CN101004748A - Method for searching 3D model based on 2D sketch - Google Patents
Method for searching 3D model based on 2D sketch Download PDFInfo
- Publication number
- CN101004748A CN101004748A CN 200610114097 CN200610114097A CN101004748A CN 101004748 A CN101004748 A CN 101004748A CN 200610114097 CN200610114097 CN 200610114097 CN 200610114097 A CN200610114097 A CN 200610114097A CN 101004748 A CN101004748 A CN 101004748A
- Authority
- CN
- China
- Prior art keywords
- model
- sketch
- view
- feature
- fourier
- 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.)
- Granted
Links
Images
Abstract
A method for indexing 3-D model based on 2-D sketch includes generating character databank of 3-D model according to character of each calculated view, picking up shape character of sketch by client end of indexing system, using server to match sketch character with characters in said character databank , selecting different algorithms to calculate similar distance of 2-D sketch to each 3-D model, sequencing obtained similar distances and returning precedence and index image as well as URL of 3-D mode being sequenced at front to client end according to displayed model number assigned by system.
Description
Technical field
The present invention relates to a kind of content-based method for searching three-dimension model, particularly a kind of method for searching three-dimension model based on two-dimentional sketch.
Background technology
Rapid development of Internet is brought the exponential increase of quantity of information.People need fast and effeciently to search for and visit required information resources urgently.Text retrieval technical development based on keyword is comparatively ripe, and Google, Baidu have all become common tool in people's life, but is not suitable for the retrieval of multimedia messages.Content-based multimedia information retrieval comprises the retrieval to voice, image, video, animation and three-dimensional model, directly the content of multimedia object is analyzed, extract the feature and the semanteme of content of multimedia, utilize these features to set up index database then, the line retrieval of going forward side by side.
Three-dimensional model is widely used in a lot of fields such as computer-aided design (CAD) (CAD), virtual reality, recreation, the making of film special efficacy, archaeology, biology, chemistry.As a kind of three-dimensional information resource, the quantity of the three-dimensional model on the internet is considerable, and presents the trend of quick increase.How reusing existing three-dimensional model resource becomes a problem of needing solution badly, arises at the historic moment based on the three-dimensional model search engine of Web.Existing three-dimensional modeling data resource has positive effect to three-dimensional model search on the network to efficiently obtaining and reusing, and can save and spend in the time and efforts that makes up in the three-dimensional model process in a large number.Three-dimensional model search is based on the hot issue in the multimedia information retrieval field of content, is with a wide range of applications, and lot of domestic and international mechanism just is being devoted to the research of this direction.
Owing to represent the shape of ferret out with two-dimentional sketch, the use habit that meets the people, domestic consumer is not subjected to any training, it is on screen, draw the roughly geometric configuration of searched targets of available mouse, therefore utilize two-dimentional sketch to retrieve the three-dimensional model search mode that three-dimensional model becomes a kind of close friend, have friendly interface, operation easy-to-use advantage directly perceived.
At present in the external disclosed document, Loffler J. " Content-based retrieval of 3Dmodels in distributed Web databases by visual shape information " In:Proceedings of IEEE International Conference on InformationVisualization 2000, London, UK proposes among the 2000:82-87 to retrieve three-dimensional model with two-dimensional approach under distributed network environment the earliest.Okada Y. " 3D model database system byhand sketch query " In:Proceedings of IEEE International Conference onMultimedia and Expo 2002, Lausanne, Switzerland, having provided one among the 2002:889-892 is the three-dimensional model searching system of searching request with two-dimentional sketch, obtain 3 width of cloth perspective views of model from 3 major axes orientations, the range distribution of pixel on the perspective view profile has been adopted in feature extraction.Min P., Chen J., Funkhourser T. " A 2D Sketch Interface for a 3D Model SearchEngine " .In:Proceedings of SlGGRAPH 2002 Sketches﹠amp; Applications, San Antonio, Texas, USA has provided the two-dimentional sketch drawn with several users method as query requests among the 2002:138.
But the solution thinking of above-mentioned document has 3 deficiencies: do not prejudge object when (1) uses contour feature and whether be fit to this character description method, be subject to connective influence, poor robustness; (2) some has only used single feature, but because every kind of feature has only been considered object shapes characteristic in a certain respect, therefore, only exists not enough according to single signature search; (3) though other have used various features, every kind of feature is put on an equal footing, fail to reflect the retrieval precision difference of feature to different classes of query requests.
Summary of the invention
Technology of the present invention is dealt with problems and is: overcome the deficiencies in the prior art, a kind of method based on two-dimentional sketch retrieval three-dimensional model is provided, it has considered that contour feature is subjected to connectedness to influence big characteristics, adopt pre-service and judge the mode that is communicated with set, guarantee only the destination object that is fit to be used contour feature; And the mode by various features self-adaptation combined retrieval further improves the precision of retrieval, the deficiency when avoiding adopting single feature.
Technical solution of the present invention: based on the method for searching three-dimension model of two-dimentional sketch, it is characterized in that: needs are handled the three-dimensional modeling data storehouse in advance, at first obtain the viewdata storehouse by the three-dimensional modeling data storehouse, calculate the feature of every width of cloth view then, final generating feature database, the client of searching system is extracted the shape facility of two-dimentional sketch, and feature submitted to server, server is at first with the feature of sketch and the characteristic matching in the property data base, according to the kind of sketch feature, select the different algorithm computation two dimension sketches and the similarity distance of each three-dimensional model during coupling; Then similarity distance is sorted; According to the homepage display model number of system's appointment, the precedence of the three-dimensional model that ordering is forward, index image, URL etc. return to client at last.For can be when the combined retrieval dynamic calculation feature weight, also to make up training set according to existing classified information, make the model number of each classification equate during structure, and guarantee that the model set of known class belong to model database.
Concrete as follows by step:
(1) each model in the three-dimensional modeling data storehouse is handled, obtained the perspective view of three-dimensional model;
(2) with the perspective view of all three-dimensional models, generate the viewdata storehouse;
(3) processing of every width of cloth view process characteristic extracting module in the viewdata storehouse to generation, the generating feature database;
(4) to the Shape Feature Extraction of the two-dimentional sketch of client, comprise Zernike moment characteristics and Fourier boundary descriptor, identical with the method for extracting shape features of view in the step (3);
(5) pattern search module is responsible for the two-dimentional sketch feature of client submission and the characteristic matching in the property data base are calculated similarity distance and ordering, finally generates result for retrieval.
The present invention's advantage compared with prior art is:
(1), before the tracking target contours of objects, taked Bridge operation in two value filterings and Clean to operate and improved robustness at the problem of contour feature than the poor robustness of provincial characteristics.
(2) when describing the feature of destination object, having adopted regional Zernike square is method main, that the Fourier profile is described as assisting.Whether the target area is fit to adopt the Fourier profile to describe to make is prejudged.
(3) utilize existing classified information in the model database, make up training set, when search, utilize the training set weight of calculated characteristics automatically, thereby improve the accuracy of retrieval.The present invention can adjust feature weight automatically according to query requests, can reflect the retrieval precision difference of different characteristic for different classes of query object, maximizes favourable factors and minimizes unfavourable ones.
Description of drawings
Fig. 1 is the main modular and the flow process of the three-dimensional model searching system based on two-dimentional sketch of the present invention;
Fig. 2 is a model aircraft of the present invention and its three views;
Fig. 3 is the calculation process of the Zernike square of view of the present invention;
Fig. 4 is the calculation process of the Fourier boundary descriptor of view of the present invention;
The calculation process of the feature weight when Fig. 5 retrieves for the characteristics combination among the present invention;
The two-dimentional sketch example that Fig. 6 describes for the two-dimensional graphics tool set among the present invention and user;
Fig. 7 is the retrieval example of a people's who draws with the user of the present invention two-dimentional sketch as query requests.
Embodiment
In conjunction with Fig. 1, describe system flow of the present invention and main modular in detail.
1. each model in the three-dimensional modeling data storehouse is handled, obtained the perspective view of three-dimensional model.
The perspective view that does not comprise model in the original three-dimensional modeling data storehouse, the present invention obtains the perspective view of three-dimensional model by model projection figure generation module.This module realizes that based on OpenGL concrete steps are as follows:
(1) standardization of model coordinate comprises translation transformation standardization and rotational transform standardization.At first carry out the translation transformation standardization, the barycenter of computation model moves to true origin with barycenter.Adopt improved continuous P CA method to determine three main shafts of model then, it is rotated the conversion standardization, its coordinate axis and main shaft are overlapped around the initial point rotating model.
(2) adopt the projection of orthogonal projection mode, surround the position that the size of ball is calculated six cutting faces of view frustums according to the minimum of model, the minimum of model is surrounded ball all the time in view frustums when guaranteeing projection.
(3) define the negative direction that direction of visual lines is three coordinate axis successively, preserve content displayed in the window, be three width of cloth perspective views of model.
2. for the perspective view of all models, generate the viewdata storehouse by the perspective view processing module.
The purpose of perspective view processing module is: perspective view is implemented a series of image processing operations, comprise the Close computing in image binaryzation and the two-value shape filtering, to avoid the interference of color, illumination, texture, interior details and bad other factors such as triangle gridding structure.Thus,, obtained the viewdata storehouse of model by processing to the three-dimensional modeling data storehouse, corresponding three views of each model in the three-dimensional modeling data storehouse, view is binary picture, and the prospect that guarantees is 1, and background is 0.
Concrete grammar is as follows:
(1) perspective view is carried out image binaryzation, make that it becomes that prospect is 1, background is 0 binary picture.
(2) image is carried out Close computing in the two-value shape filtering, avoid interior details too much and the zone fill the influence of insufficient these two problems that may occur.
Fig. 2 provides a model aircraft and its 3 width of cloth views.Observe aircraft from the y axle, obtain top right plot; Observe aircraft from the z axle, obtain scheming in the right side; Observe aircraft from the x axle, obtain bottom-right graph.
3. the processing of every width of cloth view process characteristic extracting module in the viewdata storehouse to generation, the generating feature database.
The function of characteristic extracting module is: every width of cloth view is calculated view feature, the present invention in the characteristics of calculating the view feature stage is: only the Fourier boundary descriptor is calculated in the target area that is fit to, this is because the Fourier boundary descriptor belongs to contour feature, the Zernike square belongs to provincial characteristics, when calculating, provincial characteristics used regional all interior picture elements, be subjected to the influence of noise etc. little, robustness is better than contour feature.
Concrete steps are as follows:
(1) the Zernike moment characteristics of calculating view.
As Fig. 3, the algorithm flow that calculates the Zernike moment characteristics of view can be specifically described as:
(a) take Close computing in the two-value shape filtering.
(b) regional barycenter is moved to coordinate origin, edges of regions is zoomed to 1 to the ultimate range of barycenter, this two steps operation is in order to guarantee that the target area feature is not subjected to the influence of translation, transformation of scale.
(c) calculate the Zernike square.
(d) the plural coefficient that obtains is asked the spoke value, the standardization of spoke value is by realizing the spoke value divided by the pixel number in the foreground area.
Spoke value after the standardization is as the Zernike moment characteristics of view the most at last.
(2) the Fourier boundary descriptor is calculated in the target area that is fit to, as contour feature.
As Fig. 4, the algorithm flow that calculates the Fourier boundary descriptor can be specifically described as:
(a) before the tracing area profile, take Bridge operation and Clean in the two-value shape filtering to operate the connectedness that as far as possible guarantees destination object.Bridge operation with the target area near but disconnected part couple together, the Clean operation can be removed isolated bright spot.This two steps shape filtering operation all is in order to improve the robustness of algorithm.
(b) connectedness of judgement target area.For through still disconnected target area behind the first two steps filtering operation, it is considered herein that this kind target is not suitable for using contour feature.Because if the target area still is not communicated with, the object fracture in the view is described or has a plurality of objects, therefore be not suitable for adopting the Fourier boundary descriptor as feature.
(c) if the target area is communicated with, the outline of tracking target at first then; To the edge pixel sequences, calculate their distances successively then to regional center; Next, carry out Fourier transform; At last to obtaining one group of plural number coefficient { a behind the Fourier transform
n, but { a
nWith image in the selection of the rotation, convergent-divergent of target and different profile starting point relevant, therefore must carry out standardization, through standardization obtain | b
n|, n=1 ... N-1}, wherein b
n=a
n/ a
0, as the Fourier boundary descriptor of view.
4. the method for extracting shape features of view is similar in method for extracting shape features of the two-dimentional sketch of client (comprising Zernike moment characteristics and Fourier boundary descriptor) and the step 3.At first two-dimentional sketch is carried out Close computing in binary conversion treatment and the two-value shape filtering, calculate the Zernike moment characteristics then,, calculate the description of Fourier profile for through meeting the target of connectivity platform after Bridge operation and the Clean operation.
In order to alleviate the burden of server end, the feature extraction of two-dimentional sketch is placed on client calculates, only submit the characteristic vector data that calculates to server.Another advantage of doing like this is: the characteristic vector data amount is much smaller than the pictorial data amount of two-dimentional sketch, and the data volume of Network Transmission is little, helps to improve the response speed of system.
5. pattern search module is responsible for the two-dimentional sketch feature of client submission and the characteristic matching in the property data base are calculated similarity distance and ordering, finally generates result for retrieval.
Wherein, characteristic matching and similarity distance calculate and ordering is an emphasis of the present invention, and concrete steps are as follows:
(1) makes up training set.Department pattern in the model database has comprised classification information (perhaps artificially department pattern being classified), the classification here refers to the semantic category according to people's cognition, from each classification, select K typical model and form training set, the span of K is [1, other model sum of infima species].
(2) when connectivity platform is satisfied in the target area in the sketch, the sketch feature comprises the Zemike square and the Fourier profile is described two kinds of features, then finds the solution result for retrieval in conjunction with these two kinds of features, mainly needs the weight of calculated characteristics, and key step is:
(a) respectively according to the similarity distance of each model in these two kinds of feature calculation two dimension sketches and the database, adopt the minimum value of the characteristic distance of sketch and 3 views of model to measure, if the similarity distance of two-dimentional sketch and model be d (sketch, model), then computing formula is as follows;
d(sketch，model)＝min(d(sketch，view
1)，d(sketch，view
2)，d(sketch，view
3))
D (sketch, view wherein
1), d (sketch, view
2), d (sketch, view
3) be respectively the distance between the feature of the same race of 3 views of sketch feature and model.
(b) for according to these two kinds of similarity distances that feature obtains, adopt the quicksort method that the similarity distance ascending order is arranged respectively;
(c) for preceding K forward result for retrieval of precedence in ordering, analyze the model number that belongs to known class in the training set, determine the model classification that the model number is maximum;
(d) if according to these the two kinds of resulting model classification of feature unanimities, be made as classification C, the feature weight that Zernike square and Fourier profile are described is used w respectively
z, w
fExpression, then
Before wherein obtaining according to the Zernike square in K result for retrieval C class model number be k
z, describe according to the Fourier profile that C class model number is k in preceding K the result for retrieval obtaining
f
(e) if inconsistent, then when measuring the similarity distance of sketch and model view, adopt following computing formula with Weighted distance by the model classification that obtains in (c):
w
z＝0.5，w
f＝0.5
So far, the weight that needs when calculating the characteristics combination retrieval, algorithm flow such as Fig. 5.
(f) make up this two kinds of features, calculate the similarity distance of two-dimentional sketch and model, at similarity distance d (sketch, the view of tolerance sketch and model view
i) time adopt Weighted distance, computing formula is as follows:
D wherein
zAnd d
fBe respectively to describe the sketch that obtains and the similarity distance of model view, d according to Zernike square and Fourier profile
ZmaxAnd d
FmaxBe respectively to describe the sketch that obtains and the maximum similarity distance of model view according to Zernike square and Fourier profile.
(g) adopt the quicksort method that the similarity distance ascending order of three-dimensional model after according to weighting arranged.
(3) when connectivity platform is not satisfied in the target area in the sketch, directly calculate similarity distance, adopt the quicksort method that three-dimensional model is sorted according to the Zernike square.
When generating result for retrieval, to the restriction of homepage display model result for retrieval quantity and the requirement of content, the index image of the model that ordering is forward, title, URL etc. return to client according to client.
6. use Java Applet technology to realize the two-dimensional graphics tool set, final interface and a sketch example such as the Fig. 6 that the user draws.Drawing tool collection shown in Figure 6 also provides and has wiped, fills, cancelled previous action, emptied functions such as making graph region except the drafting function that fundamental figure (line segment, rectangle, ellipse, round rectangle) is provided; Make a car for drawing in the graph region with these instruments.
Based on interface such as Fig. 7 of the three-dimensional model searching system of two-dimentional sketch, the left side is two-dimentional sketch drafting zone, and what the right showed is the result for retrieval that returns from server.Provide the people that draws with user retrieval example among Fig. 7 as query requests, what show in the result for retrieval is the index image of three-dimensional model, each width of cloth index image is represented a three-dimensional model, homepage has shown ordering at preceding 20 three-dimensional model, clicks " Next page " and can browse the three-dimensional model that comes the back.
Claims (8)
1,, it is characterized in that realizing by following steps based on the method for searching three-dimension model of two-dimentional sketch:
(1) each model in the three-dimensional modeling data storehouse is handled, obtained the perspective view of three-dimensional model;
(2) with the perspective view of all three-dimensional models, generate the viewdata storehouse;
(3) processing of every width of cloth view process characteristic extracting module in the viewdata storehouse to generation, the generating feature database;
(4) to the Shape Feature Extraction of the two-dimentional sketch of client, comprise Zernike moment characteristics and Fourier boundary descriptor;
(5) two-dimentional sketch feature that client is submitted to and the characteristic matching in the property data base are calculated similarity distance and ordering, finally generate result for retrieval.
2, the method for searching three-dimension model based on two-dimentional sketch according to claim 1 is characterized in that: described step (1) is handled each model in the three-dimensional modeling data storehouse, and the concrete steps of perspective view of obtaining three-dimensional model are as follows:
(1) standardization of model coordinate comprises translation transformation standardization and rotational transform standardization, at first carries out the translation transformation standardization, and the barycenter of computation model moves to true origin with barycenter; Adopt improved continuous P CA method to determine three main shafts of model then, it is rotated the conversion standardization, its coordinate axis and main shaft are overlapped around the initial point rotating model;
(2) adopt the projection of orthogonal projection mode, surround the position that the size of ball is calculated six cutting faces of view frustums according to the minimum of model, the minimum of model is surrounded ball all the time in view frustums when guaranteeing projection;
(3) define the negative direction that direction of visual lines is three coordinate axis successively, preserve content displayed in the window, be three width of cloth perspective views of model.
3, the method for searching three-dimension model based on two-dimentional sketch according to claim 1 is characterized in that: described step (2) is for the perspective view of all three-dimensional models, and the method that generates the viewdata storehouse by the perspective view processing module is as follows:
(1) perspective view is carried out image binaryzation, make that it becomes that prospect is 1, background is 0 binary picture;
(2) image is carried out Close computing in the two-value shape filtering.
4, the method for searching three-dimension model based on two-dimentional sketch according to claim 1 is characterized in that: described step (3) is to every width of cloth view is through the processing of characteristic extracting module in the viewdata storehouse that generates, and the step of generating feature database is as follows:
(1) the Zernike moment characteristics of calculating view;
(2) the Fourier boundary descriptor is calculated in the target area that is fit to, as contour feature.
5, the method for searching three-dimension model based on two-dimentional sketch according to claim 4 is characterized in that: the method for the Zernike moment characteristics of calculating view is as follows in the described step (1):
(a) take Close computing in the two-value shape filtering;
(b) regional barycenter is moved to coordinate origin, edges of regions is zoomed to 1 to the ultimate range of barycenter, this two steps operation is in order to guarantee that the target area feature is not subjected to the influence of translation, transformation of scale;
(c) calculate the Zernike square;
(d) the plural coefficient that obtains is asked the spoke value, the standardization of spoke value is by realizing the spoke value divided by the pixel number in the foreground area;
(e) the most at last the spoke value after the standardization as the Zernike moment characteristics of view.
6, the method for searching three-dimension model based on two-dimentional sketch according to claim 4 is characterized in that: the method for calculating the Fourier boundary descriptor in the described step (2) is as follows:
(a) before the tracing area profile, take Bridge operation and Clean in the two-value shape filtering to operate the connectedness that as far as possible guarantees destination object, Bridge operation with the target area near but disconnected part couple together, the Clean operation can be removed isolated bright spot;
(b) connectedness of judgement target area.For through still disconnected target area behind this two steps filtering operation, do not calculate the Fourier boundary descriptor;
(c) if the target area is communicated with, the outline of tracking target at first then; To the edge pixel sequences, calculate their distances successively then to regional center;
(d) carry out Fourier transform;
(e) at last to obtaining one group of plural number coefficient { a behind the Fourier transform
nCarry out standardization, through standardization obtain | b
n|, n=1 ... N-1}, wherein b
n=a
n/ a
0, as the Fourier boundary descriptor of view.
7, the method for searching three-dimension model based on two-dimentional sketch according to claim 1 is characterized in that: the step of calculating similarity distance in the described step (5) and ordering is as follows:
(1) make up training set, the department pattern in the model database has comprised classification information, selects K typical model and form training set from each classification, and the span of K is other model sum of 1-infima species;
(2) when connectivity platform is satisfied in the target area in the sketch, the sketch feature comprises the Zernike square and the Fourier profile is described two kinds of features, then finds the solution result for retrieval in conjunction with these two kinds of features, mainly needs the weight of calculated characteristics;
(3) when connectivity platform is not satisfied in the target area in the sketch, directly calculate similarity distance, adopt the quicksort method that three-dimensional model is sorted according to the Zernike square.
8, the method for searching three-dimension model based on two-dimentional sketch according to claim 7 is characterized in that: the weight key step of calculated characteristics is in the described step (2):
(a) respectively according to the similarity distance of each model in these two kinds of feature calculation two dimension sketches and the database, adopt the minimum value of the characteristic distance of sketch and 3 views of model to measure, if the similarity distance of two-dimentional sketch and model be d (sketch, model), then computing formula is as follows;
d(sketch，model)＝min(d(sketch，view
1)，d(sketch，view
2)，d(sketch，view
3))
D (sketch, view wherein
1), d (sketch, view
2), d (sketch, view
3) be respectively the distance between the feature of the same race of 3 views of sketch feature and model.
(b) for according to these two kinds of similarity distances that feature obtains, adopt the quicksort method that the similarity distance ascending order is arranged respectively;
(c) for preceding K forward result for retrieval of precedence in ordering, analyze the model number that belongs to known class in the training set, determine the model classification that the model number is maximum;
(d) if according to these the two kinds of resulting model classification of feature unanimities, be made as classification C, the feature weight that Zernike square and Fourier profile are described is used w respectively
z, w
fExpression, then
Before wherein obtaining according to the Zernike square in K result for retrieval C class model number be k
z, describe according to the Fourier profile that C class model number is k in preceding K the result for retrieval obtaining
f
(e) if inconsistent, then when measuring the similarity distance of sketch and model view, adopt following computing formula with Weighted distance by the model classification that obtains in (c):
w
z=0.5，w
f=0.5
The weight that needs when so far, calculating the characteristics combination retrieval;
(f) make up this two kinds of features, calculate the similarity distance of two-dimentional sketch and model, at similarity distance d (sketch, the view of tolerance sketch and each model 3 width of cloth view
i) time adopt Weighted distance, computing formula is as follows:
D wherein
zAnd d
fBe respectively to describe the sketch that obtains and the similarity distance of model view, d according to Zernike square and Fourier profile
ZmaxAnd d
FmaxBe respectively to describe the sketch that obtains and the maximum similarity distance of model view according to Zernike square and Fourier profile;
(g) adopt the quicksort method that the similarity distance ascending order of three-dimensional model after according to weighting arranged.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2006101140974A CN100456300C (en) | 2006-10-27 | 2006-10-27 | Method for searching 3D model based on 2D sketch |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CNB2006101140974A CN100456300C (en) | 2006-10-27 | 2006-10-27 | Method for searching 3D model based on 2D sketch |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN101004748A true CN101004748A (en) | 2007-07-25 |
| CN100456300C CN100456300C (en) | 2009-01-28 |
Family
ID=38703890
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CNB2006101140974A Expired - Fee Related CN100456300C (en) | 2006-10-27 | 2006-10-27 | Method for searching 3D model based on 2D sketch |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN100456300C (en) |
Cited By (51)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2009030124A1 (en) * | 2007-09-06 | 2009-03-12 | Huawei Technologies Co., Ltd. | Method, device, and system for searching multimedia model |
| CN100593785C (en) * | 2008-05-30 | 2010-03-10 | 清华大学 | Three-dimensional model search method based on multiple characteristic related feedback |
| CN101937549A (en) * | 2010-10-09 | 2011-01-05 | 姚建 | Network shopping guidance system |
| CN101950400A (en) * | 2010-10-09 | 2011-01-19 | 姚建 | Network shopping guiding method |
| CN101951398A (en) * | 2010-09-08 | 2011-01-19 | 檀朝东 | Engineering calculation method, client side, server side and system |
| CN102073738A (en) * | 2011-01-20 | 2011-05-25 | 清华大学 | Intelligent retrieval view selection-based three-dimensional object retrieval method and device |
| CN102110159A (en) * | 2011-02-21 | 2011-06-29 | 清华大学 | CAD three-dimensional model retrieval method and system |
| CN102163343A (en) * | 2011-04-11 | 2011-08-24 | 西安交通大学 | Three-dimensional model optimal viewpoint automatic obtaining method based on internet image |
| CN102221963A (en) * | 2010-04-16 | 2011-10-19 | 卡西欧计算机株式会社 | Display apparatus and computer readable medium |
| CN101196930B (en) * | 2008-01-04 | 2012-01-04 | 覃征 | Three-dimensional model searching system |
| CN101201845B (en) * | 2007-09-24 | 2012-01-11 | 覃征 | Method for searching three-dimensional model based on axis point-set hierarchical helix information |
| CN101373481B (en) * | 2007-08-24 | 2012-06-27 | 株式会社理光 | Image search apparatus, and image search method |
| CN102667855A (en) * | 2009-10-19 | 2012-09-12 | Metaio有限公司 | Method for determining the pose of a camera and for recognizing an object of a real environment |
| CN101398896B (en) * | 2007-09-28 | 2012-10-17 | 三星电子株式会社 | Device and method for extracting color characteristic with strong discernment for image forming apparatus |
| CN102810160A (en) * | 2012-06-06 | 2012-12-05 | 北京京东世纪贸易有限公司 | Method and device for searching images |
| CN103295025A (en) * | 2013-05-03 | 2013-09-11 | 南京大学 | Automatic selecting method of three-dimensional model optimal view |
| CN103460252A (en) * | 2011-03-03 | 2013-12-18 | 南洋理工大学 | A method for reconstructing a 3d surface from a 2d sketch |
| CN103902657A (en) * | 2014-03-03 | 2014-07-02 | 浙江大学 | Three-dimensional model retrieval method based on sketch |
| CN103914868A (en) * | 2013-12-20 | 2014-07-09 | 柳州腾龙煤电科技股份有限公司 | Method for mass model data dynamic scheduling and real-time asynchronous loading under virtual reality |
| CN104520850A (en) * | 2012-08-06 | 2015-04-15 | 微软公司 | Three-dimensional object browsing in documents |
| WO2015109992A1 (en) * | 2014-01-22 | 2015-07-30 | Tencent Technology (Shenzhen) Company Limited | Information retrieval method, apparatus and system |
| CN105006023A (en) * | 2015-08-04 | 2015-10-28 | 北京航空航天大学 | Multi-category three-dimension model combination modeling method supporting aided design |
| CN105243137A (en) * | 2015-09-30 | 2016-01-13 | 华南理工大学 | Draft-based three-dimensional model retrieval viewpoint selection method |
| CN105243139A (en) * | 2015-10-10 | 2016-01-13 | 天津大学 | Deep learning based three-dimensional model retrieval method and retrieval device thereof |
| CN105320764A (en) * | 2015-10-26 | 2016-02-10 | 天津大学 | 3D model retrieval method and 3D model retrieval apparatus based on slow increment features |
| CN105447205A (en) * | 2016-01-05 | 2016-03-30 | 腾讯科技（深圳）有限公司 | Retrieved result sorting method and device |
| CN105631932A (en) * | 2015-12-22 | 2016-06-01 | 北京航空航天大学 | Three-dimensional model re-construction method with contour line guidance |
| CN106484692A (en) * | 2015-08-25 | 2017-03-08 | 北京师范大学 | A kind of method for searching three-dimension model |
| CN107122396A (en) * | 2017-03-13 | 2017-09-01 | 西北大学 | Three-dimensional model searching algorithm based on depth convolutional neural networks |
| CN107209631A (en) * | 2015-01-09 | 2017-09-26 | 三星电子株式会社 | User terminal and its method for displaying image for display image |
| CN107688599A (en) * | 2017-07-05 | 2018-02-13 | 国网山东省电力公司泰安供电公司 | A kind of method of quick-searching threedimensional model |
| CN108416028A (en) * | 2018-03-09 | 2018-08-17 | 北京百度网讯科技有限公司 | A kind of method, apparatus and server of search content resource |
| CN108537887A (en) * | 2018-04-18 | 2018-09-14 | 北京航空航天大学 | Sketch based on 3D printing and model library 3-D view matching process |
| CN108875080A (en) * | 2018-07-12 | 2018-11-23 | 百度在线网络技术（北京）有限公司 | A kind of image search method, device, server and storage medium |
| CN108921045A (en) * | 2018-06-11 | 2018-11-30 | 佛山科学技术学院 | A kind of space characteristics of threedimensional model extract and matching process and its device |
| CN108932258A (en) * | 2017-05-25 | 2018-12-04 | 华为技术有限公司 | Data directory processing method and processing device |
| CN109213884A (en) * | 2018-11-26 | 2019-01-15 | 北方民族大学 | A kind of cross-module state search method based on Sketch Searching threedimensional model |
| CN109857886A (en) * | 2018-12-28 | 2019-06-07 | 浙江工业大学 | A kind of method for searching three-dimension model approached based on minimax value theory of games view |
| CN110019914A (en) * | 2018-07-18 | 2019-07-16 | 王斌 | A kind of three-dimensional modeling data storehouse search method for supporting three-dimensional scenic interaction |
| US10380175B2 (en) | 2017-06-06 | 2019-08-13 | International Business Machines Corporation | Sketch-based image retrieval using feedback and hierarchies |
| CN110134693A (en) * | 2019-05-17 | 2019-08-16 | 南京大学 | Temporal index method for building up based on Hash and PCA |
| CN110162655A (en) * | 2019-04-30 | 2019-08-23 | 厦门大学 | A kind of threedimensional model partial retrieval, device, terminal device based on sketch |
| CN110188228A (en) * | 2019-05-28 | 2019-08-30 | 北方民族大学 | Cross-module state search method based on Sketch Searching threedimensional model |
| CN110569386A (en) * | 2019-09-16 | 2019-12-13 | 哈尔滨理工大学 | Three-dimensional model retrieval method based on hand-drawn sketch integrated descriptor |
| CN110704653A (en) * | 2019-09-09 | 2020-01-17 | 上海慧之建建设顾问有限公司 | Method for searching component by graph in BIM (building information modeling) model and graph-text searching system |
| CN110737788A (en) * | 2019-10-16 | 2020-01-31 | 哈尔滨理工大学 | Rapid three-dimensional model index establishing and retrieving method |
| CN111328396A (en) * | 2017-11-15 | 2020-06-23 | 高通科技公司 | Pose estimation and model retrieval for objects in images |
| US10739142B2 (en) | 2016-09-02 | 2020-08-11 | Apple Inc. | System for determining position both indoor and outdoor |
| CN113032613A (en) * | 2021-03-12 | 2021-06-25 | 哈尔滨理工大学 | Three-dimensional model retrieval method based on interactive attention convolution neural network |
| US11244517B2 (en) | 2015-02-17 | 2022-02-08 | Samsung Electronics Co., Ltd. | Device for generating printing information and method for generating printing information |
| CN114020953A (en) * | 2021-10-27 | 2022-02-08 | 北京中知智慧科技有限公司 | Multi-image retrieval method and device for appearance design product |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11735306B2 (en) | 2019-11-25 | 2023-08-22 | Dentsply Sirona Inc. | Method, system and computer readable storage media for creating three-dimensional dental restorations from two dimensional sketches |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3498643B2 (en) * | 1999-09-10 | 2004-02-16 | 日本電気株式会社 | Three-dimensional video identification device and computer-readable recording medium |
| EP1429264B1 (en) * | 2002-12-05 | 2016-06-29 | Samsung Electronics Co., Ltd. | Method of perceptual 3D shape description and method and apparatus for searching 3D graphics model database using the description method |
| JP4693454B2 (en) * | 2005-03-29 | 2011-06-01 | 富士通株式会社 | 3D shape comparison program and 3D similar shape search program |
-
2006
- 2006-10-27 CN CNB2006101140974A patent/CN100456300C/en not_active Expired - Fee Related
Cited By (93)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101373481B (en) * | 2007-08-24 | 2012-06-27 | 株式会社理光 | Image search apparatus, and image search method |
| US8082263B2 (en) | 2007-09-06 | 2011-12-20 | Huawei Technologies Co., Ltd. | Method, apparatus and system for multimedia model retrieval |
| CN101382934B (en) * | 2007-09-06 | 2010-08-18 | 华为技术有限公司 | Search method for multimedia model, apparatus and system |
| WO2009030124A1 (en) * | 2007-09-06 | 2009-03-12 | Huawei Technologies Co., Ltd. | Method, device, and system for searching multimedia model |
| CN101201845B (en) * | 2007-09-24 | 2012-01-11 | 覃征 | Method for searching three-dimensional model based on axis point-set hierarchical helix information |
| CN101398896B (en) * | 2007-09-28 | 2012-10-17 | 三星电子株式会社 | Device and method for extracting color characteristic with strong discernment for image forming apparatus |
| CN101196930B (en) * | 2008-01-04 | 2012-01-04 | 覃征 | Three-dimensional model searching system |
| CN100593785C (en) * | 2008-05-30 | 2010-03-10 | 清华大学 | Three-dimensional model search method based on multiple characteristic related feedback |
| US9218665B2 (en) | 2009-10-19 | 2015-12-22 | Metaio Gmbh | Method for determining the pose of a camera and for recognizing an object of a real environment |
| CN106650723A (en) * | 2009-10-19 | 2017-05-10 | Metaio有限公司 | Method for determining the pose of a camera and for recognizing an object of a real environment |
| US8942418B2 (en) | 2009-10-19 | 2015-01-27 | Metaio Gmbh | Method of providing a descriptor for at least one feature of an image and method of matching features |
| US10650546B2 (en) | 2009-10-19 | 2020-05-12 | Apple Inc. | Method of providing a descriptor for at least one feature of an image and method of matching features |
| US8837779B2 (en) | 2009-10-19 | 2014-09-16 | Metaio Gmbh | Method for determining the pose of a camera and for recognizing an object of a real environment |
| CN102667855A (en) * | 2009-10-19 | 2012-09-12 | Metaio有限公司 | Method for determining the pose of a camera and for recognizing an object of a real environment |
| US10580162B2 (en) | 2009-10-19 | 2020-03-03 | Apple Inc. | Method for determining the pose of a camera and for recognizing an object of a real environment |
| US10062169B2 (en) | 2009-10-19 | 2018-08-28 | Apple Inc. | Method of providing a descriptor for at least one feature of an image and method of matching features |
| US10229511B2 (en) | 2009-10-19 | 2019-03-12 | Apple Inc. | Method for determining the pose of a camera and for recognizing an object of a real environment |
| US8487934B2 (en) | 2010-04-16 | 2013-07-16 | Casio Computer Co., Ltd. | Display apparatus and computer readable medium |
| CN102221963A (en) * | 2010-04-16 | 2011-10-19 | 卡西欧计算机株式会社 | Display apparatus and computer readable medium |
| CN102221963B (en) * | 2010-04-16 | 2013-07-03 | 卡西欧计算机株式会社 | Display apparatus |
| CN101951398A (en) * | 2010-09-08 | 2011-01-19 | 檀朝东 | Engineering calculation method, client side, server side and system |
| CN101950400A (en) * | 2010-10-09 | 2011-01-19 | 姚建 | Network shopping guiding method |
| CN101937549B (en) * | 2010-10-09 | 2014-04-16 | 姚建 | Picture guidance system for network shopping guidance |
| CN101950400B (en) * | 2010-10-09 | 2014-04-16 | 姚建 | Picture retrieving method of network shopping guiding method |
| CN101937549A (en) * | 2010-10-09 | 2011-01-05 | 姚建 | Network shopping guidance system |
| CN102073738B (en) * | 2011-01-20 | 2013-04-17 | 清华大学 | Intelligent retrieval view selection-based three-dimensional object retrieval method and device |
| CN102073738A (en) * | 2011-01-20 | 2011-05-25 | 清华大学 | Intelligent retrieval view selection-based three-dimensional object retrieval method and device |
| CN102110159B (en) * | 2011-02-21 | 2013-05-22 | 清华大学 | CAD three-dimensional model retrieval method and system |
| CN102110159A (en) * | 2011-02-21 | 2011-06-29 | 清华大学 | CAD three-dimensional model retrieval method and system |
| CN103460252B (en) * | 2011-03-03 | 2017-02-22 | 南洋理工大学 | A method for reconstructing a 3d surface from a 2d sketch |
| CN103460252A (en) * | 2011-03-03 | 2013-12-18 | 南洋理工大学 | A method for reconstructing a 3d surface from a 2d sketch |
| CN102163343A (en) * | 2011-04-11 | 2011-08-24 | 西安交通大学 | Three-dimensional model optimal viewpoint automatic obtaining method based on internet image |
| CN102163343B (en) * | 2011-04-11 | 2013-11-06 | 西安交通大学 | Three-dimensional model optimal viewpoint automatic obtaining method based on internet image |
| CN102810160A (en) * | 2012-06-06 | 2012-12-05 | 北京京东世纪贸易有限公司 | Method and device for searching images |
| CN104520850A (en) * | 2012-08-06 | 2015-04-15 | 微软公司 | Three-dimensional object browsing in documents |
| CN104520850B (en) * | 2012-08-06 | 2018-08-07 | 微软技术许可有限责任公司 | Three dimensional object browses in document |
| CN103295025A (en) * | 2013-05-03 | 2013-09-11 | 南京大学 | Automatic selecting method of three-dimensional model optimal view |
| CN103295025B (en) * | 2013-05-03 | 2016-06-15 | 南京大学 | A kind of automatic selecting method of three-dimensional model optimal view |
| CN103914868A (en) * | 2013-12-20 | 2014-07-09 | 柳州腾龙煤电科技股份有限公司 | Method for mass model data dynamic scheduling and real-time asynchronous loading under virtual reality |
| CN103914868B (en) * | 2013-12-20 | 2017-02-22 | 柳州腾龙煤电科技股份有限公司 | Method for mass model data dynamic scheduling and real-time asynchronous loading under virtual reality |
| WO2015109992A1 (en) * | 2014-01-22 | 2015-07-30 | Tencent Technology (Shenzhen) Company Limited | Information retrieval method, apparatus and system |
| CN103902657B (en) * | 2014-03-03 | 2017-04-19 | 浙江大学 | Three-dimensional model retrieval method based on sketch |
| CN103902657A (en) * | 2014-03-03 | 2014-07-02 | 浙江大学 | Three-dimensional model retrieval method based on sketch |
| CN107209631A (en) * | 2015-01-09 | 2017-09-26 | 三星电子株式会社 | User terminal and its method for displaying image for display image |
| US11244517B2 (en) | 2015-02-17 | 2022-02-08 | Samsung Electronics Co., Ltd. | Device for generating printing information and method for generating printing information |
| CN105006023B (en) * | 2015-08-04 | 2018-01-19 | 北京航空航天大学 | A kind of multi-class threedimensional model compositional modeling method for supporting Computer Aided Design |
| CN105006023A (en) * | 2015-08-04 | 2015-10-28 | 北京航空航天大学 | Multi-category three-dimension model combination modeling method supporting aided design |
| CN106484692A (en) * | 2015-08-25 | 2017-03-08 | 北京师范大学 | A kind of method for searching three-dimension model |
| CN106484692B (en) * | 2015-08-25 | 2019-11-29 | 北京师范大学 | A kind of method for searching three-dimension model |
| CN105243137A (en) * | 2015-09-30 | 2016-01-13 | 华南理工大学 | Draft-based three-dimensional model retrieval viewpoint selection method |
| CN105243137B (en) * | 2015-09-30 | 2018-12-11 | 华南理工大学 | A kind of three-dimensional model search viewpoint selection method based on sketch |
| CN105243139B (en) * | 2015-10-10 | 2018-10-23 | 天津大学 | A kind of method for searching three-dimension model and its retrieval device based on deep learning |
| CN105243139A (en) * | 2015-10-10 | 2016-01-13 | 天津大学 | Deep learning based three-dimensional model retrieval method and retrieval device thereof |
| CN105320764B (en) * | 2015-10-26 | 2019-05-07 | 天津大学 | A kind of 3D model retrieval method and its retrieval device based on the slow feature of increment |
| CN105320764A (en) * | 2015-10-26 | 2016-02-10 | 天津大学 | 3D model retrieval method and 3D model retrieval apparatus based on slow increment features |
| CN105631932A (en) * | 2015-12-22 | 2016-06-01 | 北京航空航天大学 | Three-dimensional model re-construction method with contour line guidance |
| US10049503B2 (en) | 2015-12-22 | 2018-08-14 | Beihang University | Line guided three-dimensional model reshaping method |
| CN105631932B (en) * | 2015-12-22 | 2018-06-12 | 北京航空航天大学 | A kind of three-dimensional model reconfiguration of contour line guidance makes method |
| CN105447205A (en) * | 2016-01-05 | 2016-03-30 | 腾讯科技（深圳）有限公司 | Retrieved result sorting method and device |
| CN105447205B (en) * | 2016-01-05 | 2023-10-24 | 腾讯科技（深圳）有限公司 | Method and device for sorting search results |
| US10739142B2 (en) | 2016-09-02 | 2020-08-11 | Apple Inc. | System for determining position both indoor and outdoor |
| US11859982B2 (en) | 2016-09-02 | 2024-01-02 | Apple Inc. | System for determining position both indoor and outdoor |
| CN107122396B (en) * | 2017-03-13 | 2019-10-29 | 西北大学 | Method for searching three-dimension model based on depth convolutional neural networks |
| CN107122396A (en) * | 2017-03-13 | 2017-09-01 | 西北大学 | Three-dimensional model searching algorithm based on depth convolutional neural networks |
| CN108932258A (en) * | 2017-05-25 | 2018-12-04 | 华为技术有限公司 | Data directory processing method and processing device |
| US10380175B2 (en) | 2017-06-06 | 2019-08-13 | International Business Machines Corporation | Sketch-based image retrieval using feedback and hierarchies |
| CN107688599B (en) * | 2017-07-05 | 2018-11-16 | 国网山东省电力公司 | A kind of method of quick-searching threedimensional model |
| CN107688599A (en) * | 2017-07-05 | 2018-02-13 | 国网山东省电力公司泰安供电公司 | A kind of method of quick-searching threedimensional model |
| CN111328396A (en) * | 2017-11-15 | 2020-06-23 | 高通科技公司 | Pose estimation and model retrieval for objects in images |
| CN108416028B (en) * | 2018-03-09 | 2021-09-21 | 北京百度网讯科技有限公司 | Method, device and server for searching content resources |
| CN108416028A (en) * | 2018-03-09 | 2018-08-17 | 北京百度网讯科技有限公司 | A kind of method, apparatus and server of search content resource |
| CN108537887A (en) * | 2018-04-18 | 2018-09-14 | 北京航空航天大学 | Sketch based on 3D printing and model library 3-D view matching process |
| CN108921045B (en) * | 2018-06-11 | 2021-08-03 | 佛山科学技术学院 | Spatial feature extraction and matching method and device of three-dimensional model |
| CN108921045A (en) * | 2018-06-11 | 2018-11-30 | 佛山科学技术学院 | A kind of space characteristics of threedimensional model extract and matching process and its device |
| CN108875080A (en) * | 2018-07-12 | 2018-11-23 | 百度在线网络技术（北京）有限公司 | A kind of image search method, device, server and storage medium |
| CN110019914A (en) * | 2018-07-18 | 2019-07-16 | 王斌 | A kind of three-dimensional modeling data storehouse search method for supporting three-dimensional scenic interaction |
| CN109213884B (en) * | 2018-11-26 | 2021-10-19 | 北方民族大学 | Cross-modal retrieval method based on sketch retrieval three-dimensional model |
| CN109213884A (en) * | 2018-11-26 | 2019-01-15 | 北方民族大学 | A kind of cross-module state search method based on Sketch Searching threedimensional model |
| CN109857886B (en) * | 2018-12-28 | 2020-12-15 | 浙江工业大学 | Three-dimensional model retrieval method based on minimum maximum value game theory view approximation |
| CN109857886A (en) * | 2018-12-28 | 2019-06-07 | 浙江工业大学 | A kind of method for searching three-dimension model approached based on minimax value theory of games view |
| CN110162655A (en) * | 2019-04-30 | 2019-08-23 | 厦门大学 | A kind of threedimensional model partial retrieval, device, terminal device based on sketch |
| CN110162655B (en) * | 2019-04-30 | 2021-07-27 | 厦门大学 | Sketch-based three-dimensional model local retrieval method and device and terminal equipment |
| CN110134693A (en) * | 2019-05-17 | 2019-08-16 | 南京大学 | Temporal index method for building up based on Hash and PCA |
| CN110134693B (en) * | 2019-05-17 | 2023-05-05 | 南京大学 | Space-time index establishing method based on hash and PCA |
| CN110188228A (en) * | 2019-05-28 | 2019-08-30 | 北方民族大学 | Cross-module state search method based on Sketch Searching threedimensional model |
| CN110188228B (en) * | 2019-05-28 | 2021-07-02 | 北方民族大学 | Cross-modal retrieval method based on sketch retrieval three-dimensional model |
| CN110704653A (en) * | 2019-09-09 | 2020-01-17 | 上海慧之建建设顾问有限公司 | Method for searching component by graph in BIM (building information modeling) model and graph-text searching system |
| CN110569386A (en) * | 2019-09-16 | 2019-12-13 | 哈尔滨理工大学 | Three-dimensional model retrieval method based on hand-drawn sketch integrated descriptor |
| CN110737788A (en) * | 2019-10-16 | 2020-01-31 | 哈尔滨理工大学 | Rapid three-dimensional model index establishing and retrieving method |
| CN110737788B (en) * | 2019-10-16 | 2022-05-31 | 哈尔滨理工大学 | Rapid three-dimensional model index establishing and retrieving method |
| CN113032613B (en) * | 2021-03-12 | 2022-11-08 | 哈尔滨理工大学 | Three-dimensional model retrieval method based on interactive attention convolution neural network |
| CN113032613A (en) * | 2021-03-12 | 2021-06-25 | 哈尔滨理工大学 | Three-dimensional model retrieval method based on interactive attention convolution neural network |
| CN114020953A (en) * | 2021-10-27 | 2022-02-08 | 北京中知智慧科技有限公司 | Multi-image retrieval method and device for appearance design product |
Also Published As
| Publication number | Publication date |
|---|---|
| CN100456300C (en) | 2009-01-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN100456300C (en) | Method for searching 3D model based on 2D sketch | |
| Zhou et al. | Semantic understanding of scenes through the ade20k dataset | |
| Lai et al. | Fast mesh segmentation using random walks | |
| CN109213884B (en) | Cross-modal retrieval method based on sketch retrieval three-dimensional model | |
| CN104850633B (en) | A kind of three-dimensional model searching system and method based on the segmentation of cartographical sketching component | |
| Papadakis et al. | PANORAMA: A 3D shape descriptor based on panoramic views for unsupervised 3D object retrieval | |
| CN101477529B (en) | Three-dimensional object retrieval method and apparatus | |
| CN110069656B (en) | Method for searching three-dimensional model based on two-dimensional picture of generated countermeasure network | |
| CN101201845B (en) | Method for searching three-dimensional model based on axis point-set hierarchical helix information | |
| Tam et al. | Deformable model retrieval based on topological and geometric signatures | |
| CN108170823B (en) | Hand-drawn interactive three-dimensional model retrieval method based on high-level semantic attribute understanding | |
| CN108537887A (en) | Sketch based on 3D printing and model library 3-D view matching process | |
| JP2002288687A (en) | Device and method for calculating feature amount | |
| CN113330455A (en) | Finding complementary digital images using conditional generative countermeasure networks | |
| Li et al. | SHREC'16 Track: 3D Sketch-Based 3D Shape Retrieval | |
| Bae et al. | Content-based 3D model retrieval using a single depth image from a low-cost 3D camera | |
| CN111354076A (en) | Single-image three-dimensional part combined modeling method based on embedding space | |
| Álvarez et al. | Junction assisted 3d pose retrieval of untextured 3d models in monocular images | |
| Cailliere et al. | 3D mirror symmetry detection using Hough transform | |
| Lei et al. | A novel sketch-based 3D model retrieval method by integrating skeleton graph and contour feature | |
| CN111382301B (en) | Three-dimensional model generation method and system based on generation countermeasure network | |
| CN109767496B (en) | Network virtual assembly method based on symmetric constraint | |
| Liang et al. | Decorin: An automatic method for plane-based decorating | |
| Hlavatý et al. | A survey for methods for 3D model feature extraction | |
| CN111339332A (en) | Three-dimensional volume data retrieval method based on tree structure topological graph |
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 | ||
| C17 | Cessation of patent right | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20090128 Termination date: 20131027 |