CN103093498A - Three-dimensional human face automatic standardization method - Google Patents
Three-dimensional human face automatic standardization method Download PDFInfo
- Publication number
- CN103093498A CN103093498A CN2013100296453A CN201310029645A CN103093498A CN 103093498 A CN103093498 A CN 103093498A CN 2013100296453 A CN2013100296453 A CN 2013100296453A CN 201310029645 A CN201310029645 A CN 201310029645A CN 103093498 A CN103093498 A CN 103093498A
- Authority
- CN
- China
- Prior art keywords
- dimensional
- template
- texture image
- dimensional face
- planar
- 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.)
- Pending
Links
Images
Landscapes
- Image Generation (AREA)
- Processing Or Creating Images (AREA)
Abstract
The invention discloses a three-dimensional human face automatic standardization method. All the models in a three-dimensional human face bank have equal numbers of vertex and alignment based on the features are achieved. The three-dimensional human face automatic standardization method includes three aspects: (1) an automatic generation method of a planar triangular mesh template is designed according to a texture image of the three dimensional human face models. (2) the standardization method based on the planar triangular mesh template resampling is designed so that after the standardization, different three dimensional human faces have the same numbers of vertex and topological structures. (3) the three dimensional meshes based on a planar pattern is established. For any three dimensional human faces, after the processing over the three dimensional human face data base, the three-dimensional human face automatic standardization method can achieve the applications like a linear operation, the human face model with multi-resolution and the mixed human models and the like.
Description
Technical field
The present invention relates to the fields such as computer graphics, Digital Image Processing, three-dimensional measurement, specifically for the three-dimensional face data, is a kind of three-dimensional face model point cloud, standardized method of grid data of realizing.。
Background technology
Along with three-dimensional acquisition equipment (as laser scanner, structured light collector etc.) price is progressively cheap, the means that people obtain three-dimensional data are more and more abundanter.Three-dimensional face opposite planar photo is more lively, can fully show facial a plurality of visual angles information, all has broad application prospects at aspects such as 3 D-printing, recognitions of face.
In a lot of application scenarios (as three-dimensional face identification, 3 D human face animation etc.), the three-dimensional face storehouse that establishment comprises a plurality of three-dimensional face models is very necessary, and then it is very necessary that the model in three-dimensional face model storehouse is carried out standardization.For the ease of problem description, suppose to exist a three-dimensional face storehouse, the below carries out formalization to three-dimensional face model data wherein: the three-dimensional face data are arranged according to the summit order, and any one faceform is with following two vector representations
Wherein, s
iThe shape vector of i faceform in face database, by the D coordinates value (x on each summit
j, y
j, z
j) arranged sequentially combining; t
iBe this faceform's texture, similarly, it is by the color value (r on each summit
k, g
k, b
k) combine 1≤j≤n
i, n
iIt is the summit number of this people's face.Clearly, can't directly perform mathematical calculations between the three-dimensional face data of directly obtaining by spatial digitizer, reason is:
1. due to the individual difference of people's face, the summit number that consists of different three-dimensional faces is not quite similar, namely when i ≠ j (i, j refer to respectively different models), and n
i≠ n
jThereby two people's face vector dimensions are different, can't directly carry out vector operation;
2. the order the when numbering on summit is only with collection is relevant, and irrelevant with feature, namely k summit of k summit of i people's face and j people's face is not to be people's same unique point on the face.
Above two aspect factors have affected the operability of three-dimensional face database, can't directly carry out linear operation between people's face data, need standardization.From mathematics, the face database of Criterion is exactly to be unified vector form with the face representation in database, make all three-dimensional face data have identical number of vertex and the feature set up between people's face corresponding.After standardization, identical unique point should be in identical position in different people's face vectors.For example, people nose summit and another one people nose summit on the face on the face is all k summit in vector.
In fact, set up based on the point-to-point accurate correspondence of feature very difficultly on three-dimensional data, some scholars also did relevant work both at home and abroad.Due to the individual difference of different people face, the three-dimensional geometry of people's face is widely different, and the three-dimensional face data are dense point sets, and data volume is very large, is difficult to use conventional method to set up this corresponding relation.The people such as Vetter are at " Estimating coloured 3D face models from single images:An example based approach " Lecture Notes in Computer Science.1998, three-dimensional face being carried out cylinder unwrapping in 1407:499-513 is shape and texture image, utilizes color texture information and shape image information to set up the corresponding relation of three-dimensional data and the standard faces of any people's face by optical flow algorithm.the people such as Vetter are at " A bootstrapping algorithm for learning linear models of object classes " IEEE Conference on Computer Vision and Pattern Recognition.PuertoRico, USA, 1997:40-47 improves optical flow algorithm with Bootstrapping, but because the prerequisite of optical flow algorithm is to suppose that the variation of light stream between two width images is smooth continuously, to with the near people's appearance of reference man's appearance to effectively, and to the people face larger with reference man's face difference, the hypotheses of optical flow computation can not satisfy, there is certain error in result of calculation.In order to improve the alignment effect of three-dimensional face, the people such as Gu Chunliang are at " Resampling Based Method for Pixel-wise Correspondence between3D Faces " International Conference on Information Technology:Coding and Computing, 2004.Las Vegas, Nevada, USA has proposed the method based on mesh resampling in 2004:614-619.At first use unified dividing method to carry out piecemeal everyone face according to face characteristic, then each dough sheet is carried out identical resampling.Can make the corresponding piecemeal of different people face have identical topological structure owing to resampling, therefore can naturally set up the alignment of people's face, and obtain reasonable matching effect.On this basis, Yin is precious just waits the people at " the non-uniform sampling alignment algorithm of three-dimensional face " the journal .2007 of Beijing University of Technology, 33 (2): proposed again the three-dimensional face alignment algorithm based on non-uniform sampling in 213-218, analyze the curvature information in people's face various zones territory, carried out mesh resampling heterogeneous by the complex-shaped property that curvature reflects.These methods all need the manual partitioned mode of determining.
Summary of the invention
Technical matters to be solved by this invention is the method in design a kind of automatic Criterion three-dimensional face storehouse, planar triangulations template synthetic according to three-dimensional face model, and then realize all three-dimensional models standardization efficiently and accurately in the three-dimensional face storehouse.The present invention mainly comprises following three aspects: content:
The present invention realizes that the technical scheme that its goal of the invention adopts is:
A kind of three-dimensional face automatic standardizing method based on the planar triangulations template, planar triangulations template synthetic according to three-dimensional face model, and then realize all three-dimensional models standardization efficiently and accurately in the three-dimensional face storehouse, comprise following steps:
(1) automatically create the planar triangulations template
1.1 generating three-dimensional faceform's planar grains image;
1.2 the AAMs model of unique point on the training texture image (Active Appearance Models, active appearance models);
1.3 create the planar triangulations template;
(2) three-dimensional model based on the planar triangulations template resamples
The plane template that adopts (1) to obtain resamples to the resolution of each triangular plate on each input people face texture maps according to corresponding triangular plate on plane template.According to the one-to-one relationship between 2 d texture image and three-dimensional point cloud, the three-dimensional face model after being resampled makes different three-dimensional faces have identical number of vertex and topological structure after standardization;
(3) rebuild based on the 3D grid of plane template
At first the annexation according to pixel on horizontal direction on plane template and vertical direction is linked in sequence the summit in three dimensions; Next connect the lower left corner and to the diagonal line in the upper right corner, each little rectangle be divided into two triangular plates, and adopt from left to right, mode from top to bottom will be numbered summit and little triangle; At last, with all piecemeal combinations, generate the three-dimensional face model of ultimate criterion.
Adopt the present invention, for the foundation of three-dimensional face database provides a kind of standardized method, do not need manual intervention, can automatically realize the correspondence between three-dimensional face.It is to be used for by the three-dimensional face data that three-dimensional acquisition equipment obtains the important foundation that recognition of face and human face animation etc. are used.
Description of drawings
Fig. 1 is 34 unique points marking on texture maps and 4 angle points on image, the effect of having showed subdivision in figure;
Fig. 2 is that the present invention realizes three-dimensional face standardization flow process;
Fig. 3 is planar triangulations template establishment flow process.Subgraph (a) is average two-dimensional shapes, and subgraph (b) is the triangulation result, and subgraph (c) is the planar triangulations template, and subgraph (d) is the detail section of template;
Fig. 4 is that 3D grid is rebuild and the numbering schematic diagram.Subgraph (a) is scattered points and the numbering that resamples, and subgraph (b) is mesh reconstruction schematic diagram and triangular plate method for numbering serial schematic diagram.
Embodiment
Below in conjunction with accompanying drawing, the present invention is done further detailed description.
The present invention includes three steps: (1) creates the planar triangulations template automatically, and flow process is seen accompanying drawing 3; (2) based on the resampling of plane template, flow process is seen accompanying drawing 2; (3) rebuild based on the 3D grid of plane template.Process is seen accompanying drawing 4.Wherein, the 1st, 2 steps were core features of the present invention.The below introduces specific implementation method step by step:
Step 1: create the planar triangulations template
1.1 generating three-dimensional faceform's planar grains image
The present invention proposes the texture drawing generating method based on cylindrical coordinates.The texture image of three-dimensional face model is generated by the cylindrical coordinates of three-dimensional face model, and adopts the method for discretize to realize resampling, and can obtain according to demand the texture image of different resolution.
Detailed process is as follows:
Make v
j=(x
j, y
j, z
j)
TBe any one summit on three-dimensional face, its corresponding color value is r
j=(r
j, g
j, b
j)
T, wherein, 1≤j≤n, n are the summit numbers; And c
j=(h
j, θ
j, d
j)
TCylindrical coordinates corresponding to this point, wherein, h
j, θ
j, d
jRepresent respectively height, angle and radius.We can calculate c by following formula
jEach component value:
h
j＝y
j，
Wherein, d
j〉=0 ,-π≤θ
j≤ π.
Size is that the digital picture of M * N (M, N are respectively the height and width of image, can arrange voluntarily as required) can represent with f (x, y).And then we represent texture image with f (θ, h), obviously f (θ
i, h
i)=r
iBecause θ and h are successive values, also need it is sampled into the discrete picture of M * N size.Represent that with u and w according to equivalent discrete method, f (θ, h) being carried out uniform sampling can get with θ and h after resampling:
Wherein, 1≤i≤M, 1≤t≤N, R are the functions that rounds up, max, min get respectively maximal value and minimum value.Thereby we obtain the texture image after discretize: f (u
i, w
i)=r
i
1.2 the AAMs model of unique point on the training texture image
This module is a necessary links of the present invention, but is not original creation of the present invention, and we only set forth its operating process.Select 100 left and right three-dimensional face models to form training set, the method with 1.1 is carried out planar development with it, and manually demarcates 34 unique points on texture image, then adopts the AAMs algorithm that training set is trained, and generates the two-dimension human face shape.
1.3 create the planar triangulations template
The planar triangulations template is combined by average texture image and average two-dimensional shapes.
(1) at first, according to the texture image of everyone face in 1.1 method generating three-dimensional face database;
(2) secondly, generate the average texture image.The average texture image is not simply to average, each pixel on the average image is voted by all texture images that participate in training, if the texture image number of pixel is arranged greater than more than 1/2 on this position, keep this pixel (color is set to all texture images in the average color of this point) on the average texture image, otherwise for being set to sky (representing with white);
(3) adopt AAMs to carry out positioning feature point to the texture image of all models, obtain 34 unique points of everyone face, (two-dimensional shapes of i model is used to be referred to as two-dimensional shapes
Expression), calculate all faceforms' average two-dimensional shapes according to following formula
Four summits (upper left corner, the upper right corner, the lower left corner, the lower right corner all 34 points (seeing accompanying drawing 3 (a)) on average two-dimensional shapes being added epigraph, see accompanying drawing 1) as point set, adopt the Delauney algorithm to carry out triangulation to the average texture image, obtain the result of accompanying drawing 3 (b).All triangle sets after subdivision have just been generated complete people's face plane template altogether, this planar triangulations template definition number and the topological structure that resamples, also defined by this planar triangulations template the precision (being the pixel number of each triangle burst) that in the standardisation process, the three-dimensional face texture image resamples.Accompanying drawing 3 (c, d) is plane template and local detailed maps, and wherein coloured pixel represents effective summit (except the blue line on the edge) on template.
Step 2: the three-dimensional model based on the planar triangulations template resamples
Each point on two-dimensional shapes has physical meaning (as canthus, the corners of the mouth etc.), the summit on two-dimensional texture map and three-dimensional model to have one-to-one relationship, so we can obtain 34 three-dimensional vertices that unique point is corresponding on each three-dimensional model.The plane template that we can obtain by step 1, the resolution of each triangular plate on each input people face texture maps according to corresponding triangular plate on plane template is resampled, and each triangular plate in each faceform's 2 d texture image resampling process on the plane template is as unit.According to the one-to-one relationship between 2 d texture image and three-dimensional point cloud, the three-dimensional face model after can being resampled.Owing to having adopted identical template, each adopts the three-dimensional face model of same operation to have identical summit number; On the other hand, owing to being take triangular plate as unit, each triangular plate is at the ad-hoc location of face, so this method has realized the coupling based on unique point.After this step, each three-dimensional face model has identical summit number and has realized alignment based on unique point.
Step 3: the 3D grid based on plane template is rebuild
The summit that obtains after resampling is at random, need to organize generating three-dimensional models to these summits by setting up grid.Template has defined the topological structure between the summit, carries out in the following manner mesh reconstruction: at first the annexation according to pixel on horizontal direction on plane template and vertical direction is linked in sequence the summit in three dimensions; Next connect the lower left corner and to the diagonal line in the upper right corner, each little rectangle be divided into two triangular plates, and adopt from left to right, mode from top to bottom will be numbered summit and little triangle; At last, with all piecemeal combinations, generate the three-dimensional face model of ultimate criterion, as shown in Figure 4.
Claims (4)
1. three-dimensional face automatic standardizing method based on the planar triangulations template, planar triangulations template synthetic according to three-dimensional face model texture stretch-out view, and then realize all three-dimensional models standardization efficiently and accurately in the three-dimensional face storehouse, comprise following steps:
(1) automatically create the planar triangulations template
1.1 generating three-dimensional faceform's planar grains image;
1.2 the AAMs model active appearance models of unique point on the training texture image;
1.3 create the planar triangulations template;
(2) three-dimensional model based on the planar triangulations template resamples
The plane template that adopts (1) to obtain resamples to the resolution of each triangular plate on each input people face texture maps according to corresponding triangular plate on plane template.According to the one-to-one relationship between 2 d texture image and three-dimensional point cloud, the three-dimensional face model after being resampled makes different three-dimensional faces have identical number of vertex and topological structure after standardization;
(3) rebuild based on the 3D grid of plane template
At first the annexation according to pixel on horizontal direction on plane template and vertical direction is linked in sequence the summit in three dimensions; Next connect the lower left corner and to the diagonal line in the upper right corner, each little rectangle be divided into two triangular plates, and adopt from left to right, mode from top to bottom will be numbered summit and little triangle; At last, with all piecemeal combinations, generate the three-dimensional face model of ultimate criterion.
2. according to claim 1 the three-dimensional face automatic standardizing method based on the planar triangulations template, it is characterized in that: in described step (1), the texture image of three-dimensional face model is generated by the cylindrical coordinates of three-dimensional face model, and adopt the method for discretize to realize resampling, and can obtain according to demand the texture image of different resolution.
3. according to claim 1 the three-dimensional face automatic standardizing method based on the planar triangulations template, it is characterized in that: in described step (1), the planar triangulations template is combined by average texture image and average two-dimensional shapes; Wherein, if each pixel on the average texture image is produced by all texture image ballots in training set---the texture image number of value is arranged greater than more than 1/2 on this position, keep this pixel on the average texture image; Average two-dimensional shapes is determined by the mean value of 34 characteristic point positions on all texture images in training set; Four angle points that all 34 points on average two-dimensional shapes added epigraph, that is: the upper left corner, the upper right corner, the lower left corner, the lower right corner are as point set, the average texture image is carried out triangulation, all triangle sets after subdivision are generated complete people's face plane template altogether, by the precision of three-dimensional face texture image resampling in this planar triangulations template definition standardisation process.
4. according to claim 1 the three-dimensional face automatic standardizing method based on the planar triangulations template, it is characterized in that: in described step (2), each triangular plate in each faceform's 2 d texture image resampling process on the plane template is as unit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013100296453A CN103093498A (en) | 2013-01-25 | 2013-01-25 | Three-dimensional human face automatic standardization method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN2013100296453A CN103093498A (en) | 2013-01-25 | 2013-01-25 | Three-dimensional human face automatic standardization method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN103093498A true CN103093498A (en) | 2013-05-08 |
Family
ID=48206029
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN2013100296453A Pending CN103093498A (en) | 2013-01-25 | 2013-01-25 | Three-dimensional human face automatic standardization method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103093498A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103646416A (en) * | 2013-12-18 | 2014-03-19 | 中国科学院计算技术研究所 | Three-dimensional cartoon face texture generation method and device |
| CN104599325A (en) * | 2013-10-30 | 2015-05-06 | 中国科学院苏州纳米技术与纳米仿生研究所 | Three-dimensional face resampling method |
| CN104978764A (en) * | 2014-04-10 | 2015-10-14 | 华为技术有限公司 | Three-dimensional face mesh model processing method and three-dimensional face mesh model processing equipment |
| CN106462738A (en) * | 2014-05-20 | 2017-02-22 | 埃西勒国际通用光学公司 | Method for constructing a model of the face of a person, method and device for posture analysis using such a model |
| CN106919899A (en) * | 2017-01-18 | 2017-07-04 | 北京光年无限科技有限公司 | The method and system for imitating human face expression output based on intelligent robot |
| CN107123085A (en) * | 2017-03-22 | 2017-09-01 | 华南理工大学 | A kind of point cloud ratio standard method based on triangulation |
| CN107507127A (en) * | 2017-08-04 | 2017-12-22 | 深圳市易尚展示股份有限公司 | The global registration method and system of multiple views three-dimensional point cloud |
| CN107729806A (en) * | 2017-09-05 | 2018-02-23 | 西安理工大学 | Single-view Pose-varied face recognition method based on three-dimensional facial reconstruction |
| CN108510558A (en) * | 2017-02-28 | 2018-09-07 | 上海小桁网络科技有限公司 | Compression method, device and the terminal of point cloud data |
| CN109844818A (en) * | 2016-05-27 | 2019-06-04 | 米米听力科技有限公司 | For establishing the method and associated system of the deformable 3d model of element |
| CN109992809A (en) * | 2017-12-29 | 2019-07-09 | 深圳市优必选科技有限公司 | A kind of construction method of buildings model, device and storage device |
| CN110533777A (en) * | 2019-08-01 | 2019-12-03 | 北京达佳互联信息技术有限公司 | Three-dimensional face images modification method, device, electronic equipment and storage medium |
| CN111488853A (en) * | 2020-04-23 | 2020-08-04 | 杨九妹 | Big data face recognition method and system for financial institution security system and robot |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101866497A (en) * | 2010-06-18 | 2010-10-20 | 北京交通大学 | Binocular stereo vision based intelligent three-dimensional human face rebuilding method and system |
-
2013
- 2013-01-25 CN CN2013100296453A patent/CN103093498A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101866497A (en) * | 2010-06-18 | 2010-10-20 | 北京交通大学 | Binocular stereo vision based intelligent three-dimensional human face rebuilding method and system |
Non-Patent Citations (1)
| Title |
|---|
| LI XINXIN ET AL: "《Buiding a Standard 3D Face Database Based on Planar Templates》", 《3RD INTERNATIONAL CONFERENCE ON MEASURING TECHNOLOGY AND MECHATRONICS AUTOMATION》, vol. 4849, 6 January 2011 (2011-01-06), pages 280 - 283 * |
Cited By (23)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104599325A (en) * | 2013-10-30 | 2015-05-06 | 中国科学院苏州纳米技术与纳米仿生研究所 | Three-dimensional face resampling method |
| CN104599325B (en) * | 2013-10-30 | 2018-03-20 | 中国科学院苏州纳米技术与纳米仿生研究所 | A kind of three-dimensional face method for resampling |
| CN103646416A (en) * | 2013-12-18 | 2014-03-19 | 中国科学院计算技术研究所 | Three-dimensional cartoon face texture generation method and device |
| CN104978764B (en) * | 2014-04-10 | 2017-11-17 | 华为技术有限公司 | 3 d human face mesh model processing method and equipment |
| CN104978764A (en) * | 2014-04-10 | 2015-10-14 | 华为技术有限公司 | Three-dimensional face mesh model processing method and three-dimensional face mesh model processing equipment |
| CN106462738A (en) * | 2014-05-20 | 2017-02-22 | 埃西勒国际通用光学公司 | Method for constructing a model of the face of a person, method and device for posture analysis using such a model |
| US10380411B2 (en) | 2014-05-20 | 2019-08-13 | Essilor International | Method for constructing a model of the face of a person, method and device for posture analysis using such a model |
| CN106462738B (en) * | 2014-05-20 | 2020-10-09 | 依视路国际公司 | Method for constructing a model of a person's face, method and apparatus for analyzing a pose using such a model |
| CN109844818A (en) * | 2016-05-27 | 2019-06-04 | 米米听力科技有限公司 | For establishing the method and associated system of the deformable 3d model of element |
| CN106919899A (en) * | 2017-01-18 | 2017-07-04 | 北京光年无限科技有限公司 | The method and system for imitating human face expression output based on intelligent robot |
| CN106919899B (en) * | 2017-01-18 | 2020-07-28 | 北京光年无限科技有限公司 | Method and system for simulating facial expression output based on intelligent robot |
| CN108510558A (en) * | 2017-02-28 | 2018-09-07 | 上海小桁网络科技有限公司 | Compression method, device and the terminal of point cloud data |
| CN108510558B (en) * | 2017-02-28 | 2019-08-16 | 贝壳找房(北京)科技有限公司 | Compression method, device and the terminal of point cloud data |
| CN107123085A (en) * | 2017-03-22 | 2017-09-01 | 华南理工大学 | A kind of point cloud ratio standard method based on triangulation |
| CN107123085B (en) * | 2017-03-22 | 2019-10-18 | 华南理工大学 | A kind of point cloud ratio standard method based on triangulation |
| CN107507127A (en) * | 2017-08-04 | 2017-12-22 | 深圳市易尚展示股份有限公司 | The global registration method and system of multiple views three-dimensional point cloud |
| CN107507127B (en) * | 2017-08-04 | 2021-01-22 | 深圳市易尚展示股份有限公司 | Global matching method and system for multi-viewpoint three-dimensional point cloud |
| CN107729806A (en) * | 2017-09-05 | 2018-02-23 | 西安理工大学 | Single-view Pose-varied face recognition method based on three-dimensional facial reconstruction |
| CN109992809A (en) * | 2017-12-29 | 2019-07-09 | 深圳市优必选科技有限公司 | A kind of construction method of buildings model, device and storage device |
| CN109992809B (en) * | 2017-12-29 | 2023-03-10 | 深圳市优必选科技有限公司 | Building model construction method and device and storage device |
| CN110533777A (en) * | 2019-08-01 | 2019-12-03 | 北京达佳互联信息技术有限公司 | Three-dimensional face images modification method, device, electronic equipment and storage medium |
| CN111488853A (en) * | 2020-04-23 | 2020-08-04 | 杨九妹 | Big data face recognition method and system for financial institution security system and robot |
| CN111488853B (en) * | 2020-04-23 | 2020-12-11 | 中信百信银行股份有限公司 | Big data face recognition method and system for financial institution security system and robot |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN103093498A (en) | Three-dimensional human face automatic standardization method | |
| CN104330074B (en) | Intelligent surveying and mapping platform and realizing method thereof | |
| CN102426712B (en) | Three-dimensional head modeling method based on two images | |
| CN109003325A (en) | A kind of method of three-dimensional reconstruction, medium, device and calculate equipment | |
| CN106803267A (en) | Indoor scene three-dimensional rebuilding method based on Kinect | |
| CN108171780A (en) | A kind of method that indoor true three-dimension map is built based on laser radar | |
| CN104574432B (en) | Three-dimensional face reconstruction method and three-dimensional face reconstruction system for automatic multi-view-angle face auto-shooting image | |
| CN104966317B (en) | A kind of three-dimensional method for automatic modeling based on ore body contour line | |
| CN106067190A (en) | A kind of fast face threedimensional model based on single image generates and alternative approach | |
| CN107895048B (en) | Rapid drawing method based on live-action three-dimension | |
| CN104809940B (en) | Geometry stereographic projection device and projecting method | |
| CN107945285B (en) | Three-dimensional model map changing and deforming method | |
| CN108986221A (en) | A kind of three-dimensional face grid texture method lack of standardization approached based on template face | |
| CN103593869B (en) | A kind of scanning device and method for displaying image thereof | |
| CN102117488B (en) | Method for constructing surface reflection characteristics of real object in virtual environment | |
| CN104751511A (en) | 3D scene construction method and device | |
| CN104537705A (en) | Augmented reality based mobile platform three-dimensional biomolecule display system and method | |
| CN107729806A (en) | Single-view Pose-varied face recognition method based on three-dimensional facial reconstruction | |
| CN106500626A (en) | A kind of mobile phone stereoscopic imaging method and three-dimensional imaging mobile phone | |
| CN102411794A (en) | Output method of two-dimensional (2D) projection of three-dimensional (3D) model based on spherical harmonic transform | |
| CN103544733B (en) | The three-dimensional human head triangular mesh model method for building up analyzed based on Statistical Shape | |
| Trenchev et al. | Mathematical approaches for creation of complex 3D models and their application in education | |
| CN105243137A (en) | Draft-based three-dimensional model retrieval viewpoint selection method | |
| Deng et al. | Sketch2PQ: freeform planar quadrilateral mesh design via a single sketch | |
| CN109598785B (en) | Three-dimensional grid model view conversion method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C02 | Deemed withdrawal of patent application after publication (patent law 2001) | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20130508 |