CN101751698B - Method and device for extracting skeletons of three-dimensional models - Google Patents
Method and device for extracting skeletons of three-dimensional models Download PDFInfo
- Publication number
- CN101751698B CN101751698B CN2010100345460A CN201010034546A CN101751698B CN 101751698 B CN101751698 B CN 101751698B CN 2010100345460 A CN2010100345460 A CN 2010100345460A CN 201010034546 A CN201010034546 A CN 201010034546A CN 101751698 B CN101751698 B CN 101751698B
- Authority
- CN
- China
- Prior art keywords
- point
- dimensional model
- value
- height
- geodesic distance
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title abstract description 13
- 238000000605 extraction Methods 0.000 claims abstract description 38
- 238000004364 calculation method Methods 0.000 claims abstract description 29
- 238000004422 calculation algorithm Methods 0.000 claims description 10
- 239000000284 extract Substances 0.000 claims description 8
- 238000011002 quantification Methods 0.000 claims 1
- 230000008569 process Effects 0.000 abstract description 6
- 210000004556 brain Anatomy 0.000 abstract description 5
- 230000016776 visual perception Effects 0.000 abstract 1
- 238000010586 diagram Methods 0.000 description 7
- 230000001149 cognitive effect Effects 0.000 description 4
- 239000012141 concentrate Substances 0.000 description 4
- 238000010276 construction Methods 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000012067 mathematical method Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002386 leaching Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000035939 shock Effects 0.000 description 1
Images
Landscapes
- Processing Or Creating Images (AREA)
Abstract
The invention provides a method for extracting skeletons of three-dimensional models, which includes the following steps that: the curvature of each point on a three-dimensional model is calculated, and the key points of the three-dimensional model are extracted according to the curvatures; the geodesic distance between each point and each key point on the three-dimensional model is calculated, and according to the minimum in the geodesic distances, the height value of each point on the three-dimensional model is obtained; according to the height values, the three-dimensional model is divided into L height intervals, each height interval corresponds to a node, a link line exists between the nodes of two neighboring height intervals, and the nodes and the link lines are formed into the skeleton of the three-dimensional model. The invention also provides a device for extracting skeletons, which comprises a key point extraction module, a height value calculation module and a skeleton extraction module. In the process of extracting the skeleton of the three-dimensional model, the invention only takes the overall structure of the three-dimensional model into account rather than the detail of the surface of the three-dimensional model, consequently, the invention not only can greatly reduce the amount of calculation, but also accords with the characteristics of the visual perception of the human brain on three-dimensional models..
Description
Technical field
The present invention relates to the 3-D view process field, relate to a kind of framework extraction method and device of three-dimensional model specifically.
Background technology
Along with developing rapidly of image processing techniques, increasing image processing algorithm has all been obtained quantum jump in theory and practice.Three-dimensional model all has in fields such as cognitive psychology, medical treatment, exploration, Model Identification, cartoon making very widely to be used, and therefore identification and the processing for three-dimensional model is one of important subject of image processing field.The skeleton of three-dimensional model has comprised abundant topology information, therefore in Model Identification and cartoon making, has played very important effect.
At present, many mathematical methods have proposed and have been applied in the skeletal extraction of three-dimensional model, for example medial axis, shock graphs, reeb graph with morse theory or the like.Wherein, What the medialaxis algorithm mainly utilized is the track at the center of three-dimensional model inscribed sphere; The skeletal extraction process of three-dimensional model is as shown in Figure 1; Reeb graph with morse theory algorithm mainly utilizes is that the equipotentiality at the particular point place of three-dimensional model is divided into several parts with three-dimensional model, and these particular points mainly refer to salient point, concave point or saddle point etc., obtain a topological diagram as shown in Figure 2 according to the structure between the each several part then.
Although these mathematical methods are very novel and each tool advantage; But they are very responsive to the tiny noise in three-dimensional model surface; Thereby the skeleton that causes finally extracting has complicated structure; The structure of this complicacy has not only been disturbed the cognitive and identification to three-dimensional model, has also brought redundant calculated amount, therefore is difficult to be applied directly in the practical matter.
The research of cognitive psychology shows that human brain tends to ignore the details of this body surface when perception three-dimensional geometry object, and interested in the significant sags and crests of this object naturally.Human brain can decompose object at these sags and crests places naturally, removes to discern this object then step by step.Therefore when extracting a three-dimensional model skeleton that meets cognitive characteristics, the details on considered three-dimensional model surface not, and only need consider the one-piece construction of three-dimensional model.
Summary of the invention
In order to overcome the deficiency of prior art, the objective of the invention is to provide a kind of and can know demonstration integrally-built framework extraction method of three-dimensional model and device for extracting skeletons.
To achieve these goals, the invention provides a kind of framework extraction method of three-dimensional model, may further comprise the steps: the curvature of each point on the Calculation of Three Dimensional model, according to the key point of curvature extraction three-dimensional model; Each puts the geodesic distance of each key point on the Calculation of Three Dimensional model, obtains the height value of each point on the three-dimensional model according to the minimum value in the geodesic distance; It is highly interval according to height value three-dimensional model to be divided into L, each highly interval corresponding node, and adjacent two highly exist a line between the node in the interval, are made up of the skeleton of three-dimensional model node and line.
Therewith accordingly; The present invention also provides a kind of device for extracting skeletons of three-dimensional model; Comprise key point extraction module, height value calculation module and skeleton extraction module, key point extraction module is used for the curvature of each point on the Calculation of Three Dimensional model, extracts the key point of three-dimensional model according to curvature; Height value calculation module is used for that each puts the geodesic distance of each key point on the Calculation of Three Dimensional model, obtains the height value of each point on the three-dimensional model according to the minimum value in the geodesic distance; It is highly interval that skeleton extraction module is used for according to height value three-dimensional model being divided into L, each highly interval corresponding node, and adjacent two highly exist a line between the node in the interval, are made up of the skeleton of three-dimensional model node and line.
The invention has the beneficial effects as follows; In the skeletal extraction process of three-dimensional model, only consider the one-piece construction of three-dimensional model; Do not consider the details on three-dimensional model surface, not only algorithm is simple, has greatly practiced thrift calculated amount; Also be fit to the visually-perceptible characteristics of human brain, be fit to very much the application in fields such as Model Identification and cartoon making three-dimensional model.
Description of drawings
Above-mentioned and/or additional aspect of the present invention and advantage are from obviously with easily understanding becoming the description of embodiment below in conjunction with accompanying drawing, wherein:
Fig. 1 is the principle schematic of medial axis algorithm in the prior art;
Fig. 2 is the principle schematic of reeb graph with morse theory algorithm in the prior art;
Fig. 3 is the process flow diagram according to the framework extraction method of the embodiment of the invention;
Fig. 4 is the leaching process synoptic diagram according to the framework extraction method of the embodiment of the invention;
Fig. 5 is according to the extraction of the framework extraction method of embodiment of the invention synoptic diagram as a result; And
Fig. 6 is the structural representation according to the device for extracting skeletons of the embodiment of the invention.
Embodiment
Describe embodiments of the invention below in detail, the example of said embodiment is shown in the drawings, and wherein identical from start to finish or similar label is represented identical or similar elements or the element with identical or similar functions.Be exemplary through the embodiment that is described with reference to the drawings below, only be used to explain the present invention, and can not be interpreted as limitation of the present invention.
The present invention carries out the skeletal extraction of three-dimensional model to the visually-perceptible characteristics of three-dimensional model according to human brain; In the skeletal extraction process; The present invention only considers the significant particular point of three-dimensional model visual signature; The details of ignoring the three-dimensional model surface, thus the skeleton that extracts can clearly illustrate the one-piece construction of three-dimensional model, is suitable in the fields such as Model Identification and cartoon making.
Fig. 3 is the process flow diagram of framework extraction method of the three-dimensional model of the embodiment of the invention, and Fig. 4 is the synoptic diagram of the processing procedure of framework extraction method shown in Figure 3.The framework extraction method of this three-dimensional model mainly may further comprise the steps:
S310: the curvature of each point on the Calculation of Three Dimensional model, extract the key point of said three-dimensional model according to curvature.The key step of extracting key point is:
S311: calculate the remarkable value of each point of threedimensional model based on curvature, and form remarkable point set greater than the point of preset threshold value by remarkable value.Forming remarkable point set may further comprise the steps:
S3111: two principal curvatures k of each point of Calculation of Three Dimensional model
1And k
2, and calculating mean curvature avg (v) does
S3112: (v) each point of Calculation of Three Dimensional model is at metric space σ according to mean curvature avg
iIn curvature C (v, σ
i) do
Wherein, N (v, 2 σ
i) be that some v is at metric space σ
iIn neighborhood.
Each point of Calculation of Three Dimensional model is at metric space σ then
iIn remarkable value S (v, σ
i) do
S(v,σ
i)=|C(v,σ
i)-C(v,2σ
i)|(2)
S3113: according to S (v, σ
i) the remarkable value S of each point of Calculation of Three Dimensional model (v) does
S(v)=∑
i(M
i-m
i)
2S(v,σ
i) (3)
Wherein, M
iBe corresponding metric space σ
iThe maximum significantly value of middle three-dimensional model, m
iBe corresponding metric space σ
iRemove M on the middle three-dimensional model
iThe remarkable value of local maximum point in addition average with, local maximum point is for significantly being worth at neighborhood N (v, 2 σ
i) interior maximum point.
Threshold value can be regulated according to the accuracy requirement of actual conditions and Flame Image Process, in one embodiment of the invention threshold value is redefined for the maximum significantly value M of three-dimensional model
i0.9 times.Significantly (v) be higher than the point set that the point of threshold value constituted is exactly the significant point set of visual signature on the three-dimensional model to value S, promptly remarkable point set.For the three-dimensional model shown in Fig. 4 (a), extract result behind the remarkable point set shown in Fig. 4 (b).
S312: extract the salient point that significant point is concentrated, and salient point is carried out cluster.
S3121: defining two principal curvaturess is salient point greater than zero point all, concentrates all salient points all to extract significant point and forms the salient point collection.
S3122: the connectedness according to salient point is divided connected region, and with being made as the representative point of representing connected region to the minimum salient point of the geodesic distance sum of other salient point in each connected region.
If have a paths from a v1 to a v2, all points all belong to the point that salient point is concentrated on this path, then put v1 and are communicated with some v2.Connectedness according between points is divided into different connected regions with the salient point collection; From each connected region, choose the minimum point of geodesic distance sum of other each point in this connected region; And be the point of the whole connected region of representative with this fixed fire, be designated as representative point, represent point set thereby form.The calculation procedure of the geodesic distance value between calculation level v1 and the some v2 is:
S31221: the geodesic distance value d of each the some v in the connected region (v) is made as infinity, the geodesic distance value d (v1) of a v1 is made as 0, and institute is joined a little among the List that tabulates.
S31222: (v) minimum some v3 puts v2 exactly like fruit dot v3, and then the geodesic distance value d (v3) of this time point v3 is exactly the geodesic distance value between some v1 and the some v2, calculates and finishes, otherwise carry out step S31223 from List, to select a geodesic distance value d.
S31223: suppose that v4 representes to carry out the point that the limit is connected with point between the v3, the length of fillet be length (v3, v4); If d (v3)+length (v3, v4)<d (v4), the geodesic distance value d (v4) that then will put v4 is updated to d (v4)=d (v3)+length (v3; V4); And will put v4 and be inserted among the tabulation List, after having upgraded all and having put the some v4 that v3 is connected, will put v3 and from the List that tabulates, delete.
S31224: repeating step S31222 and S31223 are empty up to tabulation List.Can obtain the geodesic distance value between a v1 and the some v2 this moment.
S3123: adopt K-central point clustering algorithm (K-medoid algorithm) that representative point is carried out cluster, the point that geodesic distance is close gathers into one type, and geodesic distance point far away is distributed in the different classes.The key step of K-central point clustering algorithm is:
S31231: calculation representative point concentrates each to put the geodesic distance that this representative point is concentrated other point, and preserves.
S31232:, concentrate from representative point and choose K point arbitrarily as the medoid each type if representative point is agglomerated into the K class.
S31233:, concentrate remaining point to assign in K type representative point and go based on the principle nearest with the medoid geodesic distance.
S31234: from each type, select the new medoid of the shortest point of the average geodesic distance of other point in this type as such.
S31235: repeating step S31233 and S31234 no longer change up to the medoid of each type.At this moment, represent point set to be divided into the K class.
S313: the maximum point of remarkable value is a key point in each the salient point class after the extraction cluster.If step S312 is divided into the K class with the point on the three-dimensional model, then explain to have K key point in this three-dimensional model.Each key point is 1 point corresponding to the skeleton moderate, just the starting point of a branch or last point in the skeleton.Image among Fig. 4 (b) is extracted result after the key point shown in Fig. 4 (c).
S320: each puts the geodesic distance of each key point on the Calculation of Three Dimensional model, obtains the height value of each point on the three-dimensional model according to the minimum value in the geodesic distance.Concrete steps comprise:
S321: (v) be made as infinity, the geodesic distance value d to self (v) is made as 0, and institute is had a few adding tabulate among the List with each key point to the geodesic distance value d of each key point with each some v on the three-dimensional model.
S322: for d among the List (v) minimum some v, some v1 with put that between the v length to be arranged be length (v, fillet v1); If d (v)+length (v; V1)<d (v1), the geodesic distance value d (v1) that then will put v1 be updated to d (v1)=d (v)+length (v, v1); Having upgraded all and having put v has after the some v1 of fillet, will put v and from List, delete.
S323: repeating step of updating S322, is empty until above-mentioned List, and (minimum value v) is made as minimum value the height value of each point on the three-dimensional model thereby obtain the geodesic distance d of each point on the three-dimensional model.
S330: it is highly interval according to height value three-dimensional model to be divided into L, each highly interval corresponding node, and adjacent two highly exist a line between the node in the interval, are made up of the skeleton of three-dimensional model node and line.Concrete steps are:
S331: interval [min, max] is divided into L part with height value, obtains quantizing height value h
iFor
Wherein, min is the minimum value of height value, and max is the maximal value of height value.
S332: according to quantizing height value h
iObtain the equipotential line shown in Fig. 4 (d), to obtain height sequence of intervals { [h
I-1, h
i, i=1 ..., L, thus three-dimensional model is divided into and highly interval corresponding triangular plate collection, each triangular plate set pair is answered a node, has a line between the node that two adjacent triangular plates are concentrated.According to the signal of Fig. 4 (e), just can form the skeleton of three-dimensional model by these nodes and line.
Framework extraction method according to three-dimensional model of the present invention carries out skeletal extraction to the three-dimensional picture among Fig. 5 (a), finally can obtain the skeleton shown in Fig. 5 (b).(a) of comparison diagram 5 and can find out that (b) framework extraction method of the present invention can clearly illustrate the one-piece construction of three-dimensional model, and can not disturbed by the details on three-dimensional model surface.
Shown in Fig. 6 is the device for extracting skeletons 100 according to the three-dimensional model of the embodiment of the invention; It comprises key point extraction module 110, height value calculation module 120 and skeleton extraction module 130; Key point extraction module 110 is used for the curvature of each point on the Calculation of Three Dimensional model, extracts the key point of three-dimensional model according to curvature; Height value calculation module 120 is used for that each puts the geodesic distance of each key point on the Calculation of Three Dimensional model, obtains the height value of each point on the three-dimensional model according to the minimum value in the geodesic distance; It is highly interval that skeleton extraction module 130 is used for according to height value three-dimensional model being divided into L, each highly interval corresponding node, and adjacent two highly exist a line between the node in the interval, are made up of the skeleton of three-dimensional model node and line.
Key point extraction module 110 is at first calculated the remarkable value of each point of three-dimensional model according to curvature when extracting the key point of three-dimensional model according to curvature, and form remarkable point set by remarkable value greater than the point of preset threshold value; Extract the salient point that significant point is concentrated then, and salient point is carried out cluster; Extracting in each the salient point class after the cluster significantly at last, the point of value maximum is a key point.
Wherein, key point extraction module 110 is formed remarkable point set and is comprised: two principal curvatures k of each point of Calculation of Three Dimensional model at first
1And k
2, and calculate mean curvature
(v) each point of Calculation of Three Dimensional model is at metric space σ according to avg for extraction module then
iIn curvature C (v, σ
i) do
Wherein, N (v, 2 σ
i) be that some v is at metric space σ
iIn neighborhood, and the Calculation of Three Dimensional model each the point at metric space σ
iIn remarkable value S (v, σ
i) be S (v, σ
i)=| C (v, σ
i)-C (v, 2 σ
i) |; At last according to S (v, σ
i) the remarkable value S of each point of Calculation of Three Dimensional model (v) does
Wherein, M
iBe corresponding metric space σ
iThe maximum significantly value of middle three-dimensional model, m
iBe corresponding metric space σ
iRemove M on the middle three-dimensional model
iThe remarkable value of local maximum point in addition average with, local maximum point is for significantly being worth at neighborhood N (v, 2 σ
i) interior maximum point.
During the height value of height value calculation module 120 each point on the Calculation of Three Dimensional model; At first each some v on the three-dimensional model (v) is made as infinity to the geodesic distance value d of each key point; Geodesic distance value d to self (v) is made as 0, and institute is a bit added among the tabulation List with each key point; Then for d among the List (v) minimum some v, some v1 with put that between the v length to be arranged be length (v, fillet v1); If d (v)+length (v; V1)<d (v1), the geodesic distance value d (v1) that then will put v1 be updated to d (v1)=d (v)+length (v, v1); Having upgraded all and having put v has after the some v1 of fillet, will put v from said List, to delete; Repeating above-mentioned step of updating, is empty until List, and (minimum value v) is made as minimum value the height value of each point on the three-dimensional model thereby obtain the geodesic distance d of each point on the three-dimensional model.
One of ordinary skill in the art will appreciate that and realize that all or part of step that the foregoing description method is carried is to instruct relevant hardware to accomplish through program; Described program can be stored in a kind of computer-readable recording medium; This program comprises one of step or its combination of method embodiment when carrying out.
In addition, each functional unit in each embodiment of the present invention can be integrated in the processing module, also can be that the independent physics in each unit exists, and also can be integrated in the module two or more unit.Above-mentioned integrated module both can adopt the form of hardware to realize, also can adopt the form of software function module to realize.If said integrated module realizes with the form of software function module and during as independently production marketing or use, also can be stored in the computer read/write memory medium.
The above-mentioned storage medium of mentioning can be a ROM (read-only memory), disk or CD etc.
The above only is a preferred implementation of the present invention; Should be pointed out that for those skilled in the art, under the prerequisite that does not break away from the principle of the invention; Can also make some improvement and retouching, these improvement and retouching also should be regarded as protection scope of the present invention.
Claims (3)
1. the framework extraction method of a three-dimensional model is characterized in that, may further comprise the steps:
1) curvature of each point on the Calculation of Three Dimensional model is extracted the key point of said three-dimensional model according to said curvature;
2) each puts the geodesic distance of each said key point on the said three-dimensional model of calculating, obtains the height value of each point on the said three-dimensional model according to the minimum value in the said geodesic distance;
3) based on said height value said threedimensional model is divided into L highly interval; The interval corresponding node of each said height; There is a line between the said node in two adjacent said height intervals, forms the skeleton of said threedimensional model by said node and said line;
Wherein, step 1) specifically comprises the steps: 11) curvature of each point on the Calculation of Three Dimensional model, calculate the remarkable value of each point of said three-dimensional model according to said curvature;
12) form remarkable point set by said remarkable value greater than the point of preset threshold value;
13) extract the salient point that said significant point is concentrated, and said salient point is carried out cluster;
14) the maximum point of remarkable value is a key point described in each the salient point class after the extraction cluster;
Wherein, step 11) comprises:
Calculate two principal curvatures k of each point of said three-dimensional model
1And k
2, and calculating mean curvature avg (v) does
(v) calculate each point of said three-dimensional model according to said avg at metric space σ
iIn curvature C (v, σ
i) do
Wherein, N (v, 2 σ
i) be that some v is at metric space σ
iIn neighborhood, and calculate said three-dimensional model each the point at metric space σ
iIn remarkable value S (v, σ
i) be S (v, σ
i)=| C (v, σ
i)-C (v, 2 σ
i) |;
According to said S (v, σ
i) the remarkable value S that calculates each point of said three-dimensional model (v) does
Wherein, M
iBe corresponding metric space σ
iDescribed in the maximum significantly value of three-dimensional model,
Be corresponding metric space σ
iDescribed in remove M on the three-dimensional model
iThe remarkable value of local maximum point in addition average with, said local maximum point is for significantly being worth at neighborhood N (v, 2 σ
i) interior maximum point;
Wherein, step 13) comprises: extracting concentrated two principal curvaturess of said significant point is salient point greater than zero point all;
Connectedness according to said salient point is divided connected region;
And with being made as the representative point of representing said connected region to the minimum salient point of the geodesic distance sum of other said salient point in each said connected region;
Adopt K-central point clustering algorithm that said representative point is carried out cluster.
2. framework extraction method according to claim 1 is characterized in that, the said height value that obtains each point on the three-dimensional model comprises:
(v) be made as infinity, the geodesic distance value d to self (v) is made as 0, and institute is had a few adding tabulate among the List with each said key point to the geodesic distance value d of each key point with each some v on the said three-dimensional model;
For d among the said List (v) minimum some v, between some v1 and the said some v length being arranged is length (v, fillet v1); If d (v)+length (v; V1)<d (v1), then with the geodesic distance value d (v1) of said some v1 be updated to d (v1)=d (v)+length (v, v1); Upgraded all and said some v has after the some v1 of fillet, said some v deleted from said List;
Repeating said step of updating, is empty until said List, and (minimum value v) is made as said minimum value the height value of each point on the said three-dimensional model thereby obtain the geodesic distance d of each point on the said three-dimensional model.
3. framework extraction method according to claim 1 is characterized in that, the skeleton of said composition three-dimensional model comprises:
With the interval [min of height value; Max] be divided into L part; Obtain quantizing height value
wherein; Min is the minimum value of said height value, and max is the maximal value of said height value;
According to said quantification height value h
iObtain equipotential line, to obtain height sequence of intervals { [h
I-1, h
i]; I=1 ..., L; Thereby said three-dimensional model is divided into and the interval corresponding triangular plate collection of said height; Each said triangular plate set pair is answered a node, has a line between the said node that two adjacent said triangular plates are concentrated, and is made up of the skeleton of said three-dimensional model said node and said line.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2010100345460A CN101751698B (en) | 2010-01-22 | 2010-01-22 | Method and device for extracting skeletons of three-dimensional models |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN2010100345460A CN101751698B (en) | 2010-01-22 | 2010-01-22 | Method and device for extracting skeletons of three-dimensional models |
Publications (2)
Publication Number | Publication Date |
---|---|
CN101751698A CN101751698A (en) | 2010-06-23 |
CN101751698B true CN101751698B (en) | 2012-01-04 |
Family
ID=42478636
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN2010100345460A Expired - Fee Related CN101751698B (en) | 2010-01-22 | 2010-01-22 | Method and device for extracting skeletons of three-dimensional models |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN101751698B (en) |
Families Citing this family (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101957992B (en) * | 2010-09-30 | 2012-05-02 | 清华大学 | Data characteristic extracting and matching method for two-dimensional injective mapping curved surface |
CN105069777A (en) * | 2015-07-02 | 2015-11-18 | 广东工业大学 | Automatic extracting method of neck-edge line of preparation body grid model |
CN107398900A (en) * | 2017-05-27 | 2017-11-28 | 芜湖星途机器人科技有限公司 | Active system for tracking after robot identification human body |
CN109648841A (en) * | 2018-12-10 | 2019-04-19 | 西安交通大学 | A kind of multi-direction multiple degrees of freedom 3D printing dicing method |
CN109887072A (en) * | 2019-01-16 | 2019-06-14 | 中德(珠海)人工智能研究院有限公司 | A kind of extraction system of threedimensional model |
CN112183159A (en) * | 2019-07-03 | 2021-01-05 | 四川大学 | Construction of a skeletal model of a non-human target in an image using keypoints |
CN112785689A (en) * | 2021-01-28 | 2021-05-11 | 北京爱奇艺科技有限公司 | Three-dimensional model construction method, device and system and storage medium |
CN113409452B (en) * | 2021-07-12 | 2023-01-03 | 深圳大学 | Three-dimensional line generation method, storage medium and system |
CN114266875B (en) * | 2021-12-23 | 2024-07-23 | 中国科学院软件研究所 | Reeb graph generation method and system based on reconciliation measurable leaf structure |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1629888A (en) * | 2003-12-17 | 2005-06-22 | 中国科学院自动化研究所 | A skeletonized object rebuild method |
CN101231760A (en) * | 2007-01-24 | 2008-07-30 | 中国科学院自动化研究所 | Method for abstracting grade framework and stereo decomposing of arborescence figure |
-
2010
- 2010-01-22 CN CN2010100345460A patent/CN101751698B/en not_active Expired - Fee Related
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN1629888A (en) * | 2003-12-17 | 2005-06-22 | 中国科学院自动化研究所 | A skeletonized object rebuild method |
CN101231760A (en) * | 2007-01-24 | 2008-07-30 | 中国科学院自动化研究所 | Method for abstracting grade framework and stereo decomposing of arborescence figure |
Non-Patent Citations (2)
Title |
---|
田绪红等.基于B-Snake的三维图像骨架化方法.《计算机科学》.2008,第35卷(第9期),241-265. * |
韩国强等.三维图像骨架化方法综述.《小型微型计算机系统》.2007,(第9期),1695-1699. * |
Also Published As
Publication number | Publication date |
---|---|
CN101751698A (en) | 2010-06-23 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN101751698B (en) | Method and device for extracting skeletons of three-dimensional models | |
Maharjan et al. | Understanding rural outmigration and agricultural land use change in the Gandaki Basin, Nepal | |
CN102521884B (en) | Three-dimensional roof reconstruction method based on LiDAR data and ortho images | |
Benedikt | The valuable citizens of smart cities: The case of Songdo City. | |
CN103400372B (en) | A kind of three-dimensional topology information extracting method described based on Reeb figure | |
US20060176213A1 (en) | Method and apparatus for enhancing a digital elevation model (DEM) for topographical modeling | |
TWI309027B (en) | Computer implemented method and computer system | |
CN104850846A (en) | Human behavior recognition method and human behavior recognition system based on depth neural network | |
CN106227726A (en) | A kind of path extraction method based on track of vehicle data | |
CN103745497B (en) | Plant growth modeling method and system | |
CN108229740B (en) | Method and device for determining business circle boundary, server and storage medium | |
Harrie | An optimisation approach to cartographic generalisation | |
Svoray et al. | Demonstrating the effect of exposure to nature on happy facial expressions via Flickr data: Advantages of non-intrusive social network data analyses and geoinformatics methodologies | |
CN105701798A (en) | Point cloud extraction method and device for columnar object | |
Wichmann et al. | Roofn3d: Deep learning training data for 3d building reconstruction | |
CN112435333B (en) | Road scene generation method and related device | |
Pettit et al. | Big bicycle data processing: From personal data to urban applications | |
Fränti et al. | Averaging GPS segments competition 2019 | |
Griffin et al. | A decision tree classification model to automate trip purpose derivation | |
CN116051699B (en) | Dynamic capture data processing method, device, equipment and storage medium | |
CN112687079A (en) | Disaster early warning method, device, equipment and storage medium | |
CN105528423A (en) | Self-adaptive co-location pattern obtaining method and apparatus considering spatial instance distance weight | |
Hyun | Geography: Rethinking the ‘urban’and ‘urbanization’ | |
Zhang et al. | Urban visual intelligence: Studying cities with AI and street-level imagery | |
CN114637305B (en) | Unmanned aerial vehicle shortest path planning method and device |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20120104 Termination date: 20220122 |