set to 0, from
And generate sketch figure picture.In addition, during sketch figure picture is generated, server can provide a user the work(of amending image
Can, i.e., user can be adjusted by adjusting the method for Bezier to actual human body image, using Bezier circle
Go out human body region in actual human body image, unnecessary image-region can so be deleted, to reduce the complexity calculated；
Meanwhile after sketch figure picture is generated, server can show sketch figure picture user, user can pass through bezier curve pair
Sketch figure picture is finely adjusted, so that it is guaranteed that sketch figure picture more conforms to the actual bodily form of user.
Optionally, actual human body image can include human body direct picture and human body side image, and server can be to people
Body takes section, and human bodily form's data are determined according to the data of everyone body section, and accordingly, the processing of step 102 can be as
Under：Preset number user people's body section, user people's body section and people are chosen along human body natural's direction of extension to the human body of user
Body nature direction of extension is perpendicular；The human body direct picture and human body side image uploaded to user carries out calculating processing generation just
Face sketch figure picture and side sketch figure picture；Corresponding to determining user people's body section in positive sketch figure picture and side sketch figure picture
Transversal section, human body width data and human depth data of the user at user people's body section are determined according to the length of transversal section.
In force, user can shoot itself direct picture and side image (i.e. human body direct picture and people respectively
Body side surface image), then human body direct picture and human body side image are uploaded onto the server.One can be first built in server
The human body of individual imaginary user, preset number and human body natural then are chosen along human body natural's direction of extension to the human body of user
The perpendicular user people's body section of direction of extension.Herein, human body can be divided into trunk, arm, neck and the part of leg four, people
Body nature direction of extension is the extension direction of vertebra, the arm exhibition direction of arm and the direction of extension of leg.For the ease of follow-up
Calculate, preset number user people's body section can be uniformly chosen along human body natural's direction of extension.Afterwards, server can basis
The human body direct picture and human body side image that user uploads carry out calculating the positive sketch figure picture of processing generation and side sketch figure
Picture.And then server can determine that above-mentioned user people's body section is corresponding in positive sketch figure picture and side sketch figure picture and cut
Line segment, human body width number of user's human body at corresponding user people's body section then can be then determined according to the length of transversal section
According to human depth's data, as shown in Figure 4.
Step 103, according to human bodily form's data, deformation adjustment is carried out to basic manikin, generates the specific bodily form of user
Model.
In force, the modeling program for building model, the mould that following servers are carried out can be installed in server
Type deformation process all may be considered server by modeling program to complete.Server obtains according to actual human body image
To after human bodily form's data of user, the above-mentioned basic manikin determined can be carried out according to human bodily form's data
Deformation adjustment, so as to generate the specific bodily form model of user.The specific bodily form model of the user can be bound with the account of user,
So, server need not be again the new specific bodily form model of user's generation within a certain period of time.In view of human bodily form at any time
Between change, therefore can be with upload pictures again more new model when user's bodily form has significant change.
Optionally, the human body length data and human body width data of each user people's body section based on above-mentioned determination, can
To carry out deformation adjustment to basic manikin.It is varied to deform the mode of adjustment, herein it is exemplary give two kinds can
Capable processing mode：
Mode one, preset number model is chosen in predeterminated position along human body natural's direction of extension to basic manikin and cut
Face, the model width data and model depth data in preset number model section are obtained, wherein, user people's body section and human body
Natural direction of extension is perpendicular；Using the center in each model section as node, model is set to bind bone, according to default covering
Standard carries out covering processing to model and bone, sets corresponding deformation weight；According to the model width data in model section and
Model depth data, and human body width data and human depth data of the user at corresponding user people's body section, it is determined that
Zoom factor；Model binding bone is zoomed in and out according to zoom factor, and according to the position adjustment model of user people's body section
The position of bone is bound, generates the specific bodily form model of user.
In force, server can be according to the mode of above-mentioned interception user people body section, to basic manikin along people
Body nature direction of extension chooses preset number model section in predeterminated position, then obtains the mould in preset number model section
Molded breadth degrees of data and model depth data.Afterwards, can be using the center in each model section as node, on basic manikin
Corresponding model binding bone is set, and measures and stores size (the i.e. mould in model section corresponding to each bone node
The model width data and model depth data in type section).It is for example, equal to the torso portion of the human body of user in a step 102
It is even to have intercepted 50 sections, then need also equably to intercept the torso portion of basic manikin in 50 sections, and set with
Corresponding 50 models binding bone.Processing can carry out automatic business processing by program above, can such as in Maya
To realize automation mechanized operation using program modules such as PyMel, Maya API, excessive description is not done herein.
Further, covering processing can be carried out to the model binding bone of above-mentioned setting according to default covering standard,
The surface vertices of basic manikin, and each model binding bone can be had influence on by setting each model binding bone respectively
For the influence degree (deforming weight) of the surface vertices of each basic manikin.Default covering standard herein is with existing
Some computer role animation binding standards are reference, and here is omitted.Server can be according to the model in model section afterwards
Human body width data and human depth's number at width data and model depth data, and corresponding user people's body section
According to it is determined that the zoom factor of each model binding bone.And then server can be according to the zoom factor of determination to corresponding mould
Type binding bone zoom in and out, and according to the position of user people's body section correspondingly adjust model binding bone node position
Put, for example, a length of 1 meter of the torso portion of basic manikin, a length of 0.8 meter of the torso portion of user's actual human body, now uniformly
Choose 21 sections, then the position of the node of each model binding bone be followed successively by 0cm, 5cm, 10cm, 100cm, and use
The position of people from family body section be followed successively by 0cm, 4cm, 8cm, 80cm, so model can be bound to the node of bone
Position adjustment to 0cm, 4cm, 8cm, at 80cm, be finally reached the apex coordinate mesh for adjusting basic manikin surface
, so as to generate the specific bodily form model of user.Cut it is noted that the model of torso portion and leg binds bone in model
Elliptical non-uniform zoom is scaled in face, and arm is similar in model section with the model binding bone of neck
The uniform zoom of cylinder.In addition, it is only necessary to which deformation gauge is carried out to arm and the wherein side of leg, and the half side trunk of left or right
Calculation, other side, which need to only make simple mirror transformation, to be completed.
Mode two：Preset number model is chosen to basic manikin in predeterminated position along human body natural's direction of extension to cut
Face, the width and depth in preset number model section are obtained, wherein, user people's body section mutually hangs down with human body natural's direction of extension
Directly；Using the center in each model section as control point, model NURBS (non-uniform rational b spline) curve is created, by basic human body
The surface vertices of model are projected on model nurbs curve, it is determined that each surface vertices correspond to relative vector and the throwing of subpoint
Nurbs curve parameter value corresponding to shadow point；Existed according to the model width data and model depth data in model section, and user
Human body width data and human depth's data at corresponding user people's body section, determine zoom factor；With each user's human body
The center in section is control point, creates user's human body nurbs curve；According to each surface vertices correspond to subpoint it is relative to
Amount, nurbs curve parameter value corresponding to zoom factor and subpoint, and user's human body nurbs curve, generate user's particular volume
Shape model.
In force, server can be according to the mode of above-mentioned interception user people body section, to basic manikin along people
Body nature direction of extension chooses preset number model section in predeterminated position, then obtains the mould in preset number model section
Molded breadth degrees of data and model depth data.Further according to the model width data and model depth data in model section, and user
Human body width data and human depth's data at corresponding user people's body section, determine zoom factor.Further, can be with
Basic manikin is divided into several independent parts, the i.e. part such as trunk, leg, arm and neck, then to each portion
Divide along human body natural's direction of extension and choose preset number model section in predeterminated position, while each several part is split at section
The summit sequence number of surface vertices over different portions is corresponding to be stored.Afterwards, for each part of basic manikin, with the portion
The center in each model section is control point corresponding to point, and model nurbs curve corresponding to establishment, i.e. model nurbs curve is worn
The center in each model section is crossed, the specific establishment process of model nurbs curve is prior art, is no longer repeated herein.
It is noted that exemplified by space coordinates is when human body is stood used by this programme：Human body frontal width direction is X-axis side
To human body vertical direction is Y direction, and people's body side surface depth direction is Z-direction.Model nurbs curve can be that space is sat
The Simultaneous Equations of X, Y, Z relative to parameter T are marked, the parameter T of model nurbs curve corresponding to each part is unified to
The section of [0,1], the i.e. parameter of the initiating terminal of model nurbs curve are 0, and the parameter for terminating end is 1.And then can be by basis
The surface vertices of manikin are projected on model nurbs curve, it is determined that each surface vertices correspond to the relative vector of subpoint
With subpoint corresponding to nurbs curve parameter value T, and relative vector and parameter T storage arrive with the top of corresponding surface vertices
In the array data structure of point serial number index.For example, for the surface vertices of torso portion, can be according to the seat of surface vertices
The function expression of model nurbs curve, calculates the coordinate of subpoint and corresponding parameter value corresponding to mark and torso portion
T.Then, the coordinate of its subpoint on nurbs curve is subtracted with the coordinate value of surface vertices, that is, obtain the surface vertices with it is right
The relative vector for the subpoint answered.
Herein, the scheme of two kinds of feasible determination subpoints is given：
Scheme one, for each surface vertices, determine that passes through the surface vertices, and with body nature direction of extension
Vector is the abovementioned mathematical plane of normal, and by taking trunk as an example, the normal of the abovementioned mathematical plane is Y-axis, then bent by model NURBS again
Line and the function expression of above-mentioned plane obtain the coordinate of the joining (i.e. subpoint) and corresponding parameter value T.Generally, body
The normal of the test for intersection abovementioned mathematical plane on dry, leg and neck surface summit is all Y-axis, and the test for intersection plane of arm segment has
Two kinds of possibility, if arm is the horizontal T position stretched, its normal is X-axis；If arm be tiltedly upper or tiltedly under T-shaped appearance
Gesture, its normal are then similar cornerwise sloping shaft.
Scheme two, because all control points of model nurbs curve are known, and the array to have sorted, for example wear
All control points for crossing torso portion are the arrays to be sorted along Y-axis.So binary chop algorithm can be utilized, according to each table
The value of vertex of surface Y-axis, first determine two control points that the value of Y-axis and the value of the surface vertices Y-axis are mutually closed on, then to pass through calculating
The linear interpolation of the parameter value at above-mentioned control point determines the parameter value T of subpoint corresponding to the surface vertices, so as to true
Determine the coordinate of subpoint.
Similarly, using the center of each user people's body section as control point, user's human body nurbs curve is created, then, according to
User's human body nurbs curve deforms to the surface vertices of above-mentioned each basic manikin.Specifically, according to each basis
The summit sequence number of the surface vertices of manikin, corresponding parameter of curve T and relative vector are retrieved, then parameter T
Substitute into the function expression of user's human body nurbs curve, obtain projection of the surface vertices on user's human body nurbs curve
Point, so can according to corresponding to each surface vertices relative vector and zoom factor, calculate the surface vertices in user people
Coordinate transforming in body nurbs curve space, generate the model of various pieces in the specific bodily form model of user.For example, model
The function expression of nurbs curve is Cmodel, the function expression of user's human body nurbs curve is Cphoto, for basic human body
Coordinate on model isSurface vertices, its projection coordinate on model nurbs curve isThe surface vertices
Relative vector with corresponding subpoint isParameter of curve corresponding to the surface vertices is Tmodel,
The size in model section corresponding to the surface vertices is Sxmodel,Symodel, the size of corresponding user people's body section is Sxphoto,
Sxphoto, then scaled matrix be
Coordinate of the subpoint corresponding to the surface vertices on user's human body nurbs curve beEnter
And coordinate of the surface vertices on the specific bodily form model of user is
The model of each several part is subjected to splicing suture again, so as to generate the specific bodily form model of user.Specifically, can be with
According to the data of photo acquisition, the model of each several part is moved to its corresponding position.Such as according to the leg gathered from photo
The leg model that deformation is completed, is moved to corresponding position by portion position, at each model interface that then basis prestores
The summit sequence numbers of surface vertices the model of different piece is united by a seam.When two tables to be stapled of the model of two parts
When the space coordinates of vertex of surface is inconsistent, these surface vertices can be carried out with deformation trim process, while in order that suturing part
Divide transitions smooth, can be to carrying out deformation process, deformation extent in a range of other surface vertices of above-mentioned surface vertices
Decayed according to the distance with above-mentioned surface vertices.In addition, the suture representative points of different model parts and deformation decay are omited
There is difference：The suture of trunk and thigh is fairly simple, if space coordinates of some surface vertices in the model of torso portionWith the space coordinates in the model of leg portionDiffer, then the space coordinates of the surface vertices can be adjusted
ForMeanwhile the decay distance of torso portion is more shorter than the decay distance of leg.Neck
It is more more complex with the joining ratio of trunk, come because the size of neck can calculate from photo, the suture of this part will be with
The summit of neck is target spot size to move the summit of torso portion.So surface vertices are in the space coordinates of torso portion
ForIt is in the space coordinates of neck portionThe space coordinates of the surface vertices can be then adjusted toTrunk will be grown by deformation decay distance of the part above close to the summit of sutured portion
Some, but neck only has very short deformation decay.Arm is similar with neck with the suture of trunk, can be with from full face
Measure the size of arm, but the size at armpit but compare be difficult to measure it is very accurate, so the surface top of this part
The space coordinates of point is mainly and the surface vertices of the shoulder of trunk and chest using the surface vertices at arm interface as foundation
Deformation decay distance it is longer.So the space coordinates of torso portion surface vertices isArm segment surface vertices
Space coordinates beThe space coordinates of suture surface verticesHead and neck
Joining ratio is more special, them can be selected to be liked the head model of some preset different facial characteristics, user in this programme
Or they think themselves closest model.The topological structure of these preset head models is all by optimization
Reason, to ensure the smooth registration with the model of other parts.In order to ensure the accuracy of clothes effect at user's neckline, suture
The surface vertices of Shi Tongchang adjustment head models are to be adapted to the neck phantom of user.So space of neck portion surface vertices
Coordinate isThe space coordinates of head portion surface vertices isThe space coordinates of suture surface verticesThe suture of other parts, as hand and foot, due to general chi can only be measured from photo
It is very little, so in this case, first by the size at wrist and ankle, and the related data that photo measures, estimate substantially
Scaling opponent and foot scaling of model, are then sutured again.
It is noted that in the suture of different models, particularly there is obvious corner, as armpit, neck shoulder connect
Place etc., the shape of model may not be especially to be consistent with the outline of user.It is similar using two kinds of principles in this case, but realize different
Method the model of suture is carried out approaching deformation with photo outline, make it more accurate.It is following two that to approach deformation method same
When be applied to binding bone and nurbs curve both model deformation methods.
Method one：The specific bodily form model of user is projected to two dimensional surface generation two-dimensional projection model, establishment etc. than two dimension
Grid, the beeline and the corresponding gradient vector of beeline of each grid node and the border of sketch figure picture are obtained, according to
Gradient vector corresponding to beeline and beeline, deformation process is carried out in two-dimensional projection's model, after deformation process
Two-dimensional projection's model, deformation process is carried out to the specific bodily form model of user.
In force, the model that suture is completed is projected respectively the two dimensional surface of front and side two, and according to this
Model creates two grades than two-dimensional network in the maximum boundary frame of two-dimensional space, and above-mentioned outline is applied to than two-dimensional grid by such
In image.First grid uses scalar grid data structure, for stored symbols apart from field data (SDF, Signed
Distance Field)；Second grid uses vectorial grid data structure, for storing what is come out from symbolic distance field computation
Gradient vector field data.First, around each grid node of scalar field, random default vector of samples of generation in 360 degree.
From current mesh node, along the pixel of each stochastical sampling vector detection outline bitmap, when pixel value changes, it was demonstrated that at this
For the border of outline, just the distance between current location and sampling node are stored to an interim array.Previous step is gathered
Range data sequence, the wherein minimum value of selection is as beeline.If the grid node of present sample is in outline
Distance can then be arranged to negative by portion, on the contrary then be positive number.Herein, the range data gathered from sketch figure picture is with pixel
It is unit, it is necessary to be switched to the unit space of model, is then stored into first scalar grid data structure and is used as symbol
Apart from field data；And the vector of samples above used without conversion, only the vector need to be converted into length be 1 unit vector i.e.
It can be used.Then, its gradient vector field data can be calculated based on symbolic distance field data, and the result storage of calculating is arrived
In second vectorial grid data structure.Its calculation formula is the gradient formula of standard：
Then, the border surface summit of the manikin of two-dimensional space is incident upon along two above, two controls are respectively created
Line processed.Line segment is connected with straight line between the summit of this two lines.By the method for linear interpolation from first scalar grid number
According to the beeline d that current control line summit is obtained in structure, and current control is obtained from second vectorial grid data structure
The gradient vector of line summit processed positionThen formula can be pressed in current control line summit
The border of outline is moved to, then is entered by the surface vertices in control pair pre-determined distance by different attenuation ratios
The flexible adjustment of row, specific processing can be as follows：Centered on each surface vertices, the control of correlation is searched in default radius R
Line summit processed, distance is calculated by "current" model summit and each control line summit, and calculated according to the range data for independent variable
Go out attenuation coefficient, the motion vector on the control line summit above calculated, the attenuation coefficient calculated by previous step is according to different
Ratio is added to the surface vertices.For example, the motion vector on current control line summit isWherein,For control
Coordinate after the movement of summit,Coordinate before being moved for control vertex, Current surface summit and the control line summit for influenceing the point
The distance betweenWherein,Coordinate before being moved for surface vertices, the attenuation coefficient on Current surface summit
Decaymi=DecayFn (Dmi/ R), then the coordinate after surface vertices movement is
Finally, when after the completion of the approaching of two-dimensional space, it is necessary to which final result is fed back to the three of model surface summit
Dimension coordinate.Such as the transformation results in X/Y plane space, will change three dimensional space coordinate X and Y component values.In addition,
Directly above-mentioned processing can be carried out using the beeline and direction vector collected from sketch figure picture.And utilize symbol away from
The method left the theatre with gradient vector field, can obtain more smooth and stable interpolation vector, and the effect approached more stablizes one
A bit.
Method two, sketch figure picture is converted into the vector graphics of two dimension, it is then specific to being incident upon the user of two dimensional surface
The surface vertices of bodily form model carry out approximation process.Its principle is more similar to the first scheme, and difference is minimum distance
With the method for solving of nearest direction vector, and after beeline and direction is solved, directly can be entered using these data
Row approaches deformation, it is no longer necessary to grid data structure.This method is asked in the hope of the method for solution vector and the joining of vector graphics
Solution, its solution procedure are all based on existing computer graphical conventional algorithm.Herein, vector graphics is being created according to sketch figure picture
When, i.e., pixel data vector quantization during, carry out the conversion of bulk, make vector graphics corresponding to the outline that newly creates
It is consistent with the size of manikin.
Finally, texture replacement algorithm can be utilized, sets superficial muscle to deform to increase mould on the specific bodily form model of user
The authenticity and details of type, herein, the displacement median for replacing material can be arranged to 1, i.e. the numerical value of textures texture is less than 1
Part corresponding to the surface of model can be recessed along normal to a surface.
Optionally, after the specific bodily form model of user is generated, user can be allowed to be adjusted the model, corresponding processing
Can be as follows：Projection process, generation two-dimensional projection model are carried out to the specific bodily form model of user；Two-dimensional projection's mould is sent to user
Type, so that user carries out error correction according to actual human body image to two-dimensional projection's model；Sent based on user revised
Two-dimensional projection's model, deformation process is carried out to the specific bodily form model of user.
In force, it is contemplated that the actual human body image that user uploads there may be error, and the user ultimately generated is specific
Manikin there may be part variation with the actual bodily form of user, and server can be by the specific bodily form model projection of the user of generation
To two dimensional surface, so as to obtain two-dimensional projection's model, two-dimensional projection's model herein is different from projected image, belongs to vector number
According to front two dimensional model and side two dimensional model can be divided into.Server can be by two-dimensional projection's model and real shooting photo afterwards
Superimposed complex data is sent to user, and user then can carry out error to two-dimensional projection's model according to actual human body image and repair
Positive regulator, so that the actual bodily form of two-dimensional projection's model and user itself more matches.Specifically, user can be installed in end
In end, two-dimensional projection's model is modified in the client that virtual fitting platform provides, further, grid can be used
Deformation and the modification mode of plastic deformation.
The processing of distortion of the mesh can be as follows：According to the summit sequence number before suture, the surface vertices of the model sutured
It is divided into several set of vertices, these groups and the model of each several part before suture correspond.Pay attention to, merely just a top in logic
Point packet, and model need not be physically split, it is that the model of different set of vertices calculates maximum boundary frame, as shown in figure 5, net
The density of lattice can need to adjust into Mobile state according to user, i.e., if what the adjustment of user's needs each time can have a great influence
Scope, grid can be with more sparse；On the contrary, if necessary to the adjustment of local small range, then relatively closeer grid is needed.
After user adjusted warp mesh, the deformation ratio according to grid enters row interpolation change to the model vertices covered contained by the warp mesh
Change.Generally only need linear interpolation, naturally it is also possible to other interpolation algorithms, such as curve interpolation, to reach more flat
Sliding deformation effect.Herein, distortion of the mesh is only suitable for applying to the related model deformation processing of above-mentioned nurbs curve.
The processing of plastic deformation can be as follows：User can first select to need the target surface summit changedAnd set
Corresponding radius of influence r and deformation attenuation coefficient rs, can click on simultaneously drag target surface vertices, the application of client terminal afterwards
Program can then calculate target surface top displacement vectorThen other tables of the model in the above-mentioned radius of influence are traveled through
Vertex of surface, other surface vertices are determined according to the distance D and attenuation coefficient on other surface vertices and target surface summit
Motion vector, so as to the original position coordinate according to other surface verticesIt is determined that the position coordinates after draggingI.e.Wherein rs=D/r, DecayFn (rs) are the default decay letter for being as independent variable using rs
Number, refers to Fig. 6.The modification method is suitable for bone and nurbs curve both deformations simultaneously.
Above-mentioned distortion of the mesh and plastic deformation process can be client and the interaction of server, and server, which receives, to be used
The related setting that family inputs in client-side, and corresponding regulation is carried out according to correlation setting, interaction is more conventional, herein not
Do more detailed description.User's modification finishes, and two-dimensional projection's model after regulation can be sent into server, server can
So that based on revised two-dimensional projection's model, finishing deformation process is carried out to the specific bodily form model of user.
It is true in the bodily form Sample Storehouse prestored according to the bodily form characteristic information of user's input in the embodiment of the present invention
Determine basic manikin corresponding to bodily form characteristic information, the actual human body image uploaded to user carries out calculating processing generation outline
Image, and human bodily form's data of user are obtained from sketch figure picture, according to human bodily form's data, basic manikin is carried out
Deformation adjustment, generates the specific bodily form model of user.So, server can be real according to the actual human body image pair of user and user
Basic manikin that the border bodily form more matches carries out deformation process, so as to generation the specific bodily form model of user and user it is true
The matching degree of entity shape is higher.
Based on identical technical concept, the embodiment of the present invention additionally provides a kind of dress for generating the specific bodily form model of user
Put, as shown in fig. 7, the device includes：
Determining module 701, it is true in the bodily form Sample Storehouse prestored for the bodily form characteristic information inputted according to user
Basic manikin corresponding to the fixed bodily form characteristic information；
Acquisition module 702, the actual human body image for being uploaded to the user carry out calculating processing generation sketch figure picture,
And human bodily form's data of the user are obtained from the sketch figure picture；
Generation module 703, for according to human bodily form's data, deformation adjustment to be carried out to the basic manikin,
Generate the specific bodily form model of user.
Optionally, the actual human body image includes human body direct picture and human body side image；
The acquisition module 702, is used for：
Preset number user people's body section, the user are chosen along human body natural's direction of extension to the human body of the user
People's body section and human body natural's direction of extension are perpendicular；
The human body direct picture and the human body side image uploaded to the user carries out calculating processing generation just
Face sketch figure picture and side sketch figure picture；
Determine user people's body section corresponding transversal in the positive sketch figure picture and the side sketch figure picture
Section, determine that human body width data and human body of the user at user people's body section are deep according to the length of the transversal section
Degrees of data.
Optionally, the generation module 703, is used for：
Preset number model is chosen in predeterminated position along human body natural's direction of extension to the basic manikin
Section, the model width data and model depth data in the preset number model section are obtained, wherein, user's human body
Section and human body natural's direction of extension are perpendicular；
Using the center in each model section as node, model is set to bind bone, according to default covering standard pair
The model binding bone carries out covering processing, sets corresponding deformation weight；
According to the model width data and model depth data in the model section, and the user is in corresponding user
Human body width data and human depth's data at people's body section, determine zoom factor；
Model binding bone is zoomed in and out according to the zoom factor, and according to the position of user people's body section
The position for adjusting the model binding bone is put, generates the specific bodily form model of user.
Optionally, the generation module 703, is used for：
Preset number model is chosen in predeterminated position along human body natural's direction of extension to the basic manikin
Section, the model width data and model depth data in the preset number model section are obtained, wherein, user's human body
Section and human body natural's direction of extension are perpendicular；
Using the center in each model section as control point, model nurbs curve is created, by the basic manikin
Surface vertices project on the model nurbs curve, it is determined that each surface vertices correspond to the relative vector of subpoint
With the subpoint corresponding to nurbs curve parameter value；
According to the model width data and model depth data in the model section, and the user is in corresponding user
Human body width data and human depth's data at people's body section, determine zoom factor；
Using the center of each user people's body section as control point, user's human body nurbs curve is created；
The relative vector of subpoint is corresponded to according to each surface vertices, the zoom factor and the subpoint are corresponding
Nurbs curve parameter value, and user's human body nurbs curve, generate the specific bodily form model of user.
Optionally, described device also includes：
Deformation module, for according to human bodily form's data, carrying out deformation adjustment to the basic manikin, life
Into after the specific bodily form model of user, the specific bodily form model of the user is projected to two dimensional surface generation two-dimensional projection model,
Establishment etc. obtains the beeline of each grid node and the border of the sketch figure picture and the beeline than two-dimensional grid
Corresponding gradient vector, according to gradient vector corresponding to the beeline and the beeline, in two-dimensional projection's mould
Type carries out deformation process, and based on two-dimensional projection's model after deformation process, the specific bodily form model of the user is carried out at deformation
Reason.
Optionally, described device also includes：
Correcting module, for, according to human bodily form's data, becoming described to the first foundation manikin
Shape is adjusted, and after generating the specific bodily form model of user, projection process is carried out to the specific bodily form model of the user, two dimension is generated and throws
Shadow model, two-dimensional projection's model is sent to the user so that the user according to the actual human body image to described
Two-dimensional projection's model carries out error correction, the revised two-dimensional projection's model sent based on user, to user's particular volume
Shape model carries out deformation process.
It is true in the bodily form Sample Storehouse prestored according to the bodily form characteristic information of user's input in the embodiment of the present invention
Determine basic manikin corresponding to bodily form characteristic information, the actual human body image uploaded to user carries out calculating processing generation outline
Image, and human bodily form's data of user are obtained from sketch figure picture, according to human bodily form's data, basic manikin is carried out
Deformation adjustment, generates the specific bodily form model of user.So, server can be real according to the actual human body image pair of user and user
Basic manikin that the border bodily form more matches carries out deformation process, so as to generation the specific bodily form model of user and user it is true
The matching degree of entity shape is higher.
It should be noted that：The device for the generation specific bodily form model of user that above-described embodiment provides is specific in generation user
, can as needed will be upper only with the division progress of above-mentioned each functional module for example, in practical application during bodily form model
State function distribution to be completed by different functional modules, i.e., the internal structure of device is divided into different functional modules, to complete
All or part of function described above.In addition, above-described embodiment provide generation the specific bodily form model of user device with
The embodiment of the method for the generation specific bodily form model of user belongs to same design, and its specific implementation process refers to embodiment of the method, this
In repeat no more.
The embodiment of the present invention additionally provides a kind of server, and Fig. 8 is that the structure of server provided in an embodiment of the present invention is shown
It is intended to.The server 800 can produce bigger difference because configuration or performance are different, can include in one or more
Central processor (central processing units, CPU) 822 (for example, one or more processors) and memory
832, one or more storage application programs 842 or data 844 storage medium 830 (such as one or more sea
Measure storage device).Wherein, memory 832 and storage medium 830 can be of short duration storage or persistently storage.Storage is stored in be situated between
The program of matter 830 can include one or more modules (diagram does not mark), and each module can include in server
Series of instructions operation.Further, central processing unit 822 could be arranged to communicate with storage medium 830, in server
The series of instructions operation in storage medium 830 is performed on 800.
Server 800 can also include one or more power supplys 826, one or more wired or wireless networks
Interface 850, one or more input/output interfaces 858, one or more keyboards 856, and/or, one or one
Above operating system 841, such as Windows ServerTM, Mac OS XTM, UnixTM, LinuxTM, FreeBSDTM etc..
Server 800 can include memory, and one or more than one program, one of them or one
Procedure above is stored in memory, and be configured to by one either more than one computing device it is one or one
Procedure above includes the instruction for being used for being operated below：
The bodily form characteristic information inputted according to user, the bodily form feature letter is determined in the bodily form Sample Storehouse prestored
Basic manikin corresponding to breath；
The actual human body image uploaded to the user carries out calculating processing generation sketch figure picture, and from the sketch figure picture
The middle human bodily form's data for obtaining the user；
According to human bodily form's data, deformation adjustment is carried out to the basic manikin, generates the specific bodily form of user
Model.
Optionally, the actual human body image includes human body direct picture and human body side image；
The actual human body image uploaded to the user carries out calculating processing generation sketch figure picture, and from the outline
Human bodily form's data of the user are obtained in image, including：
Preset number user people's body section, the user are chosen along human body natural's direction of extension to the human body of the user
People's body section and human body natural's direction of extension are perpendicular；
The human body direct picture and the human body side image uploaded to the user carries out calculating processing generation just
Face sketch figure picture and side sketch figure picture；
Determine user people's body section corresponding transversal in the positive sketch figure picture and the side sketch figure picture
Section, determine that human body width data and human body of the user at user people's body section are deep according to the length of the transversal section
Degrees of data.
Optionally, it is described according to human bodily form's data, deformation adjustment is carried out to the basic manikin, generation is used
The specific bodily form model in family, including：
Preset number model is chosen in predeterminated position along human body natural's direction of extension to the basic manikin
Section, the model width data and model depth data in the preset number model section are obtained, wherein, user's human body
Section and human body natural's direction of extension are perpendicular；
Using the center in each model section as node, model is set to bind bone, according to default covering standard pair
The model binding bone carries out covering processing, sets corresponding deformation weight；
According to the model width data and model depth data in the model section, and the user is in corresponding user
Human body width data and human depth's data at people's body section, determine zoom factor；
Model binding bone is zoomed in and out according to the zoom factor, and according to the position of user people's body section
The position for adjusting the model binding bone is put, generates the specific bodily form model of user.
Optionally, it is described according to human bodily form's data, deformation adjustment is carried out to the basic manikin, generation is used
The specific bodily form model in family, including：
Preset number model is chosen in predeterminated position along human body natural's direction of extension to the basic manikin
Section, the model width data and model depth data in the preset number model section are obtained, wherein, user's human body
Section and human body natural's direction of extension are perpendicular；
Using the center in each model section as control point, model nurbs curve is created, by the basic manikin
Surface vertices project on the model nurbs curve, it is determined that each surface vertices correspond to the relative vector of subpoint
With the subpoint corresponding to nurbs curve parameter value；
According to the model width data and model depth data in the model section, and the user is in corresponding user
Human body width data and human depth's data at people's body section, determine zoom factor；
Using the center of each user people's body section as control point, user's human body nurbs curve is created；
The relative vector of subpoint is corresponded to according to each surface vertices, the zoom factor and the subpoint are corresponding
Nurbs curve parameter value, and user's human body nurbs curve, generate the specific bodily form model of user.
Optionally, deformation adjustment is carried out to the basic manikin, generated according to human bodily form's data described
After the specific bodily form model of user, methods described also includes：
The specific bodily form model of the user is projected to two dimensional surface generation two-dimensional projection model, two-dimensional mesh is compared in establishment etc.
Lattice；
Beeline and the beeline for obtaining each grid node and the border of the sketch figure picture are corresponding terraced
Degree vector；
According to gradient vector corresponding to the beeline and the beeline, become in two-dimensional projection's model
Shape processing；
Based on two-dimensional projection's model after deformation process, deformation process is carried out to the specific bodily form model of the user.
Optionally, deformation adjustment is carried out to the basic manikin, generated according to human bodily form's data described
After the specific bodily form model of user, methods described also includes：
Projection process, generation two-dimensional projection model are carried out to the specific bodily form model of the user；
Send two-dimensional projection's model to the user so that the user according to the actual human body image to described
Two-dimensional projection's model carries out error correction；
The revised two-dimensional projection's model sent based on user, is carried out at deformation to the specific bodily form model of the user
Reason.
It is true in the bodily form Sample Storehouse prestored according to the bodily form characteristic information of user's input in the embodiment of the present invention
Determine basic manikin corresponding to bodily form characteristic information, the actual human body image uploaded to user carries out calculating processing generation outline
Image, and human bodily form's data of user are obtained from sketch figure picture, according to human bodily form's data, basic manikin is carried out
Deformation adjustment, generates the specific bodily form model of user.So, server can be real according to the actual human body image pair of user and user
Basic manikin that the border bodily form more matches carries out deformation process, so as to generation the specific bodily form model of user and user it is true
The matching degree of entity shape is higher.
One of ordinary skill in the art will appreciate that hardware can be passed through by realizing all or part of step of above-described embodiment
To complete, by program the hardware of correlation can also be instructed to complete, related program can be stored in a kind of computer can
Read in storage medium, storage medium mentioned above can be read-only storage, disk or CD etc..
Presently preferred embodiments of the present invention is these are only, is not intended to limit the invention, it is all in the spirit and principles in the present invention
Within, any modification, equivalent substitution and improvements made etc., it should be included in the scope of the protection.
Claims (12)
- A kind of 1. method for generating the specific bodily form model of user, it is characterised in that methods described includes：The bodily form characteristic information inputted according to user, determines the bodily form characteristic information pair in the bodily form Sample Storehouse prestored The basic manikin answered；The actual human body image uploaded to the user carries out calculating processing generation sketch figure picture, and is obtained from the sketch figure picture Take human bodily form's data of the user；According to human bodily form's data, deformation adjustment is carried out to the basic manikin, generates the specific bodily form model of user.
- 2. according to the method for claim 1, it is characterised in that the actual human body image includes human body direct picture and people Body side surface image；The actual human body image uploaded to the user carries out calculating processing generation sketch figure picture, and from the sketch figure picture The middle human bodily form's data for obtaining the user, including：Preset number user people's body section, user's human body are chosen along human body natural's direction of extension to the human body of the user Section and human body natural's direction of extension are perpendicular；The human body direct picture and the human body side image uploaded to the user carries out calculating processing generation front and cut Shadow image and side sketch figure picture；Determine user people's body section corresponding transversal section, root in the positive sketch figure picture and the side sketch figure picture Human body width data and human depth number of the user at user people's body section are determined according to the length of the transversal section According to.
- 3. according to the method for claim 2, it is characterised in that it is described according to human bodily form's data, to the basis Manikin carries out deformation adjustment, generates the specific bodily form model of user, including：Preset number model section is chosen in predeterminated position along human body natural's direction of extension to the basic manikin, The model width data and model depth data in the preset number model section are obtained, wherein, user people's body section It is perpendicular with human body natural's direction of extension；Using the center in each model section as node, model is set to bind bone, according to default covering standard to described Model binding bone carries out covering processing, sets corresponding deformation weight；According to the model width data and model depth data in the model section, and the user is in corresponding user's human body Human body width data and human depth's data at section, determine zoom factor；Model binding bone is zoomed in and out according to the zoom factor, and adjusted according to the position of user people's body section The position of the whole model binding bone, generates the specific bodily form model of user.
- 4. according to the method for claim 2, it is characterised in that it is described according to human bodily form's data, to the basis Manikin carries out deformation adjustment, generates the specific bodily form model of user, including：Preset number model section is chosen in predeterminated position along human body natural's direction of extension to the basic manikin, The model width data and model depth data in the preset number model section are obtained, wherein, user people's body section It is perpendicular with human body natural's direction of extension；Using the center in each model section as control point, model nurbs curve is created, by the table of the basic manikin Vertex of surface is projected on the model nurbs curve, it is determined that each surface vertices correspond to relative vector and the institute of subpoint State nurbs curve parameter value corresponding to subpoint；According to the model width data and model depth data in the model section, and the user is in corresponding user's human body Human body width data and human depth's data at section, determine zoom factor；Using the center of each user people's body section as control point, user's human body nurbs curve is created；The relative vector of subpoint is corresponded to according to each surface vertices, corresponding to the zoom factor and the subpoint Nurbs curve parameter value, and user's human body nurbs curve, generate the specific bodily form model of user.
- 5. according to the method for claim 1, it is characterised in that described according to human bodily form's data, to the base Plinth manikin carries out deformation adjustment, and after generating the specific bodily form model of user, methods described also includes：The specific bodily form model of the user is projected to two dimensional surface generation two-dimensional projection model, two-dimensional grid is compared in establishment etc.；Obtain each grid node and the border of the sketch figure picture beeline and the corresponding gradient of the beeline to Amount；According to gradient vector corresponding to the beeline and the beeline, carried out in two-dimensional projection's model at deformation Reason；Based on two-dimensional projection's model after deformation process, deformation process is carried out to the specific bodily form model of the user.
- 6. according to the method for claim 1, it is characterised in that described according to human bodily form's data, to the base Plinth manikin carries out deformation adjustment, and after generating the specific bodily form model of user, methods described also includes：Projection process, generation two-dimensional projection model are carried out to the specific bodily form model of the user；Send two-dimensional projection's model to the user so that the user according to the actual human body image to the two dimension Projection model carries out error correction；The revised two-dimensional projection's model sent based on user, deformation process is carried out to the specific bodily form model of the user.
- 7. a kind of device for generating the specific bodily form model of user, it is characterised in that described device includes：Determining module, for the bodily form characteristic information inputted according to user, determined in the bodily form Sample Storehouse prestored described in Basic manikin corresponding to bodily form characteristic information；Acquisition module, the actual human body image for being uploaded to the user carry out calculating processing generation sketch figure picture, and from institute State human bodily form's data that the user is obtained in sketch figure picture；Generation module, for according to human bodily form's data, deformation adjustment being carried out to the basic manikin, generates user Specific bodily form model.
- 8. device according to claim 7, it is characterised in that the actual human body image includes human body direct picture and people Body side surface image；The acquisition module, is used for：Preset number user people's body section, user's human body are chosen along human body natural's direction of extension to the human body of the user Section and human body natural's direction of extension are perpendicular；The human body direct picture and the human body side image uploaded to the user carries out calculating processing generation front and cut Shadow image and side sketch figure picture；Determine user people's body section corresponding transversal section, root in the positive sketch figure picture and the side sketch figure picture Human body width data and human depth number of the user at user people's body section are determined according to the length of the transversal section According to.
- 9. device according to claim 8, it is characterised in that the generation module, be used for：Preset number model section is chosen in predeterminated position along human body natural's direction of extension to the basic manikin, The model width data and model depth data in the preset number model section are obtained, wherein, user people's body section It is perpendicular with human body natural's direction of extension；Using the center in each model section as node, model is set to bind bone, according to default covering standard to described Model binding bone carries out covering processing, sets corresponding deformation weight；According to the model width data and model depth data in the model section, and the user is in corresponding user's human body Human body width data and human depth's data at section, determine zoom factor；Model binding bone is zoomed in and out according to the zoom factor, and adjusted according to the position of user people's body section The position of the whole model binding bone, generates the specific bodily form model of user.
- 10. device according to claim 8, it is characterised in that the generation module, be used for：Preset number model section is chosen in predeterminated position along human body natural's direction of extension to the basic manikin, The model width data and model depth data in the preset number model section are obtained, wherein, user people's body section It is perpendicular with human body natural's direction of extension；Using the center in each model section as control point, model nurbs curve is created, by the table of the basic manikin Vertex of surface is projected on the model nurbs curve, it is determined that each surface vertices correspond to relative vector and the institute of subpoint State nurbs curve parameter value corresponding to subpoint；According to the model width data and model depth data in the model section, and the user is in corresponding user's human body Human body width data and human depth's data at section, determine zoom factor；Using the center of each user people's body section as control point, user's human body nurbs curve is created；The relative vector of subpoint is corresponded to according to each surface vertices, corresponding to the zoom factor and the subpoint Nurbs curve parameter value, and user's human body nurbs curve, generate the specific bodily form model of user.
- 11. according to the method for claim 7, it is characterised in that described device also includes：Deformation module, for according to human bodily form's data, deformation adjustment being carried out to the basic manikin, generation is used After the specific bodily form model in family, the specific bodily form model of the user is projected to two dimensional surface generation two-dimensional projection model, created Deng than two-dimensional grid, beeline and the beeline for obtaining each grid node and the border of the sketch figure picture are corresponding Gradient vector, according to gradient vector corresponding to the beeline and the beeline, enter in two-dimensional projection's model Row deformation process, based on two-dimensional projection's model after deformation process, deformation process is carried out to the specific bodily form model of the user.
- 12. device according to claim 7, it is characterised in that described device also includes：Correcting module, for according to human bodily form's data, carrying out deformation adjustment to the first foundation manikin, life Into after the specific bodily form model of user, projection process is carried out to the specific bodily form model of the user, generates two-dimensional projection's model, to The user sends two-dimensional projection's model so that the user according to the actual human body image to two-dimensional projection's mould Type carries out error correction, the revised two-dimensional projection's model sent based on user, the specific bodily form model of the user is carried out Deformation process.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610534953.5A CN107590708B (en) | 2016-07-07 | 2016-07-07 | Method and device for generating user specific body shape model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610534953.5A CN107590708B (en) | 2016-07-07 | 2016-07-07 | Method and device for generating user specific body shape model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN107590708A true CN107590708A (en) | 2018-01-16 |
| CN107590708B CN107590708B (en) | 2021-01-01 |
Family
ID=61046009
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610534953.5A Active CN107590708B (en) | 2016-07-07 | 2016-07-07 | Method and device for generating user specific body shape model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107590708B (en) |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109697748A (en) * | 2018-12-19 | 2019-04-30 | 网易（杭州）网络有限公司 | Model compression processing method, model pinup picture processing method device, storage medium |
| CN109770943A (en) * | 2019-01-28 | 2019-05-21 | 电子科技大学 | A kind of ultrasonic automatic optimization method positioned using computer vision |
| CN109829794A (en) * | 2019-02-28 | 2019-05-31 | 深圳市翰香文化科技传媒有限公司 | A kind of virtual fit method and virtual fitting system based on mobile terminal |
| CN109993689A (en) * | 2019-03-14 | 2019-07-09 | 珠海天燕科技有限公司 | A kind of makeups method and apparatus |
| WO2020019915A1 (en) * | 2018-07-25 | 2020-01-30 | 北京市商汤科技开发有限公司 | Image processing method and apparatus, and computer storage medium |
| CN110910481A (en) * | 2019-11-26 | 2020-03-24 | 浙江大学 | Three-dimensional garment mannequin modeling method based on contour deformation |
| CN111861822A (en) * | 2020-06-03 | 2020-10-30 | 四川大学华西医院 | Patient model construction method, equipment and medical education system |
| CN113837835A (en) * | 2021-09-18 | 2021-12-24 | 完美世界(北京)软件科技发展有限公司 | Role model-based wearable product display method and device |
| WO2023134709A1 (en) * | 2022-01-17 | 2023-07-20 | 阿里巴巴（中国）有限公司 | Virtual garment try-on method and apparatus |
| CN117297593A (en) * | 2023-10-30 | 2023-12-29 | 济宁医学院附属医院 | Action detection system and detection method for ankylosing spondylitis patient |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101017575A (en) * | 2007-02-15 | 2007-08-15 | 东华大学 | Method for automatically forming 3D virtual human body based on human component template and body profile |
| CN101515374A (en) * | 2008-02-20 | 2009-08-26 | 中国科学院自动化研究所 | Individualized realistic virtual character modeling method based on images |
| US20100156935A1 (en) * | 2008-12-22 | 2010-06-24 | Electronics And Telecommunications Research Institute | Method and apparatus for deforming shape of three dimensional human body model |
| CN103810750A (en) * | 2014-01-16 | 2014-05-21 | 北京航空航天大学 | Human body section ring based parametric deformation method |
-
2016
- 2016-07-07 CN CN201610534953.5A patent/CN107590708B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101017575A (en) * | 2007-02-15 | 2007-08-15 | 东华大学 | Method for automatically forming 3D virtual human body based on human component template and body profile |
| CN101515374A (en) * | 2008-02-20 | 2009-08-26 | 中国科学院自动化研究所 | Individualized realistic virtual character modeling method based on images |
| US20100156935A1 (en) * | 2008-12-22 | 2010-06-24 | Electronics And Telecommunications Research Institute | Method and apparatus for deforming shape of three dimensional human body model |
| CN103810750A (en) * | 2014-01-16 | 2014-05-21 | 北京航空航天大学 | Human body section ring based parametric deformation method |
Cited By (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2020019915A1 (en) * | 2018-07-25 | 2020-01-30 | 北京市商汤科技开发有限公司 | Image processing method and apparatus, and computer storage medium |
| CN109697748A (en) * | 2018-12-19 | 2019-04-30 | 网易（杭州）网络有限公司 | Model compression processing method, model pinup picture processing method device, storage medium |
| CN109697748B (en) * | 2018-12-19 | 2023-05-16 | 网易（杭州）网络有限公司 | Model compression processing method, model mapping processing method, model compression processing device, and storage medium |
| CN109770943B (en) * | 2019-01-28 | 2021-11-02 | 电子科技大学 | Ultrasonic automatic optimization method using computer vision positioning |
| CN109770943A (en) * | 2019-01-28 | 2019-05-21 | 电子科技大学 | A kind of ultrasonic automatic optimization method positioned using computer vision |
| CN109829794A (en) * | 2019-02-28 | 2019-05-31 | 深圳市翰香文化科技传媒有限公司 | A kind of virtual fit method and virtual fitting system based on mobile terminal |
| CN109993689A (en) * | 2019-03-14 | 2019-07-09 | 珠海天燕科技有限公司 | A kind of makeups method and apparatus |
| CN110910481A (en) * | 2019-11-26 | 2020-03-24 | 浙江大学 | Three-dimensional garment mannequin modeling method based on contour deformation |
| CN111861822A (en) * | 2020-06-03 | 2020-10-30 | 四川大学华西医院 | Patient model construction method, equipment and medical education system |
| CN111861822B (en) * | 2020-06-03 | 2023-11-21 | 四川大学华西医院 | Patient model construction method, equipment and medical education system |
| CN113837835A (en) * | 2021-09-18 | 2021-12-24 | 完美世界(北京)软件科技发展有限公司 | Role model-based wearable product display method and device |
| WO2023134709A1 (en) * | 2022-01-17 | 2023-07-20 | 阿里巴巴（中国）有限公司 | Virtual garment try-on method and apparatus |
| CN117297593A (en) * | 2023-10-30 | 2023-12-29 | 济宁医学院附属医院 | Action detection system and detection method for ankylosing spondylitis patient |
| CN117297593B (en) * | 2023-10-30 | 2024-04-05 | 济宁医学院附属医院 | Action detection system and detection method for ankylosing spondylitis patient |
Also Published As
| Publication number | Publication date |
|---|---|
| CN107590708B (en) | 2021-01-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107590708A (en) | A kind of method and apparatus for generating the specific bodily form model of user | |
| CN110443885B (en) | Three-dimensional human head and face model reconstruction method based on random human face image | |
| CN105006014B (en) | The realization method and system that virtual clothing Fast simulation is tried on | |
| Igarashi et al. | Clothing manipulation | |
| CN109427007B (en) | Virtual fitting method based on multiple visual angles | |
| CN114119908A (en) | Clothing model driving method, equipment and storage medium | |
| Yunchu et al. | Prototype garment pattern flattening based on individual 3D virtual dummy | |
| Zhu et al. | An efficient human model customization method based on orthogonal-view monocular photos | |
| Wuhrer et al. | Estimating 3D human shapes from measurements | |
| Gültepe et al. | Real-time virtual fitting with body measurement and motion smoothing | |
| CN104021589A (en) | Three-dimensional fitting simulating method | |
| Hu et al. | Personalized 3D mannequin reconstruction based on 3D scanning | |
| CN114119905A (en) | Virtual fitting method, system, equipment and storage medium | |
| CN114119906A (en) | Self-adaptive driving method and device for clothing model and storage medium | |
| CN114119910A (en) | Method, equipment and storage medium for matching clothing model with human body model | |
| CN108564619A (en) | A kind of sense of reality three-dimensional facial reconstruction method based on two photos | |
| CN112233222A (en) | Human body parametric three-dimensional model deformation method based on neural network joint point estimation | |
| CN109655011A (en) | A kind of method and system of Human Modeling dimension measurement | |
| Cetinaslan et al. | Sketching manipulators for localized blendshape editing | |
| Liu et al. | Facial animation by optimized blendshapes from motion capture data | |
| Li et al. | Computer-aided 3D human modeling for portrait-based product development using point-and curve-based deformation | |
| Fondevilla et al. | Fashion transfer: Dressing 3d characters from stylized fashion sketches | |
| Yan et al. | Skeleton-based shape deformation using simplex transformations | |
| Cutler et al. | An art-directed wrinkle system for CG character clothing and skin | |
| CN116580148A (en) | Parameterized human body model fitting method based on human body point cloud and measurement parameters |
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 |