CN108537199A - Based on the 3D facial image correction gain apparatus rebuild and method - Google Patents
Based on the 3D facial image correction gain apparatus rebuild and method Download PDFInfo
- Publication number
- CN108537199A CN108537199A CN201810349228.XA CN201810349228A CN108537199A CN 108537199 A CN108537199 A CN 108537199A CN 201810349228 A CN201810349228 A CN 201810349228A CN 108537199 A CN108537199 A CN 108537199A
- Authority
- CN
- China
- Prior art keywords
- module
- face
- facial image
- characteristic point
- model
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/161—Detection; Localisation; Normalisation
- G06V40/165—Detection; Localisation; Normalisation using facial parts and geometric relationships
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
- G06V40/171—Local features and components; Facial parts ; Occluding parts, e.g. glasses; Geometrical relationships
Abstract
The present invention discloses a kind of facial image correction gain apparatus and method rebuild based on 3D, and the facial image that pending 3D is rebuild is inputted by facial image input module, and characteristic point detection is carried out to the facial image for carrying out 3D reconstructions by characteristic point detection module;Module being adjusted by face, position adjustment being carried out to model face according to face characteristic point, adjust module by profile carries out position adjustment according to contour feature point to model silhouette;By Iterative matching module to after adjustment profile and characteristic point be iterated matching, and the appearance that module builds according to the contour feature point of face characteristic point and Iterative matching after adjustment face regression model is built by regression model;Module being baked by textures, textures baking being carried out to image according to faceform and characteristic point, module matching textures are rebuild by faceform and complete the three-dimensional reconstruction for returning face with regression model.The present invention is conducive to improve the robustness and recognition correct rate of entire face identification system training pattern.
Description
Technical field
The present invention relates to technical field of image processing, and in particular to a kind of facial image correction gain rebuild based on 3D
Device and method.
Background technology
Recognition of face is a kind of biological identification technology that the facial feature information based on people carries out identification, by adopting
Collect image or video flowing containing face, and detect and track face in the picture, and then the face to detecting carries out face
Identification, usually also referred to as Identification of Images, face recognition.Currently, in face recognition application scene, often will appear to various
The facial image of direction deflection, can be by training face image set and test facial image position restriction by face correction
In certain angle and range, to improve the accuracy rate of entire face identification system.
The prior art is the characteristic point that is got by facial image critical point detection to carry out on 2d flat
Shifting, scaling and affine transformation, to realize the correction of facial image.This method has the following defects:
1) facial image of other angles is unable to get by two dimensional image, due to human face data collection Partial angle face
The missing of image, it will influence the robustness of entire face identification system training pattern;
2) fixed face key point position can only be accomplished by two dimensional image, can not really solves pitch angle, deflection angle institute
Information on the image brought loses problem, to influence the recognition correct rate of face identification system.
Invention content
The purpose of the present invention is to provide a kind of facial image correction gain apparatus rebuild based on 3D and methods, pass through
Two dimensional image and characteristic point complete the reconstruction of face 3-D view;Facial image is carried out by the face 3-D view of reconstruction
Correction and human face data collection gain, be conducive to the robustness and the identification that improve entire face identification system training pattern
Accuracy.
To achieve the above object, the technical scheme is that:Gain apparatus is corrected based on the facial image that 3D is rebuild,
Described device includes facial image input module, characteristic point detection module, face adjustment module, profile adjustment module, iteration
With module, regression model structure module, textures bake module and faceform rebuilds module;The facial image input module
It establishes a connection with the characteristic point detection module, facial image input module is used to input the face that pending 3D is rebuild
Image;The characteristic point detection module adjusts module with the face and profile adjustment module establishes a connection, the spy
Sign point detection module is used to carry out characteristic point detection to the facial image for carrying out 3D reconstructions;The face adjustment module is used for root
Position adjustment is carried out to model face according to face characteristic point;The profile adjustment module is used for according to contour feature point to model
Profile carries out position adjustment;The Iterative matching module establishes a connection with profile adjustment module, Iterative matching module
For to after adjustment profile and characteristic point be iterated matching increase algorithm robustness;The regression model builds module
Module is adjusted with the face and Iterative matching module establishes a connection, after regression model structure module is used for according to adjustment
Face characteristic point and Iterative matching contour feature point structure face regression model appearance;The textures bake module with
The regression model structure module and characteristic point detection module establish a connection, and textures bake module and are used for according to face mould
Type carries out textures baking with characteristic point to image;The faceform rebuilds module and regression model structure module and patch
Figure bakes module and establishes a connection, and faceform's reconstruction module completes three-dimensional recurrence people for matching textures and regression model
The reconstruction of face.
Gain apparatus is corrected based on the facial image that 3D is rebuild as described above, described device further includes feature point coordinates
Searching module, the feature point coordinates searching module rebuilds module with the faceform and characteristic point detection module is established and connected
Relationship is connect, feature point coordinates searching module is used to search the corresponding model of characteristic point on faceform in three-dimensional return of reconstruction
The coordinate on surface.The corresponding model of characteristic point is found on the three-dimensional face model of reconstruction by feature point coordinates searching module
The coordinate on surface.Characteristic point refers to the point that those are at contour edge or face, can influence the outer of three-dimensional face model
It sees.
Gain apparatus is corrected based on the facial image that 3D is rebuild as described above, described device further includes model rendering mould
Block, the model rendering module establish a connection with the feature point coordinates searching module, and model rendering module is used for and right
Three-dimensional returns faceform and renders.Faceform is returned by model rendering module to three-dimensional to render, it can in rendering
To simulate light, accessories or rotation application scenarios.
Gain apparatus is corrected based on the facial image that 3D is rebuild as described above, described device further includes that facial image is defeated
Go out module, the facial image output module establishes a connection with the model rendering module, and facial image output module is used
Correction gain image after exporting rendering.
The present invention also provides a kind of facial images rebuild based on 3D to correct gain method, and the method uses above-mentioned dress
Realization is set, the method includes:
Step 1:The facial image that pending 3D is rebuild is inputted by facial image input module, is detected by characteristic point
Module carries out characteristic point detection to the facial image for carrying out 3D reconstructions;
Step 2:Module is adjusted by face, position adjustment is carried out to model face according to face characteristic point, pass through profile
It adjusts module and position adjustment is carried out to model silhouette according to contour feature point;
Step 3:By Iterative matching module to after adjustment profile and characteristic point be iterated matching, increase algorithm
Robustness, and contour feature point structure of the module according to face characteristic point and Iterative matching after adjustment is built by regression model
Build the appearance of face regression model;
Step 4:Module is baked by textures, textures baking is carried out to image according to faceform and characteristic point, pass through people
Face Model Reconstruction module matches textures and completes the three-dimensional reconstruction for returning face with regression model.
Gain method is corrected based on the facial image that 3D is rebuild as described above, the facial image correction rebuild based on 3D
Gain method further includes step 5, and faceform is returned in the three-dimensional of reconstruction by feature point coordinates searching module in step 5
The upper coordinate for searching the corresponding model surface of characteristic point.
Gain method is corrected based on the facial image that 3D is rebuild as described above, further includes passing through mould in the step 5
Type rendering module returns faceform to three-dimensional and renders, and light, accessories or rotation application scenarios are simulated in rendering.
Gain method is corrected based on the facial image that 3D is rebuild as described above, the facial image correction rebuild based on 3D
Gain method further includes step 6, exports the correction gain image after mould output renders in step 6 by facial image.
The invention has the advantages that:The facial image of other angles is obtained by the reconstruction of three-dimensional face images, is supplemented
The missing of human face data collection Partial angle facial image, improves the robustness of entire face identification system training pattern;Pass through three
The reconstruction of dimension face figure solves the problems, such as that the information on image caused by pitch angle, deflection angle is lost, by computer vision technique
And computer graphics techniques effectively increase the robustness of entire face identification system and identify correct
Rate.
Description of the drawings
Fig. 1 is that the facial image rebuild based on 3D corrects gain apparatus schematic diagram;
Fig. 2 is that the facial image rebuild based on 3D corrects gain method flow chart;
Fig. 3 is that the facial image rebuild based on 3D corrects gain method processing step schematic diagram.
Specific implementation mode
The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention..
As shown in Figure 1, correcting gain apparatus based on the facial image that 3D is rebuild, described device includes facial image input
Module 1, characteristic point detection module 2, face adjustment module 3, profile adjustment module 4, Iterative matching module 5, regression model structure
Module 6, textures bake module 7 and faceform rebuilds module 8;The facial image input module 1 is detected with the characteristic point
Module 2 establishes a connection, and facial image input module 1 is used to input the facial image that pending 3D is rebuild;The feature
Point detection module 2 adjusts module 3 with the face and profile adjustment module 4 establishes a connection, the characteristic point detection module
2 to the facial image for carrying out 3D reconstructions for carrying out characteristic point detection;The face adjustment module 3 is used for according to face feature
Point carries out position adjustment to model face;Profile adjustment module 4 is used for according to contour feature point to model silhouette into line position
Set adjustment;The Iterative matching module 5 and the profile adjustment module 4 establish a connection, Iterative matching module 5 for pair
Profile and characteristic point after adjustment are iterated the robustness that matching increases algorithm;Regression model structure module 6 with it is described
Face adjust module 3 and Iterative matching module 5 establishes a connection, and regression model builds module 6 and is used for according to five after adjustment
The contour feature point of official's characteristic point and Iterative matching builds the appearance of face regression model;The textures bake module 7 with it is described
Regression model builds module 6 and characteristic point detection module 2 establishes a connection, and textures bake module 7 and are used for according to face mould
Type carries out textures baking with characteristic point to image;The faceform rebuild module 8 and regression model structure module 6 and
Textures bake module 7 and establish a connection, and faceform rebuilds module 8 and completes three-dimensional return for matching textures and regression model
Classify the reconstruction of face.
In the one embodiment for correcting gain apparatus based on the facial image that 3D is rebuild, described device further includes characteristic point
Coordinate searching module 9, the feature point coordinates searching module 9 rebuild module 8 and characteristic point detection module with the faceform
2 establish a connection, and feature point coordinates searching module 9 is used to search characteristic point pair on faceform in three-dimensional return of reconstruction
The coordinate for the model surface answered.By feature point coordinates searching module 9 characteristic point pair is found on the three-dimensional face model of reconstruction
The coordinate for the model surface answered.Characteristic point refers to the point that those are at contour edge or face, can influence three-dimensional face
The appearance of model.
In the one embodiment for correcting gain apparatus based on the facial image that 3D is rebuild, described device further includes model wash with watercolours
Module 10 is contaminated, the model rendering module 10 establishes a connection with the feature point coordinates searching module 9, model rendering mould
Block 10 is used to and returns faceform to three-dimensional render.By model rendering module 10 to three-dimensional return faceform into
Row renders, and light, accessories or rotation application scenarios can be simulated in rendering.
In the one embodiment for correcting gain apparatus based on the facial image that 3D is rebuild, described device further includes face figure
As output module 11, the facial image output module 11 establishes a connection with the model rendering module 10, facial image
Output module 11 is used to export the correction gain image after rendering.
Fig. 2 and Fig. 3 is participated in, the present invention also provides a kind of facial images rebuild based on 3D to correct gain method, the side
Method using above-mentioned apparatus realize, the method includes:
S1:The facial image that pending 3D is rebuild is inputted by facial image input module 1, mould is detected by characteristic point
Block 2 carries out characteristic point detection to the facial image for carrying out 3D reconstructions;
S2:Module 3 is adjusted by face, position adjustment is carried out to model face according to face characteristic point, pass through profile tune
Mould preparation block 4 carries out position adjustment according to contour feature point to model silhouette;
S3:By Iterative matching module 5 to after adjustment profile and characteristic point be iterated matching, increase the Shandong of algorithm
Stick, and module 6 is built by regression model and is built according to the contour feature point of face characteristic point and Iterative matching after adjustment
The appearance of face regression model;
S4:Module 7 is baked by textures, textures baking is carried out to image according to faceform and characteristic point, pass through face
Model Reconstruction module 8 matches textures and completes the three-dimensional reconstruction for returning face with regression model.
It is to convert by the form of picture the time relationship between model and model that textures, which are baked and banked up with earth, thus shape
At a kind of textures, by the control of this textures on model, can obtain a kind of vacation but very true effect, textures are baked and banked up with earth,
Textures are found, occ or ao textures convert textures, specular map, and Diffuse Color etc. belongs to the prior art.Briefly
It is a kind of mode for max Lighting informations being rendered to textures, then this textures after baking is pasted again and is returned in scene,
Such words Lighting information becomes textures, does not need CPU and goes time-consuming calculating again, as long as common textures, institute
It is exceedingly fast with speed.
Further include S5 in the one embodiment for correcting gain method based on the facial image that 3D is rebuild:It is sat by characteristic point
Mark the coordinate that searching module 9 searches the corresponding model surface of characteristic point on the three-dimensional recurrence faceform of reconstruction.
Further include passing through model wash with watercolours in S5 in the one embodiment for correcting gain method based on the facial image that 3D is rebuild
Dye module 10 returns faceform to three-dimensional and renders, and light, accessories or rotation application scenarios are simulated in rendering.
Further include S6 in the one embodiment for correcting gain method based on the facial image that 3D is rebuild:Pass through facial image
Export the correction gain image after mould output renders.
The present invention is based on the detections that two-dimensional facial image carries out characteristic point, utilize face characteristic point and contour feature
Point carries out position adjustment to the face and profile of face regression model, and profile and the characteristic point after adjustment are iterated
Match, increases algorithm robustness, and complete the appearance of face regression model, image is pasted with characteristic point using faceform
Figure bakes, and matching textures complete the three-dimensional reconstruction for returning face with regression model, and characteristic point is found on the threedimensional model of reconstruction
The coordinate of corresponding model surface, and model is rendered.Light, accessories, the different applications such as rotation are simulated in rendering
Scene.It exports corresponding picture and completes the gain of human face data collection and the correction of facial image, to pass through three-dimensional face figure
The reconstruction of picture obtains the facial image of other angles, supplements the missing of human face data collection Partial angle facial image, improves entire
The robustness of face identification system training pattern;It solves to scheme caused by pitch angle, deflection angle by the reconstruction of three-dimensional face figure
As upper information loss problem, computer vision technique and computer graphics techniques are effectively increased whole
The robustness and recognition correct rate of a face identification system.
Although above having used general explanation and specific embodiment, the present invention is described in detail, at this
On the basis of invention, it can be made some modifications or improvements, this will be apparent to those skilled in the art.Therefore,
These modifications or improvements without departing from theon the basis of the spirit of the present invention belong to the scope of protection of present invention.
Claims (8)
1. correcting gain apparatus based on the facial image that 3D is rebuild, it is characterised in that:Described device includes facial image input mould
Block, characteristic point detection module, face adjustment module, profile adjustment module, Iterative matching module, regression model structure module, patch
Figure bakes module and faceform rebuilds module;The facial image input module is established with the characteristic point detection module and is connected
Relationship, facial image input module are used to input the facial image that pending 3D is rebuild;The characteristic point detection module with it is described
Face adjust module and profile adjustment module establishes a connection, and the characteristic point detection module is used for the people to carrying out 3D reconstructions
Face image carries out characteristic point detection;The face adjustment module is used to carry out position tune to model face according to face characteristic point
It is whole;The profile adjustment module is used to carry out position adjustment to model silhouette according to contour feature point;The Iterative matching module
With the profile adjustment module establish a connection, Iterative matching module be used for after adjustment profile and characteristic point be iterated
Matching increases the robustness of algorithm;The regression model structure module is established with face adjustment module and Iterative matching module
Connection relation, regression model build module and are used to be built according to the contour feature point of face characteristic point and Iterative matching after adjustment
The appearance of face regression model;The textures bake module and are established with regression model structure module and characteristic point detection module
Connection relation, textures bake module and are used to carry out textures baking to image according to faceform and characteristic point;The faceform
It rebuilds module to establish a connection with regression model structure module and textures baking module, faceform rebuilds module and is used for
It matches textures and completes the three-dimensional reconstruction for returning face with regression model.
2. the facial image according to claim 1 rebuild based on 3D corrects gain apparatus, it is characterised in that:Described device
Further include feature point coordinates searching module, the feature point coordinates searching module rebuilds module and characteristic point with the faceform
Detection module establishes a connection, and feature point coordinates searching module is used to search feature on faceform in three-dimensional return of reconstruction
The coordinate of the corresponding model surface of point.
3. the facial image according to claim 2 rebuild based on 3D corrects gain apparatus, it is characterised in that:Described device
Further include model rendering module, the model rendering module establishes a connection with the feature point coordinates searching module, model
Rendering module is used to and returns faceform to three-dimensional render.
4. the facial image according to claim 3 rebuild based on 3D corrects gain apparatus, it is characterised in that:Described device
Further include facial image output module, the facial image output module establishes a connection with the model rendering module, people
Face image output module is used to export the correction gain image after rendering.
5. correcting gain method based on the facial image that 3D is rebuild, the method is used and is filled as described in any one of Claims 1-4
It sets, it is characterised in that:The method includes:
Step 1:The facial image that pending 3D is rebuild is inputted by facial image input module, passes through characteristic point detection module
Facial image to carrying out 3D reconstructions carries out characteristic point detection;
Step 2:Module is adjusted by face, position adjustment is carried out to model face according to face characteristic point, adjusted by profile
Module carries out position adjustment according to contour feature point to model silhouette;
Step 3:By Iterative matching module to after adjustment profile and characteristic point be iterated matching, increase the robust of algorithm
Property, and module is built by regression model, face is built according to the contour feature point of face characteristic point and Iterative matching after adjustment
The appearance of regression model;
Step 4:Module is baked by textures, textures baking is carried out to image according to faceform and characteristic point, pass through face mould
Type rebuilds module matching textures and completes the three-dimensional reconstruction for returning face with regression model.
6. the facial image according to claim 5 rebuild based on 3D corrects gain method, it is characterised in that:Based on 3D weights
The facial image correction gain method built further includes step 5, by feature point coordinates searching module the three of reconstruction in step 5
Dimension returns the coordinate that the corresponding model surface of characteristic point is searched on faceform.
7. the facial image according to claim 6 rebuild based on 3D corrects gain method, it is characterised in that:The step
Further include that returning faceform to three-dimensional by model rendering module renders, and light, accessories or rotation are simulated in rendering in five
Turn application scenarios.
8. the facial image according to claim 6 rebuild based on 3D corrects gain method, it is characterised in that:Based on 3D weights
The facial image correction gain method built further includes step 6, after exporting mould output rendering by facial image in step 6
Correction gain image.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810349228.XA CN108537199A (en) | 2018-04-18 | 2018-04-18 | Based on the 3D facial image correction gain apparatus rebuild and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810349228.XA CN108537199A (en) | 2018-04-18 | 2018-04-18 | Based on the 3D facial image correction gain apparatus rebuild and method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108537199A true CN108537199A (en) | 2018-09-14 |
Family
ID=63477893
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810349228.XA Pending CN108537199A (en) | 2018-04-18 | 2018-04-18 | Based on the 3D facial image correction gain apparatus rebuild and method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108537199A (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101916384A (en) * | 2010-09-01 | 2010-12-15 | 汉王科技股份有限公司 | Facial image reconstruction method and device and face recognition system |
CN101968892A (en) * | 2009-07-28 | 2011-02-09 | 上海冰动信息技术有限公司 | Method for automatically adjusting three-dimensional face model according to one face picture |
CN104966316A (en) * | 2015-05-22 | 2015-10-07 | 腾讯科技(深圳)有限公司 | 3D face reconstruction method, apparatus and server |
CN106067190A (en) * | 2016-05-27 | 2016-11-02 | 俞怡斐 | A kind of fast face threedimensional model based on single image generates and alternative approach |
CN106599878A (en) * | 2016-12-28 | 2017-04-26 | 深圳市捷顺科技实业股份有限公司 | Face reconstruction correction method and device based on deep learning |
CN107358207A (en) * | 2017-07-14 | 2017-11-17 | 重庆大学 | A kind of method for correcting facial image |
-
2018
- 2018-04-18 CN CN201810349228.XA patent/CN108537199A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101968892A (en) * | 2009-07-28 | 2011-02-09 | 上海冰动信息技术有限公司 | Method for automatically adjusting three-dimensional face model according to one face picture |
CN101916384A (en) * | 2010-09-01 | 2010-12-15 | 汉王科技股份有限公司 | Facial image reconstruction method and device and face recognition system |
CN104966316A (en) * | 2015-05-22 | 2015-10-07 | 腾讯科技(深圳)有限公司 | 3D face reconstruction method, apparatus and server |
CN106067190A (en) * | 2016-05-27 | 2016-11-02 | 俞怡斐 | A kind of fast face threedimensional model based on single image generates and alternative approach |
CN106599878A (en) * | 2016-12-28 | 2017-04-26 | 深圳市捷顺科技实业股份有限公司 | Face reconstruction correction method and device based on deep learning |
CN107358207A (en) * | 2017-07-14 | 2017-11-17 | 重庆大学 | A kind of method for correcting facial image |
Non-Patent Citations (2)
Title |
---|
FENG LIU等: "Joint Face Alignment and 3D Face Reconstruction", 《HTTPS://LINK.SPRINGER.COM/CHAPTER/10.1007/978-3-319-46454-1_33》 * |
胡起云 等: "《三维建模基础》", 31 August 2017 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Niu et al. | Unsupervised saliency detection of rail surface defects using stereoscopic images | |
CN104992441B (en) | A kind of real human body three-dimensional modeling method towards individualized virtual fitting | |
CN101697233B (en) | Structured light-based three-dimensional object surface reconstruction method | |
CN102663820B (en) | Three-dimensional head model reconstruction method | |
CN104835144B (en) | The method for solving camera intrinsic parameter using the picture and orthogonality of the centre of sphere of a ball | |
CN104240289B (en) | Three-dimensional digitalization reconstruction method and system based on single camera | |
CN103839277B (en) | A kind of mobile augmented reality register method of outdoor largescale natural scene | |
CN106780592A (en) | Kinect depth reconstruction algorithms based on camera motion and image light and shade | |
CN104794722A (en) | Dressed human body three-dimensional bare body model calculation method through single Kinect | |
CN106910222A (en) | Face three-dimensional rebuilding method based on binocular stereo vision | |
CN106709947A (en) | RGBD camera-based three-dimensional human body rapid modeling system | |
CN101639947A (en) | Image-based plant three-dimensional shape measurement and reconstruction method and system | |
CN107240129A (en) | Object and indoor small scene based on RGB D camera datas recover and modeling method | |
CN106327532A (en) | Three-dimensional registering method for single image | |
CN106485690A (en) | Cloud data based on a feature and the autoregistration fusion method of optical image | |
CN105279787A (en) | Method for generating three-dimensional (3D) building model based on photographed house type image identification | |
CN107392947A (en) | 2D 3D rendering method for registering based on coplanar four point set of profile | |
CN102999942A (en) | Three-dimensional face reconstruction method | |
CN106991695A (en) | A kind of method for registering images and device | |
CN110189408B (en) | System and method for obtaining human body appearance data according to human body photos | |
CN108597009A (en) | A method of objective detection is carried out based on direction angle information | |
CN103438834A (en) | Hierarchy-type rapid three-dimensional measuring device and method based on structured light projection | |
CN108389260A (en) | A kind of three-dimensional rebuilding method based on Kinect sensor | |
Ran et al. | Point cloud registration method based on SIFT and geometry feature | |
CN106709870A (en) | Close-range image straight-line segment matching method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180914 |