CN108492355A - The three-dimensional object surface Variability Metric method transmitted based on line segment - Google Patents
The three-dimensional object surface Variability Metric method transmitted based on line segment Download PDFInfo
- Publication number
- CN108492355A CN108492355A CN201810192387.3A CN201810192387A CN108492355A CN 108492355 A CN108492355 A CN 108492355A CN 201810192387 A CN201810192387 A CN 201810192387A CN 108492355 A CN108492355 A CN 108492355A
- Authority
- CN
- China
- Prior art keywords
- line segment
- dimensional
- vertex
- posture
- dimensional object
- 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
- G06T15/00—3D [Three Dimensional] image rendering
- G06T15/08—Volume rendering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T19/00—Manipulating 3D models or images for computer graphics
- G06T19/20—Editing of 3D images, e.g. changing shapes or colours, aligning objects or positioning parts
Abstract
The invention discloses a kind of three-dimensional object surface Variability Metric methods transmitted based on line segment, belong to computer vision, field of artificial intelligence research.Main function is to carry out quantisation metric to three-dimensional object surface dynamically changeable ability, can be applied to object part division, modeling, description etc..Transmission by carrying out all connecting line segments in a variety of different posture three-dimensional triangulation shape grid datas of object is realized:Each connecting line segment will sequence through all training samples.By supervising the situation of change of connecting line segment in transmittance process, the microcosmic measurement to body surface changeability is obtained, and feed back to corresponding grid vertex, to obtain the dynamically changeable ability measurement as unit of vertex.The invention has the advantages that:It relies only on a small amount of static data and measurement to body surface dynamic variability can be realized;Realize have microcosmic high-precision measured capabilities based on body surface minimum description scheme;It is low that step simplifies operation strength, it can be ensured that the real-time implementation in larger three-dimensional data.
Description
Technical field
Along with the rapid development of three-dimension object technology for information acquisition and equipment in recent years, research for three-dimension object and
Using more and more common, three-dimension object information is increasingly becoming the most important research pair in the fields such as computer vision, pattern-recognition
As one of carrier.The present invention researchs and analyses three-dimensional object surface attributive character, it is proposed that a kind of to body surface varying ability
The quantisation metric method of (property).Changeability is one of most important objective dynamic attribute of body surface, be it is flexible to body surface,
The direct description of the dynamic changes ability such as ductility.Possessing approximate changeability, to mean that corresponding local surface areas has close
Mobility and dynamic attribute, to show that these regions more likely come from homologous subassembly.The characteristic is for object
The automatic division (classification) of component, description, modeling plays an important roll and direct application value.The method of the present invention belongs to computer
Vision, pattern-recognition, artificial intelligence application technical field of research.
Background technology
In recent years, although the research for three-dimension object and three-dimension curved surface obtains fast development, about three-dimensional object surface
The research of attributive character also obtains many achievements, however currently, is more needle for the analysis and research of three-dimensional object surface attribute
To static state aspect, such as, curvature, gradient, concavity and convexity, directionality etc..And in terms of descriptive modelling, although to reach more high-precision
Degree, some newly also pay close attention to object dynamic change into the model of proposition, but the target of such research is answered object dynamic change
Now and reconstruct rather than quantisation metric, therefore belong to different research directions from the latter.
Although the work for carrying out measurement research specifically for three-dimension object dynamic attribute feature in the past is on the low side, object dynamic
The importance of varying ability but can not be ignored.In fact, dynamic variability is one of core feature of body surface, such as:Many institutes
Known, the changeability on animal type objects surface is relatively strong (especially on each joint part), and the changeability of plant body surface
Universal weaker, this is that (this difference is also embodied directly in them the most important intuitive ga s safety degree difference of this two type objects
In name).Therefore for the present invention for the prior art in the deficiency of this aspect, design proposition is a kind of can to three-dimensional object surface dynamic
Change ability carries out the general approach of quantisation metric, and the variable energy of description body surface is reacted by intuitive digitized forms
Power.
Invention content
The purpose of the present invention is overcoming the deficiencies in the prior art, propose that one kind being capable of effective quantisation metric three-dimensional article
The method of body surface face varying ability.The triangular mesh being made of vertex and line segment is that current most common object mode of appearance is retouched
State mode.Along with the variation of object appearance posture, describing the length of grid middle conductor and direction can occur centainly to change with suitable
Posture is answered to change.It is clear that the variation of these line segments can directly embody the appearance similarities and differences of the object under different postures, to
Disclose the dynamically changeable ability of body surface.Based on the thought, the method for the present invention utilizes vertex in three-dimensional triangulation shape grid data
The measurement to body surface varying ability is realized in the transmission of connecting line segment.Line segment is the most base for constituting three-dimensional object surface structure
This constitutive requirements, can objectively respond the local appearance characteristic of object smallest dimension, therefore it is between different objects posture
Situation of change is the direct reflection to body surface Micro dynamic changeability, is retouched to the most careful of body surface dynamic attribute
It states.
In terms of specific implementation, it is contemplated that vertex is the core element of composition description object dimensional surface mesh, and into
The main object carrier of the related works such as row object component clustering, the changeability information that the present invention obtains in being transmitted from line segment turn
It moves on on corresponding vertex, forms the body surface Variability Metric side for main way of realization with opposite vertexes varying ability measurement
Method.
The main contents (method flow) of the method for the present invention include:
1. collecting the three-dimensional triangulation shape grid description of a variety of different postures lower surface forms of object to be measured;
2. by all posture number consecutivelies, the posture sequence of object to be measured is formed;
3. every connecting line segment on object dimensional grid is transmitted in posture sequence, and it is obtained successively in phase
Corresponding variable quantity when being transmitted in adjacent two postures:
Wherein, (xi, xj) indicate to connect the line segment of vertex i and j on grid, t represents posture number;
4. the variable quantity of line segment is transferred on corresponding grid vertex:
Wherein, U (xi) indicate and vertex xiBetween have line segment connect vertex set, MiIndicate the included vertex of the set
Quantity;
5. the variable quantity information that vertex in entire posture sequence is reacted is integrated, the variable of object three-dimensional surface is formed
Property description:
Wherein, N represents posture number.
The present invention is a kind of measure of three-dimensional object surface dynamically changeable ability, is capable of direct reactant surface
Dynamic variability feature all has intuitive changeability quantisation metric ability to any type object.Compared with the prior art, this
The advantage of invention is:
(1) valid metric can be carried out to the dynamic variability on any type object dimensional surface, meanwhile, it needs not rely on
In high density datas such as 3 D videos, it is based only upon a small amount of static attitude sample and can be realized.
(2) due to realizing that the present invention can be to minimum based on the minimum microstructure on grid line segment this three-dimension object
The dynamic variability of nanoscale object surface region is described, and realizes microcosmic precisive.
(3) the method for the present invention step is simplified, and realization process does not depend on complicated high intensity operation, therefore operation efficiency is high, full
Sufficient requirement of real-time, the universal larger three-dimension object data of reply scale, the advantage seem especially prominent.
Description of the drawings
Fig. 1 shows the method for the present invention basic procedure.
Fig. 2 shows the correspondence line segment on the different posture triangle mesh datas of three-dimension object two.
Fig. 3 shows transmittance process of the line segment in posture sequence by taking a triangular mesh as an example.
Specific implementation mode
The specific implementation mode of the present invention is introduced below:
The present invention is based on the direct measurements of sample acquisition its surface dynamic variability under object certain amount difference posture.
In specific implementation, it is necessary first to obtain a certain number of training samples.Sample data used in the present invention uses three-dimensional article
Most common network of triangle case form in body description.To realize that the line segment for connecting each vertex in grid data is suitable in each posture
Profit is transmitted, and the present invention claims the complete correspondences that each training sample data should meet vertex and connecting line segment
(correspondence).Since the correspondence of three-dimensional mesh data is to carry out the most basic precondition of three-dimensional data analysis modeling
One of, previous correlative study is more, also forms many reliable methods, therefore in actual operation, grid corresponds to not
It is a unapproachable condition, ensure that the method for the present invention convenient can be implemented.
In the quantitative aspects of training sample, it is therefore apparent that it is more for the displaying multifarious sample size of gestures of object, this
The effect that inventive method describes body surface changeability is better, but to ensure that basic effect, training posture number of samples are not answered
Less than 5.After getting these samples, the posture sequence to form object is sequentially numbered to it, is passed in case carrying out line segment
It passs.Since transmittance process will cover all posture samples, the method for the present invention to sample number rule without particular/special requirement, it is defeated according to sample
Enter sequence to be numbered.
After the completion of sample posture sequence construct, each connecting line segment on object dimensional triangular mesh is unfolded to pass
It passs, traverses each sample in posture sequence.The variable quantity for obtaining and being occurred when line segment transmits between each two posture is calculated,
Using vector description form, both including length or including direction.The variable quantity be when changing the two postures between object
The objective metric of the line segment present position surface variation degree, can directly reactant micro- regional area dynamically changeable
Property.
Due to dividing, modeling in object part, in the work such as shape control, the line segment of the apex ratio of three-dimensional mesh data has
The outstanding role become apparent and status and most basic process object, therefore the present invention is reflected line segment variations
Body surface varying ability transfer feed back in respective vertices, formed using vertex as the object changeability description side of base unit
Method.When specific operation, the varying ability of each grid vertex is determined by the mean value varying ability of coupled all line segments.
Reflected vertex changing capability progress mean value integration is transmitted two-by-two after the completion of transmittance process, then by all,
Form the comprehensive measurement to the body surface varying ability.
Claims (6)
1. based on the three-dimensional object surface Variability Metric method that line segment transmits, include the following steps:
Step 1:Obtain the three-dimensional surface morphology description under the several different postures of an object;
Step 2:It is numbered for all postures, forms posture sequence;
Step 3:To each line segment on three-dimensional surface, sequentially transmitted in posture sequence, and metric record its adjacent
Variable quantity when being transmitted between two postures,
Step 4:According to the variable quantity that line segment transmits, seek the correspondence variable quantity on each vertex on three-dimensional surface, i.e., opposite vertexes into
Row Variability Metric,
Step 5:The changeability on all vertex in posture sequence is integrated, calculates and obtains to the direct of the object dimensional surface changeability
Measurement,
2. the three-dimensional object surface Variability Metric method transmitted as described in claim 1 based on line segment, which is characterized in that step
In rapid 1, gestures of object number should be no less than 5;The description of object dimensional configuration of surface is presented with network of triangle case form, i.e., by pushing up
It point and connects its line segment and forms direct description to object outer surface, while the triangle surface covering being made of line segment
Body all surface region (this describes method for three-dimensional surface well known to current industry);Vertex and line segment in each posture is strictly right
It answers.
3. the three-dimensional object surface Variability Metric method transmitted as described in claim 1 based on line segment, which is characterized in that step
In rapid 2, number procedural order carries out, by all postures sequentially incrementally number be 1,2,3,4,5....
4. the three-dimensional object surface Variability Metric method transmitted as described in claim 1 based on line segment, which is characterized in that step
In rapid 3, (xi, xj) indicate to connect the line segment of vertex i and j, t represents posture number.
5. the three-dimensional object surface Variability Metric method transmitted as described in claim 1 based on line segment, which is characterized in that step
In rapid 4, U (xi) indicate and vertex xiBetween have line segment connect vertex set, MiIndicate the quantity on the included vertex of the set.
6. the three-dimensional object surface Variability Metric method transmitted as described in claim 1 based on line segment, which is characterized in that step
In rapid 5, N represents posture number, i.e. posture number maximum value.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810192387.3A CN108492355A (en) | 2018-03-05 | 2018-03-05 | The three-dimensional object surface Variability Metric method transmitted based on line segment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810192387.3A CN108492355A (en) | 2018-03-05 | 2018-03-05 | The three-dimensional object surface Variability Metric method transmitted based on line segment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108492355A true CN108492355A (en) | 2018-09-04 |
Family
ID=63338080
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810192387.3A Pending CN108492355A (en) | 2018-03-05 | 2018-03-05 | The three-dimensional object surface Variability Metric method transmitted based on line segment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108492355A (en) |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103077555A (en) * | 2013-01-18 | 2013-05-01 | 南京大学 | Automatic marking method composed of three-dimensional models |
CN104408396A (en) * | 2014-08-28 | 2015-03-11 | 浙江工业大学 | Action recognition method of locality matching window based on temporal pyramid |
CN107247961A (en) * | 2017-05-10 | 2017-10-13 | 西安交通大学 | A kind of trajectory predictions method of application blurring trajectorie sequence |
-
2018
- 2018-03-05 CN CN201810192387.3A patent/CN108492355A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103077555A (en) * | 2013-01-18 | 2013-05-01 | 南京大学 | Automatic marking method composed of three-dimensional models |
CN104408396A (en) * | 2014-08-28 | 2015-03-11 | 浙江工业大学 | Action recognition method of locality matching window based on temporal pyramid |
CN107247961A (en) * | 2017-05-10 | 2017-10-13 | 西安交通大学 | A kind of trajectory predictions method of application blurring trajectorie sequence |
Non-Patent Citations (1)
Title |
---|
YUELONG LI: "Vertex-Level Three-Dimensional Shape Deformability Measurement Based on Line Segment Advection", 《IET COMPUTER VISION》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Niu et al. | Unsupervised saliency detection of rail surface defects using stereoscopic images | |
Huang et al. | Combinatorial manifold mesh reconstruction and optimization from unorganized points with arbitrary topology | |
CN110415342A (en) | A kind of three-dimensional point cloud reconstructing device and method based on more merge sensors | |
Huang et al. | Automatic data segmentation for geometric feature extraction from unorganized 3-D coordinate points | |
CN108269302B (en) | Three-dimensional human body rapid reconstruction method based on simple measurement garment | |
CN104715493A (en) | Moving body posture estimating method | |
CN109410242A (en) | Method for tracking target, system, equipment and medium based on double-current convolutional neural networks | |
CN103295025B (en) | A kind of automatic selecting method of three-dimensional model optimal view | |
CN110516388A (en) | Surface tessellation point cloud model ring cutting knife rail generating method based on reconciliation mapping | |
CN108319957A (en) | A kind of large-scale point cloud semantic segmentation method based on overtrick figure | |
CN109685848A (en) | A kind of neural network coordinate transformation method of three-dimensional point cloud and three-dimension sensor | |
CN104115192A (en) | Improvements in or relating to three dimensional close interactions | |
Wu et al. | Active recognition and pose estimation of household objects in clutter | |
CN104346824A (en) | Method and device for automatically synthesizing three-dimensional expression based on single facial image | |
CN105740915B (en) | A kind of collaboration dividing method merging perception information | |
CN104933757A (en) | Method of three-dimensional garment modeling based on style descriptor | |
CN105120517B (en) | Indoor WLAN signal plan structure and localization method based on multi-dimentional scale MDS analyses | |
CN102881047B (en) | Automatic non-closed implicit curved surface reconstruction method | |
CN109344790A (en) | A kind of human body behavior analysis method and system based on posture analysis | |
CN106709203A (en) | Three-dimensional simulating method for Texgen-based weft-knitted fabric model | |
CN108280852A (en) | A kind of door and window point cloud shape detecting method and system based on laser point cloud data | |
CN110223382A (en) | Single-frame images free view-point method for reconstructing three-dimensional model based on deep learning | |
CN104134236A (en) | Boolean operation method of three-dimensional planar solids | |
CN108010122B (en) | Method and system for reconstructing and measuring three-dimensional model of human body | |
CN103236043A (en) | Plant organ point cloud restoration 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 | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180904 |
|
WD01 | Invention patent application deemed withdrawn after publication |