CN103455501A - Clothing database generating method, clothing model building method and fitting method - Google Patents

Abstract

The invention relates to a clothing database generating method, a clothing model building method and a fitting method and system. The clothing database generating method includes the steps: acquiring clothing front photos and clothing size data; removing backgrounds of the clothing front photos; setting clothing key points of the clothing front photos; setting size data of the clothing front photos according to the clothing size data; storing the clothing front photos, the clothing key points and the size data into a clothing database. The resource utilization rate of existing clothing plane photos and clothing size data can be sufficiently increased, the clothing database for building a clothing model is directly generated, manpower and time consumption for generating the clothing database is reduced, building and operating costs of the online fitting system are greatly reduced, and commercial viability of the online fitting system is improved.

Description

Garment data generation method, dress form method for building up and fitting method

Technical field

The present invention relates to a kind of garment data generation method, set up the method for dress form and the method and system of fitting.

Background technology

At present, online dressing system usually need to be set up a large amount of dress forms and try on the net to offer the user, existing online dressing system has two kinds usually, a kind of online dressing system that is based on the 2D technology, and another kind is based on the online dressing system of 3D technology.Online dressing system based on the 2D technology generally need to be through every clothes the model who sets posture and take pictures with it, and deposits the clothes storehouse in after picture is processed; And the online dressing system based on the 3D technology need to be set up complicated 3D model to every clothes.Thereby, no matter be based on the online dressing system of 2D technology, the online dressing system that also is based on the 3D technology all can relate to the work that a large amount of manpowers, time loss etc. need to be paid sky high cost, therefore, the commercial viability of existing online dressing system is not high, needs badly at present and seeks a kind of method of making dress form cost that reduces to improve the commerciality of online dressing system.

Summary of the invention

The object of the present invention is to provide a kind of garment data generation method, set up the method for dress form and the method and system of fitting, can fully improve the resource utilization of existing dress planar photo and clothes dimensional data, directly generate for setting up the garment data of dress form, reduce and generate garment data manpower and construction and the operation cost of time loss significantly to reduce online dressing system, improve the commercial viability of online dressing system.

For addressing the above problem, the invention provides a kind of garment data generation method, comprising:

Gather clothes full face and clothes dimensional data;

Remove the background of described clothes full face;

The clothes key point of described clothes full face is set;

The dimensional data of described clothes full face is set according to described garment dimension data; And

Described clothes full face and clothes key point thereof and dimensional data are stored in garment data.

Further, in said method, gather from network described clothes full face in existing clothes photo.

Further, in said method, the resolution of described clothes full face is more than or equal to the 512*512 pixel.

Further, in said method, the background that adopts the picture processing instrument to remove described clothes full face.

Further, in said method, described picture software handling implement is the stingy figure instrument in Photo Shop.

Further, in said method, described clothes key point comprises a kind of or combination in any in the position data that characterizes collar, shoulder, sleeve, the bottom, waistband, front wave, trouser legs or leg opening.

Further, in said method, described clothes key point means in the mode of point.

Further, in said method, described clothes key point means in the mode of Points And lines.

Further, in said method, the dimensional data of described clothes full face comprises that clothing length, shoulder breadth, collar, chest measurement, waistline, hip circumference, sleeve length, sleeve enclose, trousers length, front wave, waistline, hip circumference, thigh circumference, calf circumference or the leg opening a kind of or combination in any in enclosing.

The present invention also provides a kind of and adopts above-mentioned garment data generation method to set up the method for dress form, comprising:

Set up according to the clothes full face obtained and clothes key point thereof and dimensional data the dress planar grid group formed by a plurality of the second grids on isoplanar from garment data;

Generate the second normal of the steric information that means each the second grid according to described clothes photo; And

The clothes grid model that generation is generated by described the second normal and clothes plane grid group.

Further, in said method, after generating the step of the clothes grid model generated by described the second normal and clothes plane grid group, also comprise according to described dress planar grid group and described the second normal and render the clothes solid model.

Further, in said method, according to described dress planar grid group and described the second normal, render in the step of clothes solid model, along the axle perpendicular to described dress planar grid group, render described clothes solid model.

Further, in said method, described the second grid is triangle.

Further, in said method, each summit of each the second grid is corresponding one by one with each pinup picture coordinate of described clothes full face.

The present invention also provides a kind of method that adopts the above-mentioned method of setting up dress form to be fitted, and comprising:

Generate the body area network lattice model according to human body photo and build data;

Described body area network lattice model is combined and aligns to generate the fitting grid model with the clothes grid model; And

Render the fitting solid model according to described fitting grid model.

Further, in said method, according to the step of human body photo and build data generation body area network lattice model, comprise:

Generate according to human body photo and build data the human body plane grid group formed by a plurality of human joint pointses and a plurality of the first grids at grade;

Generate the first normal of the steric information that means each the first grid according to described human body photo.

Further, in said method, described the first grid is triangle.

Further, in said method, generate in the step of the human body plane grid group formed by a plurality of human joint pointses and a plurality of the first grids at grade according to human body photo and build data, in the place of described human joint points, described the first grid is carried out to radial distribution centered by described human joint points.

Further, in said method, generate in the step of the human body plane grid group formed by a plurality of human joint pointses and a plurality of the first grids at grade according to human body photo and build data, described the first grid is rotated along the axle perpendicular to described human body plane grid group around described human joint points.

Further, in said method, each summit of each the first grid is corresponding one by one with each pinup picture coordinate of described human body photo.

Further, in said method, set up the step of the dress planar grid group formed by a plurality of the second grids on isoplanar according to the clothes full face obtained and clothes key point thereof and dimensional data from garment data after, also comprise according to the body area network lattice model and adjust described clothes grid model.

Further, in said method, described body area network lattice model and clothes grid model are combined and align to generate in the step of fitting grid model, by described human joint points and clothes key point, described body area network lattice model is combined and aligns with described clothes grid model.

Further, in said method, render according to described fitting grid model in the step of fitting solid model, by described human body plane grid group, the first normal, dress planar grid group and the second normal edge, perpendicular to the axle of described fitting grid model, play up described fitting solid model.

According to another side of the present invention, a kind of garment data generation system is provided, comprising:

Acquisition module, for gathering the clothes full face;

Remove the background module, for removing the background of described clothes full face;

The key point module, for arranging the clothes key point of described clothes full face;

Size module, for arranging the dimensional data of described clothes full face;

The garment data module, for storing described clothes full face and clothes key point and size.

Further, in said system, state acquisition module and gather described clothes full face in existing clothes photo from network.

The present invention also provides a kind of system of setting up dress form, comprising:

Above-mentioned garment data generation system; Clothes grid group module, the clothes full face obtained from garment data for basis and clothes key point thereof and dimensional data are set up the dress planar grid group be comprised of a plurality of the second grids isoplanar;

The second method wire module, for generating the second normal of the steric information that means each the second grid according to described clothes photo; And

Clothes grid model module, for generating the clothes grid model generated by described the second normal and clothes plane grid group.

Further, in said system, also comprise that a rendering module is for rendering the clothes solid model according to described dress planar grid group and described the second normal.

Further, in said system, described rendering module, render described clothes solid model for the axle along perpendicular to described dress planar grid group.

The present invention also provides a kind of dressing system, it is characterized in that, comprising:

The above-mentioned system of setting up dress form;

Body area network lattice model module, for generating the body area network lattice model according to human body photo and build data;

Fitting grid model module, for combining described body area network lattice model to align to generate the fitting grid model with the clothes grid model; And

Fitting solid model module, for rendering the fitting solid model according to described fitting grid model.

Further, in said system, described body area network lattice model module comprises:

Human body grid pack module, for generating according to human body photo and build data the human body plane grid group be comprised of a plurality of human joint pointses and a plurality of the first grids at grade;

The first method wire module, for generating the first normal of the steric information that means each the first grid according to described human body photo.

Further, in said system, described human body grid pack module, for carrying out radial distribution by described the first grid outside described human joint points centered by described human joint points.

Compared with prior art, the present invention is by gathering the clothes full face, remove the background of described clothes full face, then the clothes key point that described clothes full face is set arranges the dimensional data of described clothes full face according to described garment dimension data, and described clothes full face and clothes key point thereof and dimensional data are stored in garment data, can fully improve the resource utilization of existing dress planar photo and clothes dimensional data, directly generate for setting up the garment data of dress form, reduce and generate garment data manpower and construction and the operation cost of time loss significantly to reduce online dressing system, improve the commercial viability of online dressing system.

In addition, the present invention sets up according to the clothes full face obtained from garment data and clothes key point thereof and dimensional data the dress planar grid group be comprised of a plurality of the second grids on isoplanar, and generate the second normal of the steric information that means each the second grid according to described clothes photo, then generate the clothes grid model generated by described the second normal and clothes plane grid group, can generate fast, realistically the clothes solid model.

In addition, the present invention is by generating the body area network lattice model according to human body photo and build data, and described body area network lattice model is combined and aligns to generate the fitting grid model with the clothes grid model, then render the fitting solid model according to described fitting grid model, can quick generation true to nature and show user's clothes effect, whether the user can understand garment dimension intuitively suitable.

The accompanying drawing explanation

Fig. 1 is that the garment data of the embodiment of the present invention one generates method flow diagram;

Fig. 2 is the upper dress full face of the removal background of the embodiment of the present invention one;

Fig. 3 is the lower dress full face of the removal background of the embodiment of the present invention one;

Fig. 4 is the clothes key point schematic diagram that dress is set of the embodiment of the present invention one;

Fig. 5 is the clothes key point schematic diagram of dress under the setting of the embodiment of the present invention one;

Fig. 6 is the interface schematic diagram that the dress dimensional data is set of the embodiment of the present invention one;

Fig. 7 is the interface schematic diagram of dress dimensional data under the setting of the embodiment of the present invention one;

Fig. 8 is the method flow diagram of setting up dress form of the embodiment of the present invention two;

Fig. 9 is the fitting method process flow diagram of the embodiment of the present invention three;

Figure 10 is the schematic diagram of the human body plane grid group of the embodiment of the present invention three;

Figure 11 is the schematic diagram that the body area network lattice model of the embodiment of the present invention three aligns with the combination of clothes grid model;

Figure 12 is the clothes effect figure of the embodiment of the present invention three fitting solid models;

Figure 13 is the module diagram of the garment data generation system of the embodiment of the present invention four;

Figure 14 is the module diagram of the system of setting up dress form of the embodiment of the present invention five;

Figure 15 is the dressing system module diagram of the embodiment of the present invention six.

Embodiment

For above-mentioned purpose of the present invention, feature and advantage can be become apparent more, below in conjunction with the drawings and specific embodiments, the present invention is further detailed explanation.

Embodiment mono-

As shown in Figure 1, the invention provides a kind of garment data generation method, comprising:

Step S11, gather clothes full face and clothes dimensional data;

Wherein, because clothes electricity business generally can provide a large amount of existing clothes data resources that comprises clothes full face and corresponding garment dimension data on the webpage of garment marketing, although these clothes full faces are not taken according to the requirement of appointment, do not form corresponding 3D model data yet, can not directly in dressing system, use on the net, but by the present invention these online existing clothes data resources can be processed obtain can for online dressing system can the clothes data, for user's try-on garment on the net, specifically, but acquisition resolution is not less than the positive clear pictures of the online existing clothes of 512*512 pixel, so i.e. direct copying clothes data resource from network,

Step S12, remove the background of described clothes full face;

This step is called in iconology scratches figure, specifically can adopt the background of common picture processing software except described clothes full face, described picture processing software can be the stingy figure instrument in PhotoShop, the photo fairly simple for background can be removed by the magic wand tool in PhotoShop automatically as color background, photo for the background more complicated, it is often clean not that software is removed background automatically, can carry out manual removal by picture processing software, it shown in Fig. 2, is the upper dress full face of removing background, it shown in Fig. 3, is the lower dress full face of removing background,

Step S13, arrange the clothes key point of described clothes full face;

Preferably, described clothes key point can comprise a kind of or combination in any in the position data that characterizes collar, shoulder, sleeve, the bottom, waistband, front wave, trouser legs or leg opening; Described clothes key point can mean in the mode of point, certainly, in order to obtain more intuitive effect, described clothes key point also can mean in the mode of Points And lines, specifically, for upper dress, the method that the clothes key point of clothes full face is set is to click the position of collar, shoulder, sleeve or the bottom on the clothes photo with mouse, as shown in Figure 4, the clothes key point of upper dress shows in the mode of Points And lines; For trousers, the method that the clothes key point of clothes full face is set is to click and to comprise the positions such as waistband position, front wave, trouser legs or leg opening on the clothes photo with mouse, and as shown in Figure 5, the clothes key point of lower dress shows in the mode of Points And lines;

Step S14, arrange the dimensional data of described clothes full face according to described garment dimension data;

Wherein, the garment dimension data representation clothes there is the different size of how many kinds of, every kind of concrete numerical value of size is how many, accordingly, as shown in Figure 6, the dimensional data of the clothes full face of upper dress comprises clothing length, shoulder breadth, collar, chest measurement, waistline, hip circumference, sleeve length, the sleeve a kind of or combination in any in enclosing, and as shown in Figure 7, the dimensional data of the clothes full face of lower dress comprises trousers length, front wave, waistline, hip circumference, thigh circumference, calf circumference or the leg opening a kind of or combination in any in enclosing;

Step S15, be stored into described clothes full face and clothes key point thereof and dimensional data in garment data.

To sum up, the present embodiment can be applicable in online dressing system, and online dressing system needs fast, while setting up garment data at low cost, can generate by the method for the present embodiment.

Embodiment bis-

As shown in Figure 8, the present invention also provides a kind of method of setting up dress form, adopts the described garment data generation of embodiment mono-method to set up dress form, specifically comprises:

Step S21, gather clothes full face and clothes dimensional data;

As described in embodiment mono-, because clothes electricity business generally can provide a large amount of existing clothes data resources that comprises clothes full face and corresponding garment dimension data on the webpage of garment marketing, although these clothes full faces are not taken according to the requirement of appointment, do not form corresponding 3D model data yet, can not directly in dressing system, use on the net, but by the present invention these online existing clothes data resources can be processed obtain can for online dressing system can the clothes data, for user's try-on garment on the net, specifically, but acquisition resolution is not less than the positive clear pictures of the online existing clothes of 512*512 pixel, as direct copying clothes data resource from network,

Step S22, remove the background of described clothes full face;

This step is called in iconology scratches figure, specifically can adopt the background of common picture processing software except described clothes full face, described picture processing software can be the stingy figure instrument in PhotoShop, the photo fairly simple for background can be removed by the magic wand tool in PhotoShop automatically as color background, photo for the background more complicated, it is often clean not that software is removed background automatically, can carry out manual removal by picture processing software;

Step S23, arrange the clothes key point of described clothes full face;

Preferably, described clothes key point can comprise a kind of or combination in any in the position data that characterizes collar, shoulder, sleeve, the bottom, waistband, front wave, trouser legs or leg opening; Described clothes key point can mean in the mode of point, certainly, in order to obtain more intuitive effect, described clothes key point also can mean in the mode of Points And lines, specifically, for upper dress, the method that the clothes key point of clothes full face is set is to click the position of collar, shoulder, sleeve or the bottom on the clothes photo with mouse, and the clothes key point of upper dress shows in the mode of Points And lines; For trousers, the method that the clothes key point of clothes full face is set is to click and to comprise the positions such as waistband position, front wave, trouser legs or leg opening on the clothes photo with mouse, and the clothes key point of lower dress shows in the mode of Points And lines;

Step S24, arrange the dimensional data of described clothes full face according to described garment dimension data;

Wherein, the garment dimension data representation clothes there is the different size of how many kinds of, every kind of concrete numerical value of size is how many, accordingly, the dimensional data of the clothes full face of upper dress comprises clothing length, shoulder breadth, collar, chest measurement, waistline, hip circumference, sleeve length, the sleeve a kind of or combination in any in enclosing, and the dimensional data of the clothes full face of lower dress comprises trousers length, front wave, waistline, hip circumference, thigh circumference, calf circumference or the leg opening a kind of or combination in any in enclosing;

Step S25, be stored into described clothes full face and clothes key point thereof and dimensional data in garment data;

Step S26, set up according to the clothes full face obtained from garment data and clothes key point thereof and dimensional data the dress planar grid group be comprised of a plurality of the second grids on isoplanar;

Wherein, described the second grid can be triangle, and all triangle forms a complete dress planar grid group, can certainly be set to according to the described grid of the requirement of actual computer graph rendering other shapes that need such as square, rectangle or rhombus; Described dress planar grid group is set up according to the profile of clothes, and dress planar grid group inside is segmented by triangle;

Preferably, each summit of each the second grid is corresponding one by one with each pinup picture coordinate of described clothes full face, as three summits of triangular mesh with as described in each pinup picture coordinate of clothes photo corresponding one by one, the information that each summit of all grids comprises the corresponding pinup picture coordinate in described clothes photo, the pinup picture coordinate just can reflect the corresponding relation between each summit of grid and each relevant position of clothes full face like this, by pinup picture coordinate and grid vertex position, can render dress form;

Step S27, generate the second normal of the steric information that means each the second grid according to described clothes photo;

Concrete, step S26 and S27 are clothes F3D modeling process, clothes F3D modeling process is a kind of plane three-dimensional technology, generate the dress planar grid group (being called the F3D grid) with normal, the F3D grid refers to sets up a kind of 3D grid model at grade, means the position on each summit of each grid with the x, y, z coordinate, the z=0 on the summit of F3D grid, to guarantee all F3D grids fully at grade, grid can only the edge axle vertical with grid surface be that the z axle rotates;

Mean the fluctuating of clothes F3D grid surface in step S26 with normal, it should be noted that, the normal here no longer only refers to the ray perpendicular to each grid surface, in order to mean the different lightnesses of each grid, different normals can be corresponding and angled perpendicular to the ray of grid surface, due to the lightness difference of each face, normal may, to left avertence, to right avertence or be not offset, be stereoscopic sensation to show different grid surface lightnesses with respect to the ray perpendicular to grid surface;

Step S28, generate the clothes grid model generated by described the second normal and clothes plane grid group.

Preferably, also can comprise according to described dress planar grid group and described the second normal and render the clothes solid model after step S28, specifically can render along the axle perpendicular to described dress planar grid group described clothes solid model, generate dress form as with camera points or the complete vertical z direction of principal axis of dress planar grid group grid, played up clothes F3D grid, in order to normal, catoptric impact being shown to stereoscopic sensation.

To sum up, in the present embodiment, set up by obtain data from garment data the clothes grid model that comprises that dress planar grid group and normal form, thereby can render fast, realistically the clothes solid model.

Embodiment tri-

As shown in Figure 9, the present invention also provides a kind of method of fitting, adopts the described method of setting up dress form of embodiment bis-to be fitted, and comprising:

Step S31, generate the body area network lattice model according to human body photo and build data;

Wherein, step S31 specifically can comprise:

Step S311, generate the human body plane grid group be comprised of a plurality of human joint pointses and a plurality of the first grids at grade as shown in figure 10 according to human body photo and build data;

Preferably, described the first grid can be triangle, all triangles can form a complete human body plane grid group, can certainly be set to other shape needed such as square, rectangle or rhombus according to the described grid of the requirement of actual computer graph rendering;

Preferably, can described the first grid be carried out to radial distribution centered by described human joint points at described human joint points place, described grid can be rotated along the axle perpendicular to described human body plane grid group around described human joint points like this, to realize human posture's adjustment;

Further, each summit of each the first grid is corresponding one by one with each pinup picture coordinate of described human body photo, as three summits of triangle the first grid with as described in each pinup picture coordinate of human body photo corresponding one by one, the information that each summit of all grids comprises the corresponding pinup picture coordinate in described human body photo, the pinup picture coordinate just can reflect the corresponding relation between each summit of grid and each relevant position of human body photo like this, by pinup picture coordinate and grid vertex position, can render manikin, specifically can camera points or human body plane grid group fully vertical z direction of principal axis play up human body plane grid group and generate manikin, in order to normal, catoptric impact being shown to stereoscopic sensation,

Step S312, generate the first normal of the steric information that means each the first grid according to described human body photo;

Specifically, step S311 and S312 are human body F3D modeling process, the F3D modeling process is kind of a plane three-dimensional technology, generate the human body plane grid group (being called the F3D grid) with normal, the F3D grid refers to sets up a kind of 3D grid model at grade, the position that means each summit of each grid with the x, y, z coordinate, the z=0 on the summit of F3D grid, to guarantee all F3D grids fully at grade.Grid can only rotate by the edge z axle vertical with grid surface, for example grid can rotate around human joint points, described human joint points comprises head, health, the extremities joint point, as comprised neck node, shoulder node, ancon node, wrist node, palm node, wrist node, waist node, buttocks node, thigh node, knee node, ankle node etc.;

Mean the fluctuating of human body F3D grid surface in step S312 with normal, it should be noted that, the normal here no longer only refers to the ray perpendicular to each grid surface, in order to mean the different lightnesses of each grid, different normals can be corresponding and angled perpendicular to the ray of grid surface, due to the difference of lightness, normal may, to left avertence, to right avertence or be not offset, be stereoscopic sensation to show different grid surface lightnesses with respect to the ray perpendicular to grid surface;

Step S32, gather clothes full face and clothes dimensional data;

Wherein, because clothes electricity business generally can provide a large amount of existing clothes data resources that comprises clothes full face and corresponding garment dimension data on the webpage of garment marketing, although these clothes full faces are not taken according to the requirement of appointment, do not form corresponding 3D model data yet, can not directly in dressing system, use on the net, but by the present invention these online existing clothes data resources can be processed obtain can for online dressing system can the clothes data, for user's try-on garment on the net, specifically, but acquisition resolution is not less than the positive clear pictures of the online existing clothes of 512*512 pixel, as direct copying clothes data resource from network,

Step S33, remove the background of described clothes full face;

This step is called in iconology scratches figure, specifically can adopt the background of common picture processing software except described clothes full face, described picture processing software can be the stingy figure instrument in PhotoShop, the photo fairly simple for background can be removed by the magic wand tool in PhotoShop automatically as color background, photo for the background more complicated, it is often clean not that software is removed background automatically, can carry out manual removal by picture processing software;

Step S34, arrange the clothes key point of described clothes full face;

Preferably, described clothes key point can comprise a kind of or combination in any in the position data that characterizes collar, shoulder, sleeve, the bottom, waistband, front wave, trouser legs or leg opening; Described clothes key point can mean in the mode of point, certainly, in order to obtain more intuitive effect, described clothes key point also can mean in the mode of Points And lines, specifically, for upper dress, the method that the clothes key point of clothes full face is set is to click the position of collar, shoulder, sleeve or the bottom on the clothes photo with mouse; For trousers, the method that the clothes key point of clothes full face is set is to click and to comprise the positions such as waistband position, front wave, trouser legs or leg opening on the clothes photo with mouse;

Step S35, arrange the dimensional data of described clothes full face according to described garment dimension data;

Wherein, the garment dimension data representation clothes there is the different size of how many kinds of, every kind of concrete numerical value of size is how many, accordingly, the dimensional data of the clothes full face of upper dress comprises clothing length, shoulder breadth, collar, chest measurement, waistline, hip circumference, sleeve length, the sleeve a kind of or combination in any in enclosing, and the dimensional data of the clothes full face of lower dress comprises trousers length, front wave, waistline, hip circumference, thigh circumference, calf circumference or the leg opening a kind of or combination in any in enclosing;

Step S36, be stored into described clothes full face and clothes key point thereof and dimensional data in garment data;

Step S37, set up according to the clothes full face obtained from garment data and clothes key point thereof and dimensional data the dress planar grid group be comprised of a plurality of the second grids on isoplanar;

Wherein, described the second grid can be triangle, and all triangle forms a complete dress planar grid group, can certainly be set to according to the described grid of the requirement of actual computer graph rendering other shapes that need such as square, rectangle or rhombus; Described dress planar grid group is set up according to the profile of clothes, and dress planar grid group inside is segmented by triangle,

Preferably, each summit of each the second grid is corresponding one by one with each pinup picture coordinate of described clothes full face, as three summits of triangular mesh with as described in each pinup picture coordinate of clothes photo corresponding one by one, the information that each summit of all grids comprises the corresponding pinup picture coordinate in described clothes photo, the pinup picture coordinate just can reflect the corresponding relation between each summit of grid and each relevant position of clothes full face like this, by pinup picture coordinate and grid vertex position, can render dress form;

Step S38, generate the second normal of the steric information that means each the second grid according to described clothes photo;

Concrete, step S37 and S38 are clothes F3D modeling process, clothes F3D modeling process is a kind of plane three-dimensional technology, generate the dress planar grid group (being called the F3D grid) with normal, the F3D grid refers to sets up a kind of 3D grid model at grade, means the position on each summit of each grid with the x, y, z coordinate, the z=0 on the summit of F3D grid, to guarantee all F3D grids fully at grade, grid can only the edge axle vertical with grid surface be that the z axle rotates;

Mean the fluctuating of clothes F3D grid surface in step S37 with normal, it should be noted that, the normal here no longer only refers to the ray perpendicular to each grid surface, in order to mean the different lightnesses of each grid, different normals can be corresponding and angled perpendicular to the ray of grid surface, due to the lightness difference of each face, normal with respect to the ray perpendicular to grid surface may to left avertence, also may be to right avertence or be not offset, to show different grid surface lightnesses, be stereoscopic sensation;

Step S39, generate the clothes grid model generated by described the second normal and clothes plane grid group;

Preferably, after step S39, also can comprise according to the body area network lattice model and adjust described clothes grid model, be that the clothes grid model is determined size by body area network lattice model lattice, thereby make the clothes human body of can fitting, for example, when garment dimension small during in human dimension, can automatically adjust the dress planar grid, simulate the effect of clothes tension, avoid occurring that some position of human body is exposed to the effect outside clothes; When garment dimension is greater than human dimension, can adjusts the dress planar grid, thereby make the clothes human body of as far as possible can fitting;

Step S40, combine described body area network lattice model to align to generate the fitting grid model with the clothes grid model;

Wherein, can described body area network lattice model be combined and align with described clothes grid model by described human joint points and clothes key point, for example upper garment snaps to upper half of human body, trousers snap to the human body lower part of the body, obtain the fitting grid model that body area network lattice model 1 as shown in figure 11 combines after aliging withclothes grid model 2;

Step S41, render the fitting solid model according to described fitting grid model.

Preferably, can be by described human body plane grid group, the first normal, dress planar grid group and the second normal along perpendicular to the axle of described fitting grid model, playing up described fitting solid model, obtain the fitting solid model after user as shown in figure 12 wears shirt and jeans, make the user check easily clothes effect.

To sum up, the generation fitting can be quick true to nature of the data in the garment data generated by use in the present embodiment solid model, show user's clothes effect, and whether the user can understand garment dimension intuitively suitable.

Embodiment tetra-

As shown in figure 13, the present invention also provides a kind of garment data generation system, comprises

Acquisition module 11, removal background module 12, key point module 13, size module 14 and clothes database module 15.

Acquisition module 11 is for gathering the clothes full face.

Wherein, described acquisition module gathers described clothes full face in existing clothes photo from network, but concrete acquisition resolution is not less than the positive clear pictures of online existing clothes that 512 pixels are multiplied by 512 pixels.

Remove background module 12 for removing the background of described clothes full face, specifically can adopt the background of common picture processing software except described clothes full face, described picture processing software can be the stingy figure instrument in PhotoShop, the photo fairly simple for background can be removed by the magic wand tool in PhotoShop automatically as color background, photo for the background more complicated, it is often clean not that software is removed background automatically, can carry out manual removal by picture processing software.

Key point module 13 is for arranging the clothes key point of described clothes full face.

Preferably, described clothes key point can comprise a kind of or combination in any in the position data that characterizes collar, shoulder, sleeve, the bottom, waistband, front wave, trouser legs or leg opening; Described clothes key point can mean in the mode of point, and certainly, in order to obtain more intuitive effect, described clothes key point also can mean in the mode of Points And lines.

Size module 14 is for arranging the dimensional data of described clothes full face.

Wherein, the garment dimension data representation clothes there is the different size of how many kinds of, every kind of concrete numerical value of size is how many, accordingly, the dimensional data of the clothes full face of upper dress comprises clothing length, shoulder breadth, collar, chest measurement, waistline, hip circumference, sleeve length, the sleeve a kind of or combination in any in enclosing, and the dimensional data of the clothes full face of lower dress comprises trousers length, front wave, waistline, hip circumference, thigh circumference, calf circumference or the leg opening a kind of or combination in any in enclosing.

Garment data module 15 is for storing described clothes full face and clothes key point and size.

To sum up, the present embodiment can be applicable in online dressing system, and online dressing system needs fast, while setting up garment data at low cost, can generate by the method for the present embodiment.

Embodiment five

As shown in figure 14, the present invention also provides a kind of system of setting up dress form, comprises acquisition module 21, removes background module 22, key point module 23, size module 24, garment data module 25, clothes grid group module 26, the second normal module 27 and clothes grid model module 28.

Acquisition module 21 is for gathering the clothes full face.

Wherein, described acquisition module gathers described clothes full face in existing clothes photo from network, but concrete acquisition resolution is not less than the positive clear pictures of online existing clothes that 512 pixels are multiplied by 512 pixels.

Remove background module 22 for removing the background of described clothes full face, specifically can adopt the background of common picture processing software except described clothes full face, described picture processing software can be the stingy figure instrument in PhotoShop, the photo fairly simple for background can be removed by the magic wand tool in PhotoShop automatically as color background, photo for the background more complicated, it is often clean not that software is removed background automatically, can carry out manual removal by picture processing software.

Key point module 23 is for arranging the clothes key point of described clothes full face.

Preferably, described clothes key point can comprise a kind of or combination in any in the position data that characterizes collar, shoulder, sleeve, the bottom, waistband, front wave, trouser legs or leg opening; Described clothes key point can mean in the mode of point, and certainly, in order to obtain more intuitive effect, described clothes key point also can mean in the mode of Points And lines.

Size module 24 is for arranging the dimensional data of described clothes full face.

Wherein, the garment dimension data representation clothes there is the different size of how many kinds of, every kind of concrete numerical value of size is how many, accordingly, the dimensional data of the clothes full face of upper dress comprises clothing length, shoulder breadth, collar, chest measurement, waistline, hip circumference, sleeve length, the sleeve a kind of or combination in any in enclosing, and the dimensional data of the clothes full face of lower dress comprises trousers length, front wave, waistline, hip circumference, thigh circumference, calf circumference or the leg opening a kind of or combination in any in enclosing.

Garment data module 25 is for storing described clothes full face and clothes key point and size.

The clothes full face that clothes grid group module 26 is obtained from garment data for basis and clothes key point thereof and dimensional data are set up the dress planar grid group be comprised of a plurality of the second grids isoplanar.

Wherein, described the second grid can be triangle, and all triangle forms a complete dress planar grid group, can certainly be set to according to the described grid of the requirement of actual computer graph rendering other shapes that need such as square, rectangle or rhombus; Described dress planar grid group is set up according to the profile of clothes, and dress planar grid group inside is segmented by triangle.

Preferably, each summit of each the second grid is corresponding one by one with each pinup picture coordinate of described clothes full face, as three summits of triangular mesh with as described in each pinup picture coordinate of clothes photo corresponding one by one, the information that each summit of all grids comprises the corresponding pinup picture coordinate in described clothes photo, the pinup picture coordinate just can reflect the corresponding relation between each summit of grid and each relevant position of clothes full face like this, by pinup picture coordinate and grid vertex position, can render dress form.

The second normal module 27 is for generating the second normal of the steric information that means each the second grid according to described clothes photo.

Clothes grid model module 28 is for generating the clothes grid model generated by described the second normal and clothes plane grid group.

Preferably, the system of setting up dress form also comprises that a rendering module 29 is for according to described dress planar grid group and described the second normal, rendering the clothes solid model, and described rendering module 29 renders described clothes solid model for the axle along perpendicular to described dress planar grid group.

To sum up, in the present embodiment, set up by obtain data from garment data the clothes grid model that comprises that dress planar grid group and normal form, thereby can render fast, realistically the clothes solid model.

Embodiment six

As shown in figure 15, the present invention also provides a kind of dressing system, collection module 31, removal background module 32, key point module 33, size module 34, garment data module 35, clothesgrid group module 36, the second normal module 37, clothes grid model module 38, body area network lattice model module 39, fitting grid model module 40 and fitting solid model module 41.

Acquisition module 31 is for gathering the clothes full face.

Wherein, described acquisition module gathers described clothes full face in existing clothes photo from network, but concrete acquisition resolution is not less than the positive clear pictures of online existing clothes that 512 pixels are multiplied by 512 pixels.

Remove background module 32 for removing the background of described clothes full face, specifically can adopt the background of common picture processing software except described clothes full face, described picture processing software can be the stingy figure instrument in PhotoShop, the photo fairly simple for background can be removed by the magic wand tool in PhotoShop automatically as color background, photo for the background more complicated, it is often clean not that software is removed background automatically, can carry out manual removal by picture processing software.

Key point module 33 is for arranging the clothes key point of described clothes full face.

Preferably, described clothes key point can comprise a kind of or combination in any in the position data that characterizes collar, shoulder, sleeve, the bottom, waistband, front wave, trouser legs or leg opening; Described clothes key point can mean in the mode of point, and certainly, in order to obtain more intuitive effect, described clothes key point also can mean in the mode of Points And lines.

Size module 34 is for arranging the dimensional data of described clothes full face.

Wherein, the garment dimension data representation clothes there is the different size of how many kinds of, every kind of concrete numerical value of size is how many, accordingly, the dimensional data of the clothes full face of upper dress comprises clothing length, shoulder breadth, collar, chest measurement, waistline, hip circumference, sleeve length, the sleeve a kind of or combination in any in enclosing, and the dimensional data of the clothes full face of lower dress comprises trousers length, front wave, waistline, hip circumference, thigh circumference, calf circumference or the leg opening a kind of or combination in any in enclosing.

Garment data module 35 is for storing described clothes full face and clothes key point and size.

The clothes full face that clothesgrid group module 36 is obtained from garment data for basis and clothes key point thereof and dimensional data are set up the dress planar grid group be comprised of a plurality of the second grids isoplanar.

Wherein, described the second grid can be triangle, and all triangle forms a complete dress planar grid group, can certainly be set to according to the described grid of the requirement of actual computer graph rendering other shapes that need such as square, rectangle or rhombus; Described dress planar grid group is set up according to the profile of clothes, and dress planar grid group inside is segmented by triangle.

Preferably, each summit of each the second grid is corresponding one by one with each pinup picture coordinate of described clothes full face, as three summits of triangular mesh with as described in each pinup picture coordinate of clothes photo corresponding one by one, the information that each summit of all grids comprises the corresponding pinup picture coordinate in described clothes photo, the pinup picture coordinate just can reflect the corresponding relation between each summit of grid and each relevant position of clothes full face like this, by pinup picture coordinate and grid vertex position, can render dress form.

The second normal module 37 is for generating the second normal of the steric information that means each the second grid according to described clothes photo.

Clothes grid model module 38 is for generating the clothes grid model generated by described the second normal and clothes plane grid group.

Body area network lattice model module 39 is for generating the body area network lattice model according to human body photo and build data.

Wherein, described body area network lattice model module 39 comprises volume mesh pack module 391 and first method wire module 292.

Human body grid pack module 391 is for generating according to human body photo and build data the human body plane grid group be comprised of a plurality of human joint pointses and a plurality of the first grids at grade, and wherein said human body grid pack module 291 for carrying out radial distribution by described the first grid outside described human joint points centered by described human joint points.

First method wire module 392 is for generating the first normal of the steric information that means each the first grid according to described human body photo.

Fitting grid model module 40 is for combining described body area network lattice model to align to generate the fitting grid model with the clothes grid model.

Fitting solid model module 41 is for rendering the fitting solid model according to described fitting grid model.

Data in the garment data generated by use in the present embodiment generation fitting can be quick true to nature solid model, show user's clothes effect, and whether the user can understand garment dimension intuitively suitable.

In sum, the present invention is by gathering the clothes full face, remove the background of described clothes full face, the clothes key point of described clothes full face is set and the dimensional data of described clothes full face is set according to described garment dimension data, and described clothes full face and clothes key point thereof and dimensional data are stored in garment data, can fully improve the resource utilization of existing dress planar photo and clothes dimensional data, directly generate for setting up the garment data of dress form, reduce and generate garment data manpower and construction and the operation cost of time loss significantly to reduce online dressing system, improve the commercial viability of online dressing system.

In addition, the clothes full face that the present invention obtains from garment data by basis and clothes key point thereof and dimensional data are set up the dress planar grid group be comprised of a plurality of the second grids on isoplanar, and generate the second normal of the steric information that means each the second grid according to described clothes photo, then generate the clothes grid model generated by described the second normal and clothes plane grid group, can generate fast, realistically the clothes solid model.

In addition, the present invention is by generating the body area network lattice model according to human body photo and build data, and described body area network lattice model is combined and aligns to generate the fitting grid model with the clothes grid model, then render the fitting solid model according to described fitting grid model, can quick generation true to nature and show user's clothes effect, whether the user can understand garment dimension intuitively suitable.

In this instructions, each embodiment adopts the mode of going forward one by one to describe, and what each embodiment stressed is and the difference of other embodiment that between each embodiment, identical similar part is mutually referring to getting final product.For the disclosed system of embodiment, owing to corresponding to the method disclosed in Example, so description is fairly simple, relevant part partly illustrates and gets final product referring to method.

The professional can also further recognize, unit and the algorithm steps of each example of describing in conjunction with embodiment disclosed herein, can realize with electronic hardware, computer software or the combination of the two, for the interchangeability of hardware and software clearly is described, composition and the step of each example described according to function in the above description in general manner.These functions are carried out with hardware or software mode actually, depend on application-specific and the design constraint of technical scheme.The professional and technical personnel can specifically should be used for realizing described function with distinct methods to each, but this realization should not thought and exceeds scope of the present invention.

Obviously, those skilled in the art can carry out various changes and modification and not break away from the spirit and scope of the present invention invention.Like this, if, within of the present invention these are revised and modification belongs to the scope of the claims in the present invention and equivalent technologies thereof, the present invention also is intended to comprise these change and modification.

Claims (26)

1. a garment data generation method, is characterized in that, comprising:

Gather clothes full face and clothes dimensional data;

Remove the background of described clothes full face;

The clothes key point of described clothes full face is set;

The dimensional data of described clothes full face is set according to described garment dimension data; And

Described clothes full face and clothes key point thereof and dimensional data are stored in garment data.

2. garment data generation method as claimed in claim 1, is characterized in that, described clothes key point comprises a kind of or combination in any in the position data that characterizes collar, shoulder, sleeve, the bottom, waistband, front wave, trouser legs or leg opening.

3. garment data generation method as claimed in claim 1, is characterized in that, described clothes key point means in the mode of point.

4. garment data generation method as claimed in claim 1, is characterized in that, described clothes key point means in the mode of Points And lines.

5. as the described garment data of claim 1 to 4 any one generation method, it is characterized in that, the dimensional data of described clothes full face comprises that clothing length, shoulder breadth, collar, chest measurement, waistline, hip circumference, sleeve length, sleeve enclose, trousers length, front wave, waistline, hip circumference, thigh circumference, calf circumference or the leg opening a kind of or combination in any in enclosing.

6. an employing is set up the method for dress form as garment data generation method as described in claim 1 to 5 any one, it is characterized in that, comprising:

Set up according to the clothes full face obtained and clothes key point thereof and dimensional data the dress planar grid group formed by a plurality of the second grids on isoplanar from garment data;

Generate the second normal of the steric information that means each the second grid according to described clothes photo; And

The clothes grid model that generation is generated by described the second normal and clothes plane grid group.

7. the method for setting up dress form as claimed in claim 6, it is characterized in that, after the step of the clothes grid model that generation is generated by described the second normal and clothes plane grid group, also comprise according to described dress planar grid group and described the second normal and render the clothes solid model.

8. the method for setting up dress form as claimed in claim 7, it is characterized in that, render in the step of clothes solid model according to described dress planar grid group and described the second normal, along the axle perpendicular to described dress planar grid group, render described clothes solid model.

9. the method for setting up dress form as claimed in claim 7, is characterized in that, described the second grid is triangle.

10. the method for setting up dress form as claimed in claim 7, is characterized in that, each summit of each the second grid is corresponding one by one with each pinup picture coordinate of described clothes full face.

11. the method that the employing method of setting up dress form as described as claim 6 to 10 any one is fitted, is characterized in that, comprising:

Generate the body area network lattice model according to human body photo and build data;

Described body area network lattice model is combined and aligns to generate the fitting grid model with the clothes grid model; And

Render the fitting solid model according to described fitting grid model.

12. the method for fitting as claimed in claim 11, is characterized in that, comprises according to the step of human body photo and build data generation body area network lattice model:

Generate according to human body photo and build data the human body plane grid group formed by a plurality of human joint pointses and a plurality of the first grids at grade;

Generate the first normal of the steric information that means each the first grid according to described human body photo.

13. fitting method as claimed in claim 12, is characterized in that, described the first grid is triangle.

14. the method for fitting as claimed in claim 12, it is characterized in that, generate in the step of the human body plane grid group formed by a plurality of human joint pointses and a plurality of the first grids at grade according to human body photo and build data, in the place of described human joint points, described the first grid is carried out to radial distribution centered by described human joint points.

15. the method for fitting as claimed in claim 11, it is characterized in that, generate in the step of the human body plane grid group formed by a plurality of human joint pointses and a plurality of the first grids at grade according to human body photo and build data, described the first grid is rotated along the axle perpendicular to described human body plane grid group around described human joint points.

16. fitting method as claimed in claim 11, is characterized in that, each summit of each the first grid is corresponding one by one with each pinup picture coordinate of described human body photo.

17. fitting method as claimed in claim 11, it is characterized in that, set up the step of the dress planar grid group formed by a plurality of the second grids on isoplanar according to the clothes full face obtained and clothes key point thereof and dimensional data from garment data after, also comprise according to the body area network lattice model and adjust described clothes grid model.

18. fitting method as claimed in claim 11, it is characterized in that, described body area network lattice model and clothes grid model are combined and align to generate in the step of fitting grid model, by described human joint points and clothes key point, described body area network lattice model is combined and aligns with described clothes grid model.

19. fitting method as claimed in claim 11, it is characterized in that, render according to described fitting grid model in the step of fitting solid model, by described human body plane grid group, the first normal, dress planar grid group and the second normal edge, perpendicular to the axle of described fitting grid model, play up described fitting solid model.

20. a garment data generation system, is characterized in that, comprising:

Acquisition module, for gathering the clothes full face;

Remove the background module, for removing the background of described clothes full face;

The key point module, for arranging the clothes key point of described clothes full face;

Size module, for arranging the dimensional data of described clothes full face; And

The garment data module, for storing described clothes full face and clothes key point and size.

21. a system of setting up dress form, is characterized in that, comprising:

Garment data generation system as claimed in claim 20;

Clothes grid group module, the clothes full face obtained from garment data for basis and clothes key point thereof and dimensional data are set up the dress planar grid group be comprised of a plurality of the second grids isoplanar;

The second method wire module, for generating the second normal of the steric information that means each the second grid according to described clothes photo; And

Clothes grid model module, for generating the clothes grid model generated by described the second normal and clothes plane grid group.

Set up the system of dress form 22. as claimed in claim 21, it is characterized in that, also comprise that a rendering module is for rendering the clothes solid model according to described dress planar grid group and described the second normal.

23. set up the system of dress form as claimed in claim 21, it is characterized in that, described rendering module, render described clothes solid model for the axle along perpendicular to described dress planar grid group.

24. a dressing system, is characterized in that, comprising:

The system of setting up dress form as described as any one in claim 21 to 23:

Body area network lattice model module, for generating the body area network lattice model according to human body photo and build data;

Fitting grid model module, for combining described body area network lattice model to align to generate the fitting grid model with the clothes grid model; And

Fitting solid model module, for rendering the fitting solid model according to described fitting grid model.

25. dressing system as claimed in claim 24, is characterized in that, described body area network lattice model module comprises:

Human body grid pack module, for generating according to human body photo and build data the human body plane grid group be comprised of a plurality of human joint pointses and a plurality of the first grids at grade;

The first method wire module, for generating the first normal of the steric information that means each the first grid according to described human body photo.

26. dressing system as claimed in claim 24, is characterized in that, described human body grid pack module, for carrying out radial distribution by described the first grid outside described human joint points centered by described human joint points.

Images