CN107784688A - A kind of three-dimensional modeling method based on picture - Google Patents
A kind of three-dimensional modeling method based on picture Download PDFInfo
- Publication number
- CN107784688A CN107784688A CN201710967552.3A CN201710967552A CN107784688A CN 107784688 A CN107784688 A CN 107784688A CN 201710967552 A CN201710967552 A CN 201710967552A CN 107784688 A CN107784688 A CN 107784688A
- Authority
- CN
- China
- Prior art keywords
- synthesis
- threedimensional model
- mentioned
- synthesized
- region
- 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
- G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/50—Depth or shape recovery
- G06T7/529—Depth or shape recovery from texture
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/50—Depth or shape recovery
- G06T7/55—Depth or shape recovery from multiple images
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computer Graphics (AREA)
- Geometry (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of three-dimensional modeling method based on picture, comprise the following steps, obtain multiple picture groups, multiple picture groups obtain in the following way, the camera of one group of picture is obtained using by once shooting, subject is shot from different angles multiple, based on each picture group in the multiple picture groups got, multiple threedimensional models of generation modeling main body, the synthesis threedimensional model of above-mentioned selection is divided into multiple synthesis regions, they respectively specify that corresponding with above-mentioned multiple synthesis regions and above-mentioned any one multiple being synthesized in region that are synthesized in threedimensional model.The present invention is using the image modeling of multi-angled shooting into multiple threedimensional models, segmentation and the synthesis region for specifying multiple threedimensional models, the region of this segmentation is recombined by identical characteristic point, and the depth value of distinguished point based carries out coordinate transform to synthesis region, can model more accurate threedimensional model.
Description
Technical field
The present invention relates to dimensional Modeling Technology field, more particularly to a kind of three-dimensional modeling method based on picture.
Background technology
It is well known that due to the factor numerous and complicated being related to for modeling main body progress global design, design is fully relied on
Personnel carry out layout after considering all factors, require high and unusual labor intensive not only for designer, also
Major defect be present and can not use in the carelessness or the inconsiderate scheme for causing design for being likely due to designer.
In the past, there is following technology:Utilize camera shooting historical relic for possessing right eye image pickup part and left eye image pickup part etc.
Subject, the right eye picture and left eye picture (stereotome) that are obtained based on shooting create the threedimensional model of subject.But
It is simply bigger error be present with this to create the threedimensional model of subject, and such as light stream, the moon in picture is also present
The data such as shadow and texture can be used for the depth for judging object, so as to which the comprehensive threedimensional model for realizing structure will be more accurate
Really.
The content of the invention
Based on technical problem existing for background technology, the present invention proposes a kind of three-dimensional modeling method based on picture.
1st, a kind of three-dimensional modeling method based on picture proposed by the present invention, comprises the following steps:
S1:Multiple picture groups are obtained, multiple picture groups are obtained in the following way, and one group of figure is obtained using by once shooting
The camera of piece, subject is shot from different angles multiple;
S2:Based on each picture group in the multiple picture groups got, multiple threedimensional models of generation modeling main body;
S3:The synthesis threedimensional model of above-mentioned selection is divided into multiple synthesis regions, they are respectively specified that and above-mentioned multiple synthesis
Region and above-mentioned any one multiple being synthesized in region that are synthesized in threedimensional model are corresponding;
S4:Among synthesizing region, selection is synthesized the synthesis region of threedimensional model and synthesizes the conjunction for being synthesized threedimensional model
Into the synthesis region of threedimensional model;
S5:Extract multiple fisrt feature points from the above-mentioned synthesis region for being synthesized threedimensional model selected, and from above-mentioned
Multiple second feature points are extracted in the synthesis region for the synthesis threedimensional model selected;
S6:The multiple fisrt feature points and multiple second feature point gone out based on said extracted, are obtained the synthesis threedimensional model
Coordinate is transformed to the coordinate conversion parameter of the coordinate of the coordinate system for being synthesized threedimensional model, according to the multiple characteristic point with it is right
The depth value of similarity and the multiple characteristic point between the matching characteristic point answered;
S7:Using the above-mentioned coordinate conversion parameter and depth value got, the coordinate of above-mentioned synthesis threedimensional model is transformed to
The coordinate of the above-mentioned coordinate system for being synthesized threedimensional model, according to above step by the conjunction after above-mentioned synthesis threedimensional model coordinate transform
Into the synthesis region of threedimensional model;
S8:The synthesis region of synthesis threedimensional model after above-mentioned coordinate transform is synthesized into the above-mentioned synthesis for being synthesized threedimensional model
Region;
S9:The above-mentioned synthesis region for being synthesized threedimensional model is synthesized into synthesis threedimensional model.
Preferably, the camera should be same camera, and the height of camera camera shooting and camera shooting
The elevation angle of head should be consistent.
Preferably, the region that actively synthesizes includes with any one group of corresponding synthesis region in passive synthesis region
Identical characteristic point.
Preferably, the quantity of the fisrt feature point and second feature point is 3-10, and fisrt feature point and second
Characteristic point is not identical characteristic point.
Beneficial effects of the present invention are:The present invention, into multiple threedimensional models, is split using the image modeling of multi-angled shooting
Synthesis region with multiple threedimensional models are specified, is recombined by identical characteristic point to the region of this segmentation, and base
Coordinate transform is carried out to synthesis region in the depth value of characteristic point, more accurate threedimensional model can be modeled.
Brief description of the drawings
Fig. 1 is a kind of method flow diagram of the three-dimensional modeling method based on picture proposed by the present invention.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is carried out clear, complete
Site preparation describes, it is clear that described embodiment is only part of the embodiment of the present invention, rather than whole embodiments.
A kind of reference picture 1, three-dimensional modeling method based on picture, comprises the following steps:
Obtain data:As shown in figure 1, obtaining multiple picture groups by camera first, acquisition modes are as follows, led to using camera
Cross and once shoot to obtain one group of picture, subject is shot from different angles multiple;
Generation model:According to each picture group in the multiple picture groups got in above-mentioned, generation modeling main body it is multiple
Threedimensional model;
Dividing processing:The synthesis threedimensional model of above-mentioned selection is divided into multiple synthesis regions, they respectively specify that with it is above-mentioned more
Individual synthesis region and above-mentioned any one multiple being synthesized in region that are synthesized in threedimensional model are corresponding;
Designated area:Among above-mentioned synthesis region, selection is synthesized the synthesis region of threedimensional model and synthesizes this and is synthesized
The synthesis region of the synthesis threedimensional model of threedimensional model;
Obtain feature:Multiple fisrt feature points are extracted from the above-mentioned synthesis region for being synthesized threedimensional model selected, and
Multiple second feature points are extracted from the synthesis region of the above-mentioned synthesis threedimensional model selected;
Get parms:The multiple fisrt feature points and multiple second feature point gone out based on said extracted, obtain the synthesis is three-dimensional
The coordinate of model is transformed to the coordinate conversion parameter of the coordinate of the coordinate system for being synthesized threedimensional model, according to the multiple feature
Point and the similarity and the depth value of the multiple characteristic point between the corresponding matching characteristic point;
Coordinate transforming:Using the above-mentioned coordinate conversion parameter and depth value got, by the coordinate of above-mentioned synthesis threedimensional model
The coordinate of the above-mentioned coordinate system for being synthesized threedimensional model is transformed to, according to above step by above-mentioned synthesis threedimensional model coordinate transform
The synthesis region of synthesis threedimensional model afterwards.It is specific as follows:Similarity between characteristic point and matching characteristic point is according to calculating
Average color error value determine, average color error is bigger, then the similarity between characteristic point and matching characteristic point
Also it is lower.Using the similarity between characteristic point and matching characteristic point, characteristic point can be screened.According to the plurality of spy
Sign point and the similarity between corresponding matching characteristic point, determine drop threshold, then by the plurality of characteristic point, with corresponding
The characteristic point for being less than drop threshold with the similarity between characteristic point screens out.When between characteristic point and corresponding matching characteristic point
When similarity is less than drop threshold, show that now matching characteristic point is not the actual point with corresponding to characteristic point on second picture,
And judge that now Feature Points Matching fails, and these characteristic points that it fails to match and be not involved in follow-up three-dimensional modeling and calculate.Root
Will be unified into unified coordinate system by synthesis threedimensional model according to coordinate conversion parameter, synthesis threedimensional model synthesis region.
Synthesize region:The synthesis region of synthesis threedimensional model after above-mentioned coordinate transform is synthesized and above-mentioned is synthesized three-dimensional
The synthesis region of model;
Synthetic model:The above-mentioned synthesis region for being synthesized threedimensional model is synthesized into synthesis threedimensional model.
The foregoing is only a preferred embodiment of the present invention, but protection scope of the present invention be not limited thereto,
Any one skilled in the art the invention discloses technical scope in, technique according to the invention scheme and its
Inventive concept is subject to equivalent substitution or change, should all be included within the scope of the present invention.
Claims (4)
1. a kind of three-dimensional modeling method based on picture, it is characterised in that comprise the following steps:
S1:Multiple picture groups are obtained, multiple picture groups are obtained in the following way, and one group of figure is obtained using by once shooting
The camera of piece, subject is shot from different angles multiple;
S2:Based on each picture group in the multiple picture groups got, multiple threedimensional models of generation modeling main body;
S3:The synthesis threedimensional model of above-mentioned selection is divided into multiple synthesis regions, they are respectively specified that and above-mentioned multiple synthesis
Region and above-mentioned any one multiple being synthesized in region that are synthesized in threedimensional model are corresponding;
S4:Among synthesizing region, selection is synthesized the synthesis region of threedimensional model and synthesizes the conjunction for being synthesized threedimensional model
Into the synthesis region of threedimensional model;
S5:Extract multiple fisrt feature points from the above-mentioned synthesis region for being synthesized threedimensional model selected, and from above-mentioned
Multiple second feature points are extracted in the synthesis region for the synthesis threedimensional model selected;
S6:The multiple fisrt feature points and multiple second feature point gone out based on said extracted, are obtained the synthesis threedimensional model
Coordinate is transformed to the coordinate conversion parameter of the coordinate of the coordinate system for being synthesized threedimensional model, according to the multiple characteristic point with it is right
The depth value of similarity and the multiple characteristic point between the matching characteristic point answered;
S7:Using the above-mentioned coordinate conversion parameter and depth value got, the coordinate of above-mentioned synthesis threedimensional model is transformed to
The coordinate of the above-mentioned coordinate system for being synthesized threedimensional model, according to above step by the conjunction after above-mentioned synthesis threedimensional model coordinate transform
Into the synthesis region of threedimensional model;
S8:The synthesis region of synthesis threedimensional model after above-mentioned coordinate transform is synthesized into the above-mentioned synthesis for being synthesized threedimensional model
Region;
S9:The above-mentioned synthesis region for being synthesized threedimensional model is synthesized into synthesis threedimensional model.
2. a kind of three-dimensional modeling method based on picture according to claim 1, it is characterised in that the camera should be
Same camera, and the height of camera camera shooting should be consistent with the elevation angle of camera camera.
A kind of 3. three-dimensional modeling method based on picture according to claim 1, it is characterised in that the active synthesis zone
Domain includes identical characteristic point with any one group of corresponding synthesis region in passive synthesis region.
A kind of 4. three-dimensional modeling method based on picture according to claim 1, it is characterised in that the fisrt feature point
Quantity with second feature point is 3-10, and fisrt feature point and second feature point are not identical characteristic points.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710967552.3A CN107784688A (en) | 2017-10-17 | 2017-10-17 | A kind of three-dimensional modeling method based on picture |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710967552.3A CN107784688A (en) | 2017-10-17 | 2017-10-17 | A kind of three-dimensional modeling method based on picture |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107784688A true CN107784688A (en) | 2018-03-09 |
Family
ID=61434568
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710967552.3A Pending CN107784688A (en) | 2017-10-17 | 2017-10-17 | A kind of three-dimensional modeling method based on picture |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107784688A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108665530A (en) * | 2018-04-25 | 2018-10-16 | 厦门大学 | Three-dimensional modeling implementation method based on single picture |
CN108961374A (en) * | 2018-04-09 | 2018-12-07 | 武汉金运激光股份有限公司 | A kind of image display method and device with 3-D effect |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102208116A (en) * | 2010-03-29 | 2011-10-05 | 卡西欧计算机株式会社 | 3D modeling apparatus and 3D modeling method |
CN102236911A (en) * | 2010-03-17 | 2011-11-09 | 卡西欧计算机株式会社 | 3d modeling apparatus and 3d modeling method |
US20140071131A1 (en) * | 2012-09-13 | 2014-03-13 | Cannon Kabushiki Kaisha | Image processing apparatus, image processing method and program |
CN104346824A (en) * | 2013-08-09 | 2015-02-11 | 汉王科技股份有限公司 | Method and device for automatically synthesizing three-dimensional expression based on single facial image |
-
2017
- 2017-10-17 CN CN201710967552.3A patent/CN107784688A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102236911A (en) * | 2010-03-17 | 2011-11-09 | 卡西欧计算机株式会社 | 3d modeling apparatus and 3d modeling method |
CN102208116A (en) * | 2010-03-29 | 2011-10-05 | 卡西欧计算机株式会社 | 3D modeling apparatus and 3D modeling method |
US20140071131A1 (en) * | 2012-09-13 | 2014-03-13 | Cannon Kabushiki Kaisha | Image processing apparatus, image processing method and program |
CN104346824A (en) * | 2013-08-09 | 2015-02-11 | 汉王科技股份有限公司 | Method and device for automatically synthesizing three-dimensional expression based on single facial image |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108961374A (en) * | 2018-04-09 | 2018-12-07 | 武汉金运激光股份有限公司 | A kind of image display method and device with 3-D effect |
CN108665530A (en) * | 2018-04-25 | 2018-10-16 | 厦门大学 | Three-dimensional modeling implementation method based on single picture |
CN108665530B (en) * | 2018-04-25 | 2021-12-14 | 厦门大学 | Three-dimensional modeling implementation method based on single picture |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US10685454B2 (en) | Apparatus and method for generating synthetic training data for motion recognition | |
Alexiadis et al. | An integrated platform for live 3D human reconstruction and motion capturing | |
CN104361628B (en) | A kind of three-dimensional live modeling based on aviation oblique photograph measurement | |
US8624901B2 (en) | Apparatus and method for generating facial animation | |
CN103635935B (en) | 3d street | |
Grün et al. | Photogrammetric reconstruction of the great Buddha of Bamiyan, Afghanistan | |
CN104077804B (en) | A kind of method based on multi-frame video picture construction three-dimensional face model | |
CN104661010B (en) | Method and device for establishing three-dimensional model | |
CN105976426B (en) | A kind of quick three-dimensional atural object model building method | |
CN110189399A (en) | A kind of method and system that interior three-dimensional layout rebuilds | |
CN106611441B (en) | The treating method and apparatus of three-dimensional map | |
CN105096300B (en) | Method for checking object and equipment | |
US20190073796A1 (en) | Method and Image Processing System for Determining Parameters of a Camera | |
CN108648264A (en) | Underwater scene method for reconstructing based on exercise recovery and storage medium | |
CN110473248A (en) | A kind of measurement method using picture construction human 3d model | |
CN107784688A (en) | A kind of three-dimensional modeling method based on picture | |
KR102118937B1 (en) | Apparatus for Service of 3D Data and Driving Method Thereof, and Computer Readable Recording Medium | |
CN113284249B (en) | Multi-view three-dimensional human body reconstruction method and system based on graph neural network | |
JP6555755B2 (en) | Image processing apparatus, image processing method, and image processing program | |
JP2012113438A (en) | Posture estimation apparatus and posture estimation program | |
CN106604016A (en) | Stereoscopic video capture system | |
Qin et al. | A methodology for true orthorectification of large-scale urban aerial images and automatic detection of building occlusions using digital surface model | |
JP2005123823A (en) | Distance information attaching apparatus, attached video image generating apparatus, attached video information generating method, and distance information attaching program | |
Varanasi et al. | Temporally coherent segmentation of 3D reconstructions | |
Skuratovskyi et al. | Outdoor mapping framework: from images to 3d model |
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 |
Application publication date: 20180309 |
|
RJ01 | Rejection of invention patent application after publication |