WO2015172504A1 - 一种构建三维物体的方法及计算机辅助设计系统 - Google Patents

一种构建三维物体的方法及计算机辅助设计系统 Download PDF

Info

Publication number
WO2015172504A1
WO2015172504A1 PCT/CN2014/088311 CN2014088311W WO2015172504A1 WO 2015172504 A1 WO2015172504 A1 WO 2015172504A1 CN 2014088311 W CN2014088311 W CN 2014088311W WO 2015172504 A1 WO2015172504 A1 WO 2015172504A1
Authority
WO
WIPO (PCT)
Prior art keywords
line drawing
dimensional
loop
dimensional line
auxiliary
Prior art date
Application number
PCT/CN2014/088311
Other languages
English (en)
French (fr)
Inventor
刘健庄
鞠汶奇
许春景
Original Assignee
华为技术有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Publication of WO2015172504A1 publication Critical patent/WO2015172504A1/zh

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00Computer-aided design [CAD]
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T17/00Three dimensional [3D] modelling, e.g. data description of 3D objects
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06TIMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T19/00Manipulating 3D models or images for computer graphics

Definitions

  • the invention relates to the field of computer vision, and in particular to a method for constructing a three-dimensional object and a computer aided design system.
  • CAD Computer Aided Design
  • WIMP mode Windows, Icon Icon, Menu Menu, Mouse Pointer
  • the user needs to Select different drawing tools in the menu, panel or icon of the user interaction interface, and select various parameters of the object to be drawn in the interactive window, and finally select a drawing position, a size ratio, etc. of the object by a series of operations such as clicking, dragging, and the like.
  • CAD system can provide powerful 3D object design and editing capabilities, but the operation is more complicated, only the professionals who are very familiar with the complicated menus in the design system can complete the design well.
  • using this operating system to complete the 3D object is cumbersome and time consuming. Even if you design a simple object, it takes a long time.
  • the SBIM mode (sketch-based interfaces for modeling) is used to reconstruct a three-dimensional object, wherein the sketch is a two-dimensional line drawing, and the operation interface is simple and easy to understand, and the user only needs to draw and
  • the two-dimensional line drawing (sketch) corresponding to the three-dimensional object the system can automatically generate the three-dimensional model required by the user, because the same two-dimensional line drawing corresponds to an infinite number A three-dimensional object, so the design difficulty of this kind of system is how to reconstruct the three-dimensional object that the user needs from the two-dimensional line drawing given by the user.
  • the traditional approach is to reconstruct a three-dimensional object from line drawing into an optimization problem, and use the optimization method to find a three-dimensional object that conforms to human visual perception.
  • the traditional system can only process simple or partially complex line drawing when processing. When the line drawing is more complicated or very complicated, the system will fall into local optimum due to the rapid expansion of the solution space, resulting in inaccurate 3D reconstruction.
  • the embodiment of the invention provides a method for constructing a three-dimensional object and a computer-aided design system, which processes a two-dimensional line drawing, and segments a simpler sub-line drawing conforming to the visual perception rule, and completes the reconstruction of the three-dimensional object efficiently and accurately.
  • an embodiment of the present invention provides a method for constructing a three-dimensional object, including: acquiring a two-dimensional line drawing; dividing the two-dimensional line drawing into a plurality of sub-line drawing; and performing three-dimensional reconstruction on each sub-line drawing separately; Reconstructing the three-dimensional reconstruction result of the sub-line drawing to obtain a three-dimensional object, wherein the dividing the two-dimensional line drawing into a plurality of sub-line drawing, specifically includes:
  • Extracting a surface of the two-dimensional line drawing wherein the surface is an area surrounded by an edge of the same plane in the two-dimensional line drawing, and is a convex surface or a concave surface;
  • the two-dimensional line drawing includes a concave surface, adding an auxiliary vertex on the two-dimensional line drawing according to a preset strategy, and an auxiliary side to obtain an extended two-dimensional line drawing;
  • the two-dimensional line drawing includes a concave surface, adding auxiliary vertices on the two-dimensional line drawing according to a preset strategy, and assisting Edge, get extended 2D line drawing, including:
  • auxiliary edge Adding a first type of auxiliary edge to each concave surface of the two-dimensional line drawing, wherein the first type of auxiliary side is parallel to either side of the concave surface, and one end point of the first type of auxiliary side is a concave point of the concave surface, and another end point of the first type of auxiliary side is An auxiliary vertex, the auxiliary vertex being an intersection with an edge of the concave surface and the first type of auxiliary line;
  • the first concave surface comprises a collinear edge
  • a third type of auxiliary edge is added on the first concave surface, wherein the third type auxiliary edge is a collinear side of the first concave surface The connection between.
  • the extended two-dimensional line drawing Obtaining a split surface of the two-dimensional line drawing, including:
  • the first edge is classified into the first edge set
  • the preset criterion includes: if the loop includes a degree 3D 3D bump, then the loop is not a loop of the split plane; if the loop passes through two adjacent faces f1, f2, and f1, f2 jointly have two or more sides that are not on the same straight line, then The loop is not a loop of the split plane; if the edges of the loop self-intersect, the loop is not a loop of the split plane; if the loop contains a string and the string is on the surface of the two-dimensional line drawing, the loop is not a split plane a loop; if the loop contains two non-collinear edges, and the two edges are the sides of the same face f, and there is an overlap between the region of the loop and the f, then the loop is not a split plane The loop.
  • the method further includes:
  • the segmentation surface having the smallest number of edges or the segmentation surface having the shortest circumference is selected as the pit point The split face.
  • the The line drawing is performed to obtain a plurality of sub-line drawing of the two-dimensional line drawing, including:
  • a surface having a common side with the dividing surface is classified into a first surface set
  • the second surface In the first set of faces, if there is a second surface, there is a common edge with the surface in the second set of faces, and the common edge is connected to one vertex of the split face, and the common edge is not An edge belonging to the dividing surface, the second surface is deleted from the first surface set, and is classified into a second surface set;
  • an embodiment of the present invention provides a computer-aided design system, including: an acquisition module, configured to acquire a two-dimensional line drawing; and a segmentation module, configured to The figure is divided into a plurality of sub-line drawing; the three-dimensional reconstruction module is configured to perform three-dimensional reconstruction on each sub-line drawing separately; and a recombining module is configured to recombine the three-dimensional reconstruction result of the sub-line drawing to obtain a three-dimensional object, and the feature thereof
  • the dividing module comprises:
  • a concave surface acquiring unit configured to extract a surface of the two-dimensional line drawing, wherein the surface is an area surrounded by an edge of the same plane in the two-dimensional line drawing, and is a convex surface or a concave surface;
  • Adding a unit if the two-dimensional line drawing includes a concave surface, adding an auxiliary vertex on the two-dimensional line drawing according to a preset strategy, and an auxiliary side to obtain an extended two-dimensional line drawing;
  • a segmentation plane acquiring unit configured to acquire a segmentation surface of the two-dimensional line drawing according to the extended two-dimensional line drawing
  • the sub-picture acquisition unit performs segmentation on the two-dimensional line drawing according to the segmentation, and acquires a plurality of sub-line drawing of the two-dimensional line drawing.
  • the adding unit is specifically configured to:
  • auxiliary edge Adding a first type of auxiliary edge to each concave surface of the two-dimensional line drawing, wherein the first type of auxiliary side is parallel to either side of the concave surface, and one end point of the first type of auxiliary side is a concave point of the concave surface, another end point of the first type of auxiliary side is an auxiliary vertex, and the auxiliary vertex is an intersection with an edge of the concave surface and the auxiliary line of the first type;
  • the adding unit is specifically configured to:
  • the first concave surface comprises a collinear edge
  • a third type of auxiliary edge is added on the first concave surface, wherein the third type auxiliary edge is a collinear side of the first concave surface The connection between.
  • the split plane acquiring unit is specifically configured to:
  • the first edge is classified into the first edge set
  • the preset criterion includes: if the loop includes a degree 3D 3D bump, then the loop is not a loop of the split plane; if the loop passes through two adjacent faces f1, f2, and f1, f2 jointly have two or more sides that are not on the same straight line, then The loop is not a loop of the split plane; if the edges of the loop self-intersect, the loop is not a loop of the split plane; if the loop contains a string and the string is on the surface of the two-dimensional line drawing, the loop is not a split plane a loop; if the loop contains two non-collinear edges, and the two edges are the sides of the same face f, and there is an overlap between the region of the loop and the f, then the loop is not a split plane The loop.
  • the split plane acquiring unit is further configured to:
  • the segmentation surface having the smallest number of edges or the segmentation surface having the shortest circumference is selected as the pit point The split face.
  • the sub-graph obtaining unit is specifically configured to:
  • a surface having a common side with the dividing surface is classified into a first surface set
  • the second surface In the first set of faces, if there is a second surface, there is a common edge with the surface in the second set of faces, and the common edge is connected to one vertex of the split face, and the common edge is not An edge belonging to the dividing surface, the second surface is deleted from the first surface set, and is classified into a second surface set;
  • an embodiment of the present invention provides a method for constructing a three-dimensional object and a computer-aided design system, which acquires a two-dimensional line drawing; divides the two-dimensional line drawing into a plurality of sub-line drawing; and respectively performs three-dimensional drawing on each sub-line And reconstructing the three-dimensional reconstruction result of the sub-line drawing to obtain a three-dimensional object, wherein the dividing the two-dimensional line drawing into a plurality of sub-line drawing comprises: extracting a surface of the two-dimensional line drawing Wherein the surface is an area surrounded by edges of the same plane in the two-dimensional line drawing, which is a convex surface or a concave surface; if the two-dimensional line drawing includes a concave surface, the two-dimensional line drawing is performed according to a preset strategy.
  • Adding an auxiliary vertex, and an auxiliary side acquiring an extended two-dimensional line drawing; acquiring a dividing surface of the two-dimensional line drawing according to the extended two-dimensional line drawing; facing the two-dimensional line according to the dividing The drawing is performed to obtain a plurality of sub-line drawing of the two-dimensional line drawing.
  • the two-dimensional line drawing is processed, and the simpler and visually-aware sub-line drawing is segmented, and the reconstruction of the three-dimensional object is completed efficiently and accurately; and the line drawing processed in the prior art is avoided when the drawing is complicated or complicated.
  • the system falls into local optimum due to the rapid expansion of the solution space, resulting in inaccurate defects in 3D reconstruction.
  • FIG. 1 is a schematic diagram of a basic flow of constructing a three-dimensional object based on a two-dimensional line drawing
  • FIG. 2 is a schematic structural view of a convex surface and a concave surface according to an embodiment of the present invention
  • FIG. 3 is a specific flow chart of dividing a two-dimensional line drawing into a plurality of sub-line drawing in real time according to the present invention
  • FIG. 4 is a schematic flowchart of dividing a two-dimensional line drawing 4a into a plurality of sub-line drawing according to an embodiment of the present invention
  • FIG. 5 is a schematic flowchart of dividing a two-dimensional line drawing 5a into a plurality of sub-line drawing according to an embodiment of the present invention
  • FIG. 6 is a schematic flowchart of acquiring a divided surface of a two-dimensional line drawing 6a and acquiring a plurality of sub-line drawing according to the dividing surface according to an embodiment of the present disclosure
  • FIG. 7 is a schematic flowchart of dividing a two-dimensional line drawing 7a into a plurality of sub-line drawing according to an embodiment of the present invention
  • FIG. 8 is a structural diagram of a computer-aided design system 80 according to an embodiment of the present invention.
  • FIG. 9 is a structural diagram of a segmentation module 802 according to an embodiment of the present invention.
  • the SBIM mode (sketch-based interfaces for modeling) can realize the reconstruction of three-dimensional objects, and the operation is simple and easy to understand. See Figure 1 for a three-dimensional object based on two-dimensional line drawing.
  • Basic flow The intent, as shown in FIG. 1 , may include the steps of: acquiring a two-dimensional line drawing; dividing the two-dimensional line drawing into a plurality of sub-line drawing; performing three-dimensional reconstruction on each sub-line drawing separately; The three-dimensional reconstruction results of the drawing are reconstructed to obtain a three-dimensional object.
  • the two-dimensional line drawing is a plan view composed of straight line segments, which can be represented by a two-dimensional line drawing including vertices and edges; the sides and vertices form a two-dimensional line drawing face, and the two-dimensional line drawing is composed.
  • the face may be a convex or concave surface; all the vertices on the convex face are on the convex surface of the face, all the vertices on the convex surface are bumps, and at least one vertex on the concave face is not convex on the face a surface on the shell, a point on the concave surface that is not on the convex shell of the concave surface is a pit, and a point on the convex shell of the concave surface is a bump; for example, as shown in FIG. 2, which is provided by an embodiment of the present invention Schematic diagram of the convex and concave surfaces, as shown in FIG.
  • 2a is a convex surface
  • 2b is a concave surface
  • the vertex A is a convex point
  • the vertex B is a concave point.
  • the user or the designer can draw a two-dimensional line drawing including edges and vertices on the screen of the computer or the mobile terminal by using a drawing tool such as a computer pen, a mouse, or a finger (on a touch screen).
  • the sub-line drawing is a sub-picture after the two-dimensional line drawing is divided, and is also a plan view composed of a straight line segment, and a given two-dimensional line drawing can be divided into a plurality of sub-line drawing.
  • the three-dimensional reconstruction is performed on each sub-line drawing, specifically: the objective function can be minimized according to Calculate the depth values of all vertices in the sub-line drawing; where ⁇ i (z 1 , z 2 , ⁇ ,
  • the establishment uses MSDA (Minimum Standard Deviation of Angles), plane planarity, line parallelism, equidistance, and angular orthogonality.
  • MSDA Minimum Standard Deviation of Angles
  • plane planarity plane planarity
  • line parallelism equidistance
  • angular orthogonality angular orthogonality.
  • the five rules are as follows: MSDA rule requirements The standard deviation of the angle of the three-dimensional object on all vertices is the smallest; the planarity rule of the surface requires that all vertices on one surface are coplanar in three-dimensional space; the parallelism rule of the line requires a pair of parallel lines on the line drawing in three-dimensional space.
  • the equidistance rule requires that the ratio of the edges of objects in three-dimensional space is the same as the ratio of their on-line drawing; the orthogonality of the angle requires that if an angle formed by three lines meets the orthogonal angular condition on the line drawing It is orthogonal to the three corners in three dimensions.
  • N v is the number of vertices of the sub-line drawing
  • c is the number of regular functions. For a given set of weights, this set of weights is obtained experimentally. For example, 100, 1, 20, 15, and 20 can be used to correspond to the above five rules.
  • Performing the three-dimensional reconstruction result of the sub-line drawing to obtain a three-dimensional object specifically: obtaining the combined part information of each three-dimensional reconstruction result by using the original two-dimensional line drawing, and re-adjusting by comparing the size of the combined part Combine the size of the two parts, combine all the sub-objects into a whole, and then according to the minimum objective function Adjusting the size of the reorganized three-dimensional object; where N v is the number of vertices of the reorganized three-dimensional object.
  • the combination between the three-dimensional reconstruction results may be a combination of points or a point line, for example, two three-dimensional reconstruction results that need to be combined may be the three-dimensional weight of the first sub-line drawing.
  • This combination may be the actual splicing of two sub-objects together, or it may be to dig another sub-object from a sub-object or a part of the first sub-object, for example, after implementing the splicing rule, discovering that the second sub-object actually Within the first sub-object (ie, the inner space coincides), the second sub-object should be excavated from the first sub-object with its spatially overlapping portion.
  • the sub-line drawing should not contain concave surfaces and the sub-line drawing should contain fewer possible faces.
  • the surface is a region surrounded by the edge of the same plane in the two-dimensional line drawing, and is a convex surface or a concave surface.
  • the two-dimensional line drawing includes a concave surface, add an auxiliary vertex on the two-dimensional line drawing according to a preset strategy, and an auxiliary side to obtain an extended two-dimensional line drawing.
  • the auxiliary vertex is a new vertex other than the original vertex of the two-dimensional line drawing
  • the auxiliary side is a line connecting the auxiliary vertex and the original vertex of the two-dimensional line drawing, or is two-dimensional a line between two original vertices of a line drawing, or a line between two auxiliary vertices
  • the extended two-dimensional line drawing is a two-dimensional line drawing containing the auxiliary vertices and the auxiliary side
  • the surface of the extended two-dimensional line drawing is a convex surface
  • the convex surface is the surface of the two-dimensional line drawing, and all the vertices of the convex surface are on the convex convex surface of the convex surface.
  • an auxiliary side is added to the two-dimensional line drawing to obtain an extended two-dimensional line drawing, and the following two methods are respectively described below:
  • the concave surface of the two-dimensional line drawing 4a has: abcdefgha, ijklmnopi, in the concave surface abcdefgha, with the concave point c as an end point, a parallel line cG parallel to the side ha of the concave surface, wherein, cG and concave When the edges hg intersect, then cG belongs to the first type of auxiliary lines.
  • the convex surface hgjih of the dimension line drawing 4a has the endpoints C, D, G, and H of the first type of auxiliary edges, and the endpoints A and B of the first type of auxiliary edges exist on the convex surface gfkjg, and the first type of auxiliary edges exist on the convex surface hapih.
  • the endpoints E and F respectively connect the endpoints on the same convex surface to obtain the second type of auxiliary edges: GH, CD, AB, and EF, and obtain the extended line drawing 4b of the two-dimensional line drawing 4a.
  • the first concave surface comprises a collinear edge, and a third type of auxiliary edge is added to the first concave surface, wherein the third type of auxiliary edge is the first concave surface The line between the collinear sides.
  • collinear edges are different sides of the same straight line on the same plane.
  • the two-dimensional line drawing 5a includes concave surfaces acfgjknoa, bdehilmpb, wherein the concave surface acfgjknoa includes collinear edges on and gf, and the collinear edges on and gf are connected, and the connecting line ng is the third type of auxiliary.
  • the concave surface acfgjknoa includes collinear edges on and gf, and the collinear edges on and gf are connected
  • the connecting line ng is the third type of auxiliary.
  • a loop on the two-dimensional line drawing of the segmentation surface is formed by cutting a two-dimensional line drawing, the dividing surface includes both the original side of the two-dimensional line drawing and some added on the two-dimensional line drawing surface. Edge (ie auxiliary side).
  • the obtaining the segmentation surface of the two-dimensional line drawing according to the extended two-dimensional line drawing specifically includes:
  • the first edge is classified into the first edge set; wherein the degree of the vertex is the number of edges connected to the vertex ;
  • the loop is determined as a loop of the split plane of the two-dimensional line drawing, and the split plane is obtained; wherein the preset criterion includes the following five criteria:
  • the loop includes a 3D bump of degree 3, the loop is not a loop of the split plane; wherein the 3D bump is a vertex, and the faces on which the bumps are located are convex.
  • the loop contains a chord that is on the surface of the two-dimensional line drawing, then the loop is not a loop of the split plane; wherein the chord is the upper two non-adjacent points of the connected loop, And the line segment inside the loop.
  • the loop contains two non-collinear sides, and the two sides are the sides of the same face f, and there is an overlap between the area of the circuit siege and f, then the loop is not a split plane The loop.
  • the process of acquiring the segmentation plane of the two-dimensional line drawing 6a is as follows: the degree of the vertex d is equal to 4, the line segment of hd is selected, and the loop from the endpoint d to the endpoint h is searched, and the search is started from the d point.
  • step 303 further includes:
  • the segmentation surface having the smallest number of edges or the segmentation surface having the shortest circumference is selected as the pit point Split the face.
  • the split surface of the two-dimensional line drawing obtained by the above method is: EcnFE, cnHFC, dABmd, CdmDC, wherein two split surfaces appear on the pit c: EcnFE, cnHFC, and the split plane is deleted.
  • the plurality of sub-line drawing of the two-dimensional line drawing is obtained by the segmentation facing the two-dimensional line drawing segment according to the segmentation, and specifically includes:
  • a surface having a common side with the dividing surface is classified into a first surface set
  • the second surface In the first set of faces, if there is a second surface, there is a common edge with the surface in the second set of faces, and the common edge is connected to one vertex of the split face, and the common edge is not An edge belonging to the dividing surface, the second surface is deleted from the first surface set, and is classified into a second surface set;
  • the dividing plane of the two-dimensional line drawing 6a is: jmlkj, dhgcd; taking the dividing plane jmlkj as an example for description: firstly, the plane having the common side with jmlkj in FIG. 6a, that is, the adjacent side of jmlkj is found.
  • the 6 faces are classified into the first face set; the first face ajkdhea in the first face set is taken into the second face set, then only the first face set is Remaining ⁇ abcdklmja, sjkts, klotk, mlonm, sjmns ⁇ ; find that the face abcdklmja and the first face ajkdhea in the first set have common edges: aj and kd, and the two edges and the vertices j and k of the split face jmlkj Connected, but aj and kd are not the edges of the split surface jmlkj, then the face abcdklmja is deleted from the first face set and is classified into the second face set; and the faces in the first face set klotk, ml
  • the two methods of adding the auxiliary vertex and the auxiliary line in step 302 may be used alone or in combination, and the method of adding the auxiliary vertex and the auxiliary line may be mixed.
  • the two-dimensional line drawing can also be divided into multiple sub-line drawing by the following method:
  • the first concave surface comprises a collinear edge, then in the first concave Adding a third type of auxiliary edge to the surface to obtain a first extended two-dimensional line drawing, wherein the third type of auxiliary side is a line connecting the collinear sides of the first concave surface;
  • a first type of auxiliary side is added to each concave surface of the sub-line drawing, wherein the first type of the first type
  • the auxiliary edge is parallel to either side of the concave surface, one end point of the first type of auxiliary side is a concave point of the concave surface, and the other end point of the first type auxiliary side is an auxiliary vertex, and the auxiliary vertex is An intersection of the edge of the concave surface and the first type of auxiliary line;
  • the sub-line drawing is further divided into a plurality of sub-line drawing according to the dividing surface of the sub-line drawing.
  • the concave surface utspolkgfedau of the two-dimensional line drawing 7a includes the collinear sides ad and gk
  • the concave surface bvwrqnmjhiycbv includes the collinear sides bc and hj
  • the third type of auxiliary edges, dg and ch are added to obtain the first extension.
  • the segmentation surface cdghc is acquired according to the method described in step 303, and the two sub-line drawings 7b1 and 7b2 of the two-dimensional line drawing are acquired according to the method described in step 304;
  • the sub-line drawing 7b2 includes the concave surface Autspolka and bvwrqnmjb, add the first type of auxiliary edges sE, sF, pC, pA, rF, rH, qB, qD to the sub-line drawing 7b2, add the second type of auxiliary edges EF, GH, CD, AB, to obtain the extended sub- Line drawing 7d; according to 7d, using the acquisition method described in step 303, obtaining the segmentation faces FEsrF and qpABq of the extended sub-line drawing; using the segmentation faces FEsrF and qpABq, and the method described in step 304 to obtain the sub-line drawing Multiple sub-line drawing: 7
  • the method for constructing a three-dimensional object provided by the embodiment of the present invention can be divided into Cut out a simpler, visually perceptible sub-line drawing, so that the reconstruction of the three-dimensional object can be completed efficiently and accurately; avoiding the rapid expansion of the solution space due to the complexity or complexity of the line drawing processed in the prior art. And it falls into the local optimum, which leads to the inaccuracy of 3D reconstruction.
  • a computer-aided design system 80 according to an embodiment of the present invention is described below with reference to FIG. 8. As shown in FIG. 8, the computer-aided design system 80 includes:
  • the obtaining module 801 is configured to acquire a two-dimensional line drawing.
  • the two-dimensional line drawing is a plan composed of straight line segments, which can be represented by a two-dimensional line drawing containing vertices and edges; the sides and vertices form a two-dimensional line drawing face, and the two-dimensional line drawing face can be composed of a convex or concave surface; all the vertices on the convex finger surface are on the convex surface of the convex surface, all the vertices on the convex surface are convex, and at least one vertex on the concave surface is not on the convex convex surface of the surface a face having a concave surface that is not on the convex convex surface of the concave surface is a concave point, and a point on the convex convex surface of the concave surface is a convex point; for example, as shown in FIG.
  • the convex surface and the convex surface provided by the embodiment of the present invention
  • 2b is a concave surface
  • the vertex A is a convex point
  • the vertex B is a concave point.
  • the user or the designer can draw a two-dimensional line drawing including edges and vertices on the screen of the computer or the mobile terminal by using a drawing tool such as a computer pen, a mouse, or a finger (on a touch screen). .
  • the segmentation module 802 is configured to divide the two-dimensional line drawing into a plurality of sub-line drawings.
  • the sub-line drawing is a sub-picture after the two-dimensional line drawing is divided, and is also a plan view composed of a straight line segment, and a given two-dimensional line drawing can be divided into a plurality of sub-line drawing.
  • the three-dimensional reconstruction module 803 is configured to perform three-dimensional reconstruction on each sub-line drawing separately.
  • the recombination module 804 is configured to recombine the three-dimensional reconstruction result of the sub-line drawing to obtain a three-dimensional object.
  • the three-dimensional reconstruction module 803 is specifically configured to:
  • the minimum objective function Calculate the depth values of all vertices in the sub-line drawing; where ⁇ i (z 1 , z 2 , ⁇ ,
  • MSDA Minimum Standard Deviation of Angles
  • plane planarity plane planarity
  • line parallelism line parallelism
  • equidistance equidistance
  • angular orthogonality equidistance
  • the five rules are as follows: MSDA rule requirements The standard deviation of the angle of the three-dimensional object on all vertices is the smallest; the planarity rule of the surface requires that all vertices on one surface are coplanar in three-dimensional space; the parallelism rule of the line requires a pair of parallel lines on the line drawing in three-dimensional space.
  • the equidistance rule requires that the ratio of the edges of objects in three-dimensional space is the same as the ratio of their on-line drawing; the orthogonality of the angle requires that if an angle formed by three lines meets the orthogonal angular condition on the line drawing It is orthogonal to the three corners in three dimensions.
  • N v is the number of vertices of the sub-line drawing
  • c is the number of regular functions. For a given set of weights, this set of weights is obtained experimentally. For example, 100, 1, 20, 15, and 20 can be used to correspond to the above five rules.
  • N v is the number of vertices of the sub-line drawing
  • z i ( i 1, 2, N v) i-th coordinate values of the vertices of the depth
  • the reorganization module 804 is specifically configured to:
  • the original two-dimensional line drawing is used to obtain the combined part information of each three-dimensional reconstruction result, and at the same time, the size of the two parts to be combined is re-adjusted by comparing the size of the combined part, and all the sub-objects are combined into a whole, and then the minimum target is function Adjusting the size of the reorganized three-dimensional object; where N v is the number of vertices of the reorganized three-dimensional object.
  • the combination between the three-dimensional reconstruction results may be point-and-point combination, or may be combined at the point line.
  • two three-dimensional reconstruction results that need to be combined may be the three-dimensional weight of the first sub-line drawing.
  • This combination may be the actual splicing of two sub-objects together, or it may be to dig another sub-object from a sub-object or a part of the first sub-object, for example, after implementing the splicing rule, discovering that the second sub-object actually Within the first sub-object (ie, the inner space coincides), the second sub-object should be excavated from the first sub-object with its spatially overlapping portion.
  • the sub-line drawing should not contain concave surfaces and the sub-line drawing should contain fewer possible faces.
  • a description of a segmentation module 802 provided by the present invention in real time, as shown in FIG. 9, includes:
  • the extracting unit 8021 is configured to extract a surface of the two-dimensional line drawing.
  • the surface is an area surrounded by sides of the same plane in the two-dimensional line drawing, and is a convex surface or a concave surface.
  • the adding unit 8022 is configured to: if the two-dimensional line drawing includes a concave surface, add an auxiliary vertex on the two-dimensional line drawing according to a preset strategy, and an auxiliary side to obtain an extended two-dimensional line drawing.
  • the auxiliary vertex is a new vertex other than the original vertex of the two-dimensional line drawing
  • the auxiliary side is a line connecting the auxiliary vertex and the original vertex of the two-dimensional line drawing, or is two-dimensional a line between two original vertices of a line drawing, or a line between two auxiliary vertices
  • the extended two-dimensional line drawing is a two-dimensional line drawing containing the auxiliary vertices and the auxiliary side
  • the surface of the extended two-dimensional line drawing is convex.
  • the convex surface is a surface of the two-dimensional line drawing, and all vertices of the convex surface are on the convex convex surface of the convex surface.
  • the segmentation plane obtaining unit 8023 is configured to acquire a segmentation plane of the two-dimensional line drawing according to the extended two-dimensional line drawing.
  • a loop on the two-dimensional line drawing of the segmentation surface is formed by cutting a two-dimensional line drawing, the dividing surface includes both the original side of the two-dimensional line drawing and some added on the two-dimensional line drawing surface. Edge (ie auxiliary side).
  • the sub-picture obtaining unit 8024 is configured to perform segmentation according to the segmentation facing the two-dimensional line drawing, and acquire a plurality of sub-line drawing of the two-dimensional line drawing.
  • the adding unit 8022 is specifically used in the following two cases:
  • the concave surface of the two-dimensional line drawing 4a has: abcdefgha, ijklmnopi, in the concave surface abcdefgha, with the concave point c as an end point, a parallel line cG parallel to the side ha of the concave surface, wherein, cG and concave When the edges hg intersect, then cG belongs to the first type of auxiliary lines.
  • the convex surface hgjih of the dimension line drawing 4a has the endpoints C, D, G, and H of the first type of auxiliary edges, and the endpoints A and B of the first type of auxiliary edges exist on the convex surface gfkjg, and the first type of auxiliary edges exist on the convex surface hapih.
  • the endpoints E and F respectively connect the endpoints on the same convex surface to obtain the second type of auxiliary edges: GH, CD, AB, and EF, and obtain the extended line drawing 4b of the two-dimensional line drawing 4a.
  • the first concave surface includes a collinear side
  • the A third type of auxiliary edge is added to a concave surface, wherein the third type of auxiliary side is a line connecting the collinear sides of the first concave surface.
  • collinear edges are different sides of the same straight line on the same plane.
  • the two-dimensional line drawing 5a includes concave surfaces acfgjknoa, bdehilmpb, wherein the concave surface acfgjknoa includes collinear edges on and gf, and the collinear edges on and gf are connected, and the connecting line ng is the third type of auxiliary.
  • the concave surface acfgjknoa includes collinear edges on and gf, and the collinear edges on and gf are connected
  • the connecting line ng is the third type of auxiliary.
  • split plane acquiring unit 8023 is specifically configured to:
  • the first edge is classified into the first edge set; wherein the degree of the vertex is the number of edges connected to the vertex ;
  • the loop is determined as a loop of the split plane of the two-dimensional line drawing, and the split plane is obtained; wherein the preset criterion includes the following five criteria:
  • the loop includes a 3D bump of degree 3, the loop is not a loop of the split plane; wherein the 3D bump is a vertex, and the faces on which the bumps are located are convex.
  • the loop contains a chord that is on the surface of the two-dimensional line drawing, then the loop is not a loop of the split plane; wherein the chord is the upper two non-adjacent points of the connected loop, And the line segment inside the loop.
  • the loop contains two non-collinear sides, and the two sides are the sides of the same face f, and there is an overlap between the area of the circuit siege and f, then the loop is not a split plane The loop.
  • the process of acquiring the segmentation plane of the two-dimensional line drawing 6a is as follows: the degree of the vertex d is equal to 4, the line segment of hd is selected, and the loop from the endpoint d to the endpoint h is searched, and the search is started from the d point.
  • the divided surface acquiring unit 8023 is further configured to:
  • the segmentation surface having the smallest number of edges or the segmentation surface having the shortest circumference is selected as the pit point Split the face.
  • the split surface of the two-dimensional line drawing obtained by the above method is: EcnFE, cnHFC, dABmd, CdmDC, wherein two split surfaces appear on the pit c: EcnFE, cnHFC, and the split plane is deleted.
  • sub-graph obtaining unit 8024 is specifically configured to:
  • a surface having a common side with the dividing surface is classified into a first surface set
  • the second surface In the first set of faces, if there is a second surface, there is a common edge with the surface in the second set of faces, and the common edge is connected to one vertex of the split face, and the common edge is not An edge belonging to the dividing surface, the second surface is deleted from the first surface set, and is classified into a second surface set;
  • the dividing plane of the two-dimensional line drawing 6a is: jmlkj, dhgcd; taking the dividing plane jmlkj as an example for description: firstly, the plane having the common side with jmlkj in FIG. 6a, that is, the adjacent side of jmlkj is found.
  • the 6 faces are classified into the first face set; the first face ajkdhea in the first face set is taken into the second face set, then only the first face set is Remaining ⁇ abcdklmja, sjkts, klotk, mlonm, sjmns ⁇ ; find that the face abcdklmja and the first face ajkdhea in the first set have common edges: aj and kd, and the two edges and the vertices j and k of the split face jmlkj Connected, but aj and kd are not the edges of the split surface jmlkj, then the face abcdklmja is deleted from the first face set and is classified into the second face set; and the faces in the first face set klotk, ml
  • the obtaining module in this embodiment may be a separately set processor in the computer aided design system, or may be a processor integrated in the computer aided design system, or may be in the form of program code.
  • the processor of the base station invokes and executes the functions of the above acquiring module;
  • the splitting module is similar to the acquiring module, and can be integrated with the acquiring module, or can be independently implemented;
  • the 3D reconstruction module and the acquiring The module and the segmentation module are similar, and can be integrated with the acquisition module and the segmentation module, or can be implemented independently;
  • the reorganization module is similar to the acquisition module, the segmentation module, and the three-dimensional reconstruction module, and can be combined with the acquisition module, the segmentation module, and the three-dimensional reconstruction. Modules are integrated and can be implemented independently.
  • the processor described herein may be a Central Processing Unit (CPU), or an Application Specific Integrated Circuit (ASIC), or one or more integrated systems configured to implement embodiments of the present invention. Circuit.
  • an embodiment of the present invention provides a method for constructing a three-dimensional object and a computer-aided design system, which acquires a two-dimensional line drawing; divides the two-dimensional line drawing into a plurality of sub-line drawing; and respectively performs three-dimensional drawing on each sub-line And reconstructing the three-dimensional reconstruction result of the sub-line drawing to obtain a three-dimensional object, wherein the dividing the two-dimensional line drawing into a plurality of sub-line drawing comprises: extracting a surface of the two-dimensional line drawing Wherein the surface is an area surrounded by edges of the same plane in the two-dimensional line drawing, which is a convex surface or a concave surface; if the two-dimensional line drawing includes a concave surface, the two-dimensional line drawing is performed according to a preset strategy.
  • Adding an auxiliary vertex, and an auxiliary side acquiring an extended two-dimensional line drawing; acquiring a dividing surface of the two-dimensional line drawing according to the extended two-dimensional line drawing; facing the two-dimensional line according to the dividing The drawing is performed to obtain a plurality of sub-line drawing of the two-dimensional line drawing.
  • the two-dimensional line drawing is processed, and the simpler and visually-aware sub-line drawing is segmented, and the reconstruction of the three-dimensional object is completed efficiently and accurately; and the line drawing processed in the prior art is avoided when the drawing is complicated or complicated.
  • the system falls into local optimum due to the rapid expansion of the solution space, resulting in inaccurate defects in 3D reconstruction.
  • the disclosed system, apparatus, and method may be implemented in other manners.
  • the device described above The embodiment is merely illustrative.
  • the division of the unit is only a logical function division, and the actual implementation may have another division manner, for example, multiple units or components may be combined or may be integrated into another system. Or some features can be ignored or not executed.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may be physically included separately, or two or more units may be integrated into one unit.
  • the above integrated unit can be implemented in the form of hardware or in the form of hardware plus software functional units.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Hardware Design (AREA)
  • Geometry (AREA)
  • General Engineering & Computer Science (AREA)
  • Computer Graphics (AREA)
  • Software Systems (AREA)
  • Evolutionary Computation (AREA)
  • Processing Or Creating Images (AREA)
  • Image Generation (AREA)

Abstract

一种构建三维物体的方法及计算机辅助设计系统,涉及计算机视觉领域,能够分割出更加简单的、符合视觉感知规律的子线画图,高效准确地完成三维物体的重建。该方法包括:获取二维线画图;将二维线画图分割为多个子线画图;分别对每个子线画图进行三维重构;将子线画图的三维重构结果进行重组得到三维物体,其中,将二维线画图分割为多个子线画图,具体包括:提取二维线画图的表面;若二维线画图包含凹面,则根据预设策略,在二维线画图上添加辅助顶点,以及,辅助边,获取扩展的二维线画图;根据扩展的二维线画图,获取二维线画图的分割面;根据分割面对二维线画图进行分割,获取二维线画图的多个子线画图。

Description

一种构建三维物体的方法及计算机辅助设计系统
本申请要求于2014年05月16日提交中国专利局、申请号为201410209597.0、发明名称为“一种构建三维物体的方法及计算机辅助设计系统”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
技术领域
本发明涉及计算机视觉领域,尤其涉及一种构建三维物体的方法及计算机辅助设计系统。
背景技术
通常情况下,人们借助传统的计算机辅助设计系统(Computer Aided Design,CAD系统),按照WIMP模式(视窗Windows,图标Icon,菜单Menu,鼠标Pointer)生成物体三维几何模型,首先用户需要从CAD系统的用户交互界面的菜单、面板或者图标中选取不同的绘制工具,并在交互窗口中选择要绘制物体的各种参数,最后通过点击、拖拽等一系列操作选择物体的绘制位置、大小比例等,完成三维物体的绘制。虽然CAD系统能够提供强大的三维物体设计和编辑能力,但是操作比较复杂,只有对于设计系统中繁杂菜单非常熟悉的专业人员才能很好地完成设计,同时,使用这种操作系统完成三维物体繁琐费时,即使设计一个比较简单的物体也需要比较长时间。
随着电脑技术、三维采集设备和网络的发展,近几年来,人们提出一种新的构思:利用简单易画的二维线画图(即草图)重构三维物体,并开发出新的CAD系统采用SBIM模式(sketch-based interfaces for modeling,基于用户草图绘制的三维建模)重构三维物体,其中,所述草图就是二维线画图,这类操作界面简单易懂,用户只需绘制出与三维物体对应的二维线画图(草图),系统就可以自动生成用户所需的三维模型,由于同一个二维线画图对应着无穷多 个三维物体,所以这类系统的设计难点在于如何从用户给出的二维线画图中重构出用户所需要的三维物体。传统的做法是将线画图重构三维物体建立为一个优化问题,利用最优化方法找到符合人类视觉感知的三维物体,但传统的该类系统只能处理简单的或部分复杂的线画图,当处理的线画图较为复杂或非常复杂时,系统会因解空间的急剧膨胀而陷入局部最优,导致三维重构不准确。
发明内容
本发明实施例提供一种构建三维物体的方法及计算机辅助设计系统,对二维线画图进行处理,分割出更加简单的、符合视觉感知规律的子线画图,高效准确地完成三维物体的重建。
为达到上述目的,本发明采用的技术方案是,
第一方面,本发明实施例提供一种构建三维物体的方法,包括:获取二维线画图;将所述二维线画图分割为多个子线画图;分别对每个子线画图进行三维重构;将所述子线画图的三维重构结果进行重组得到三维物体,其特征在于,所述将所述二维线画图分割为多个子线画图,具体包括:
提取所述二维线画图的表面,其中,所述表面为所述二维线画图中处于同一平面的边围成的区域,为凸面或凹面;
若所述二维线画图包含凹面,则根据预设策略,在二维线画图上添加辅助顶点,以及,辅助边,获取扩展的二维线画图;
根据所述扩展的二维线画图,获取所述二维线画图的分割面;
根据所述分割面对所述二维线画图进行分割,获取所述二维线画图的多个子线画图。
在第一方面的第一种可能的实现方式中,结合第一方面,所述若所述二维线画图包含凹面,则根据预设策略,在二维线画图上添加辅助顶点,以及,辅助边,获取扩展的二维线画图,包括:
分别在所述二维线画图的每个凹面上,增加第一类辅助边,其中,所述第一类辅助边平行于所述凹面的任一边,所述第一类辅助边的一个端点为所述凹面的凹点,所述第一类辅助边的另一端点为 辅助顶点,所述辅助顶点为与所述凹面的边与所述第一类辅助线的交点;
遍历所述二维线画图的凸面;
若存在第一凸面的不相邻的两个边上,分别包含所述辅助顶点,则在所述第一凸面上增加第二类辅助边,其中,所述第二类辅助边的两个端点分别为所述第一凸面的不相邻的两个边上的辅助顶点。
在第一方面的第二种可能的实现方式中,结合第一方面,所述若所述二维线画图包含凹面,则根据预设策略,在二维线画图上添加辅助顶点,以及,辅助边,获取扩展的二维线画图,包括:
若存在第一凹面,所述第一凹面包含共线边,则在所述第一凹面上增加第三类辅助边,其中,所述第三类辅助边为所述第一凹面的共线边之间的连线。
在第一方面的第三种可能的实现方式中,结合第一方面的第一种可能的实现方式或第一方面的第二种可能的实现方式,所述根据所述扩展的二维线画图,获取所述二维线画图的分割面,包括:
遍历所述扩展的二维线画图中的边;
若存在第一边的两顶点中至少一个顶点的度不小于4,则将所述第一边归入第一边集合;
对所述第一边集合中的每一个边,分别采用深度优先搜索算法获取从所述边的一端点到所述边的另一端点的回路;
若搜索到的回路与预设准则不矛盾,则将所述回路确定为所述二维线画图的分割面的回路,获取分割面;其中,所述预设准则包括:若回路包含一个度为3的3D凸点,则所述回路不是分割面的回路;若回路经过两个相邻面f1、f2,且f1、f2共同拥有两个或两个以上不在同一直线上的边,则所述回路不是分割面的回路;若回路的边自相交,则所述回路不是分割面的回路;若回路包含有一个弦,且该弦在二维线画图的表面上,则所述回路不是分割面的回路;若回路包含有两个不共线的边,且两个边是同一个面f的边、以及,所述回路围城的区域和f之间有重叠区域,则所述回路不是分割面 的回路。
在第一方面的第四种可能的实现方式中,结合第第一方面的第三种可能的实现方式,所述方法还包括:
在获取的所述二维线画图的分割面中,若同一个凹点上出现至少两个分割面,则选择包含边的数量最小的分割面或者周长最短的分割面为所述凹点所在的分割面。
在第一方面的第五种可能的实现方式中,结合第一方面至第一方面的第四种可能的实现方式中的任一种实现方式,所述根据所述分割面对所述二维线画图进行分割,获取所述二维线画图的多个子线画图,包括:
在所述扩展的二维线画图中,将与所述分割面有公共边的表面归入第一面集合;
选取所述第一面集合中的第一表面归入第二面集合,并将所述第一面集合中的所述第一表面删除,其中,所述第一表面为所述第一面集合中的任意一个表面;
在所述第一面集合中,若存在第二表面,与所述第二面集合中的表面有公共边,且所述公共边与所述分割面的一个顶点相连,且所述公共边不属于所述分割面的边,则将所述第二面从所述第一面集合中删除,归入第二面集合;
获取与所述分割面的顶点相连但不属于所述分割面的边的边的集合;
在所述边的集合中,若存在第一边,与所述第一边相连的所述扩展的二维线画图的两个表面与所述第二面集合中的第一表面属于同一面集合,则将所述第一边归入第一边集合,否则,归入第二边集合;
根据所述第一面集合,第二面集合,第一边集合,第二边集合对所述二维线画图进行分割,获取所述二维线画图的多个子线画图。
第二方面,本发明实施例提供一种计算机辅助设计系统,包括:获取模块,用于获取二维线画图;分割模块,用于将所述二维线画 图分割为多个子线画图;三维重构模块,用于分别对每个子线画图进行三维重构;重组模块,用于将所述子线画图的三维重构结果进行重组得到三维物体,其特征在于,所述分割模块包括:
凹面获取单元,用于提取所述二维线画图的表面,其中,所述表面为所述二维线画图中处于同一平面的边围成的区域,为凸面或凹面;
添加单元,用于若所述二维线画图包含凹面,则根据预设策略,在二维线画图上添加辅助顶点,以及,辅助边,获取扩展的二维线画图;
分割面获取单元,用于根据所述扩展的二维线画图,获取所述二维线画图的分割面;
子图获取单元,根据所述分割面对所述二维线画图进行分割,获取所述二维线画图的多个子线画图。
在第二方面的第一种可能的实现方式中,结合第二方面,所述添加单元,具体用于:
分别在所述二维线画图的每个凹面上,增加第一类辅助边,其中,所述第一类辅助边平行于所述凹面的任一边,所述第一类辅助边的一个端点为所述凹面的凹点,所述第一类辅助边的另一端点为辅助顶点,所述辅助顶点为与所述凹面的边与所述第一类辅助线的交点;
遍历所述二维线画图的凸面;
若存在第一凸面的不相邻的两个边上,分别包含所述辅助顶点,则在所述第一凸面上增加第二类辅助边,其中,所述第二类辅助边的两个端点分别为所述第一凸面的不相邻的两个边上的辅助顶点。
在第二方面的第二种可能的实现方式中,结合第二方面,所述添加单元,具体用于:
若存在第一凹面,所述第一凹面包含共线边,则在所述第一凹面上增加第三类辅助边,其中,所述第三类辅助边为所述第一凹面的共线边之间的连线。
在第二方面的第三种可能的实现方式中,结合第二方面的第一种可能的实现方式或第二方面的第二种可能的实现方式,所述分割面获取单元,具体用于:
遍历所述扩展的二维线画图中的边;
若存在第一边的两顶点中至少一个顶点的度不小于4,则将所述第一边归入第一边集合;
对所述第一边集合中的每一个边,分别采用深度优先搜索算法获取从所述边的一端点到所述边的另一端点的回路;
若搜索到的回路与预设准则不矛盾,则将所述回路确定为所述二维线画图的分割面的回路,获取分割面;其中,所述预设准则包括:若回路包含一个度为3的3D凸点,则所述回路不是分割面的回路;若回路经过两个相邻面f1、f2,且f1、f2共同拥有两个或两个以上不在同一直线上的边,则所述回路不是分割面的回路;若回路的边自相交,则所述回路不是分割面的回路;若回路包含有一个弦,且该弦在二维线画图的表面上,则所述回路不是分割面的回路;若回路包含有两个不共线的边,且两个边是同一个面f的边、以及,所述回路围城的区域和f之间有重叠区域,则所述回路不是分割面的回路。
在第二方面的第四种可能的实现方式中,结合第第二方面的第三种可能的实现方式,所述分割面获取单元,还用于:
在获取的所述二维线画图的分割面中,若同一个凹点上出现至少两个分割面,则选择包含边的数量最小的分割面或者周长最短的分割面为所述凹点所在的分割面。
在第二方面的第五种可能的实现方式中,结合第二方面至第二方面的第四种可能的实现方式中的任一种实现方式,所述子图获取单元,具体用于:
在所述扩展的二维线画图中,将与所述分割面有公共边的表面归入第一面集合;
选取所述第一面集合中的第一表面归入第二面集合,并将所述 第一面集合中的所述第一表面删除,其中,所述第一表面为所述第一面集合中的任意一个表面;
在所述第一面集合中,若存在第二表面,与所述第二面集合中的表面有公共边,且所述公共边与所述分割面的一个顶点相连,且所述公共边不属于所述分割面的边,则将所述第二面从所述第一面集合中删除,归入第二面集合;
获取与所述分割面的顶点相连但不属于所述分割面的边的边的集合;
在所述边的集合中,若存在第一边,与所述第一边相连的所述扩展的二维线画图的两个表面与所述第二面集合中的第一表面属于同一面集合,则将所述第一边归入第一边集合,否则,归入第二边集合;
根据所述第一面集合,第二面集合,第一边集合,第二边集合对所述二维线画图进行分割,获取所述二维线画图的多个子线画图。
由上可知,本发明实施例提供一种构建三维物体的方法及计算机辅助设计系统,获取二维线画图;将所述二维线画图分割为多个子线画图;分别对每个子线画图进行三维重构;将所述子线画图的三维重构结果进行重组得到三维物体,其中,所述将所述二维线画图分割为多个子线画图,具体包括:提取所述二维线画图的表面,其中,所述表面为所述二维线画图中处于同一平面的边围成的区域,为凸面或凹面;若所述二维线画图包含凹面,则根据预设策略,在二维线画图上添加辅助顶点,以及,辅助边,获取扩展的二维线画图;根据所述扩展的二维线画图,获取所述二维线画图的分割面;根据所述分割面对所述二维线画图进行分割,获取所述二维线画图的多个子线画图。如此,对二维线画图进行处理,分割出更加简单的、符合视觉感知规律的子线画图,高效准确地完成三维物体的重建;避免了现有技术中处理的线画图较为复杂或非常复杂时,系统因解空间的急剧膨胀而陷入局部最优,导致三维重构不准确的缺陷。
附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案,下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本发明的一些实施例,对于本领域普通技术人员来讲,在不付出创造性劳动的前提下,还可以根据这些附图获得其他的附图。
图1为一种基于二维线画图构建三维物体的基本流程示意图;
图2为本发明实施例提供的凸面和凹面的结构示意图;
图3为本发明实时提供的一种将二维线画图分割为多个子线画图的具体流程图;
图4为本发明实施例提供的一种将二维线画图4a分割为多个子线画图的流程示意图;
图5为本发明实施例提供的一种将二维线画图5a分割为多个子线画图的流程示意图;
图6为本发明实施例提供的获取二维线画图6a的分割面及根据分割面获取多个子线画图的流程示意图;
图7为本发明实施例提供的一种将二维线画图7a分割为多个子线画图的流程示意图;
图8为本发明实施例提供的一种计算机辅助设计系统80的结构图;
图9为本发明实施例提供的一种分割模块802的结构图。
具体实施方式
下面将结合本发明实施例中的附图,对本发明实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅仅是本发明一部分实施例,而不是全部的实施例。基于本发明中的实施例,本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例,都属于本发明保护的范围。
采用SBIM模式(sketch-based interfaces for modeling,基于用户草图绘制的三维建模),可以实现三维物体的重建,且操作简单易懂,参见图1,为一种基于二维线画图构建三维物体的基本流程示 意图,如图1所示,可以包括下述步骤:获取二维线画图;将所述二维线画图分割为多个子线画图;分别对每个子线画图进行三维重构;将所述子线画图的三维重构结果进行重组得到三维物体。
其中,所述二维线画图为由直线线段组合而成的平面图,可以用一个包含顶点和边的二维线画图表示;边和顶点组成二维线画图的面,组成的二维线画图的面可以为凸面或凹面;所述凸面指面上的所有顶点均在该面的凸壳上的面,凸面上的所有顶点均为凸点,凹面指面上的至少一个顶点不在该面的凸壳上的面,凹面上不在所述凹面的凸壳上的点为凹点,在所述凹面的凸壳上的点为凸点;例如,如图2所示,为本发明实施例提供的凸面和凹面的结构示意图,如图2所示,2a为凸面,2b为凹面,顶点A为凸点,顶点B为凹点。优选的,本发明实施例中,用户或者设计者可以使用电脑笔、鼠标、手指(在一个触摸屏上)等绘图工具在计算机或者移动终端的屏幕上绘出包含边和顶点的二维线画图。
所述子线画图为所述二维线画图分割后的子图,也为由直线线段组合而成的平面图,一个给定的二维线画图可以分割为若干个子线画图。
所述对每个子线画图进行三维重构,具体为:可以根据最小化目标函数
Figure PCTCN2014088311-appb-000001
计算出子线画图中所有顶点的深度值;其中φi(z1,z2,Λ,
Figure PCTCN2014088311-appb-000002
)的建立使用了MSDA(Minimum Standard Deviation of Angles,角的最小标准偏差)、面的平面性、线的平行性、等距性、角的正交性五条规则,它们的解释如下:MSDA规则要求三维物体在所有顶点上的角的标准偏差最小;面的平面性规则要求在一个面上的所有顶点在三维空间上共平面;线的平行性规则要求在线画图上平行的一对直线在三维空间中也平行;等距性规则要求三维空间中物体的边的比率与它们在线画图上的比率相同;角的正交性要求,如果一个由三条线形成的角在线画图上满足正交的角条件,它在三维空间中的三个角正交。Nv为子线画图的顶点的个数,zi(i=1,2,ΛNv)为第i个顶点的深度座标值,c为规则函数的个数,
Figure PCTCN2014088311-appb-000003
为一组 给定的权值,这组权值是通过实验得到,例如,可以使用100,1,20,15,20分别对应以上五个规则。
可以理解的是,上述示例只是为了帮助理解本发明的技术方案,而不应该被视为对本发明技术方案的具体限制。φi(z1,z2,Λ,
Figure PCTCN2014088311-appb-000004
)的定义可以使用文献[K.Shoji,K.Kato.And F.Toyama.3-D interpretation of single line drawings based on entropy minimization principle.IEEE Conference on Computer Vision and Pattern Recognition(CVPR),2001]、[y.Leclerc and M.Fischler.An Optimization-Based Approach to the Interpretation of Single Line.Drawings as 3D Wire Frames.International Journal of Computer Vision(IJCV),9(2):113-136,1992]、[H.Lipson and M.Shpitaln.optimization based reconstruction of a 3d object from a single freehand line drawing.Computer-Aided Design,28(7):651-663,1996]、[A.Turner,D.Chapman,and A.Penn.Sketching space.Computer and Graphics,24:869-879,2000]、[J.Liu,L.Cao,Z.Li and X.Tang.Plane-based optimization for 3D object reconstruction from single line drawings.IEEE Transactions on Pattern Analysis and Machine Intelligence(TPAMI),30(2):315-327,2008]中提到的规则函数,比如夹角的标准方差等函数,Nv为子线画图的顶点的个数,zi(i=1,2,ΛNv)为第i个顶点的深度座标值,
Figure PCTCN2014088311-appb-000005
可以是文献[K.Shoji,K.Kato.And F.Toyama.3-D interpretation of single line drawings based on entropy minimization principle.IEEE Conference on Computer Vision and Pattern Recognition(CVPR),2001]、[y.Leclerc and M.Fischler.An Optimization-Based Approach to the Interpretation of Single Line.Drawings as 3D Wire Frames.International Journal of Computer Vision(IJCV),9(2):113-136,1992]、[H.Lip son and M.Shpitaln.optimization based reconstruction of a 3d object from a single freehand line drawing.Computer-Aided Design,28(7):651-663, 1996]、[A.Turner,D.Chapman,and A.Penn.Sketching space.Computer and Graphics,24:869-879,2000]、[J.Liu,L.Cao,Z.Li and X.Tang.Plane-based optimization for 3D object reconstruction from single line drawings.IEEE Transactions on Pattern Analysis and Machine Intelligence(TPAMI),30(2):315-327,2008]等中提到的权值,c为规则函数的个数。
所述将所述子线画图的三维重构结果进行重组得到三维物体,具体为:通过原始二维线画图的得到各个三维重构结果的结合部位信息,同时通过比较结合部位的尺寸重新调整需要结合的两个部分的大小,把所有子物体结合为整体之后,再根据最小化目标函数
Figure PCTCN2014088311-appb-000006
调整重组的三维物体的大小;其中Nv为重组的三维物体的顶点的个数。
可以理解的是,三维重构结果之间的结合可以是点点处结合,也可以是点线处结合,例如需要结合的两个三维重构结果可能结合部位是第一个子线画图的三维重构结果的一个点和第二个子线画图的三维重构结果的一个点;也可能是第一个子线画图的三维重构结果的一个点和第二个子线画图的三维重构结果的一条线的某个位置;也可能是第一个子线画图的三维重构结果的一个点和第二个子线画图的三维重构结果的一个面上的某个位置;也可能是第一个子线画图的三维重构结果的一条线和第二个子线画图的三维重构结果的一条线的某个位置;也可能是第一个子线画图的三维重构结果的一条线和第二个子线画图的三维重构结果的一个面的某个位置;也可以第一个子线画图的三维重构结果的一个面和第二个子线画图的三维重构结果的一个面的结合。这种结合可以是两个子物体的真正拼接在一起,也可以是从一个子物体中挖除另外一个子物体或者第一子物体的一部分,例如当按照拼接规则实施之后,发现第二子物体实际在第一子物体内部(即内部空间重合),则应该从第一子物体中挖除掉第二子物体与其在空间上重叠的部分。
不难发现,在利用二维线画图构建三维物体的基本过程中,若 分割后的子线画图比较复杂,则会造成子线画图的三维重建计算复杂且不够准确,从而使得重组后的三维物体不够准确。
为此,本发明实施例中,需要找到更符合人们思维方式的,简单的,更易于进行三维重构的子线画图,以使得三维物体的重构高效准确;优选的,本发明实施例中,子线画图应该不包含凹面且子线画图应该包含可能少的面。
下面结合图3,对本发明实时提供的一种将二维线画图分割为多个子线画图的具体过程进行说明,如图3所示,可以包括以下步骤:
301:提取二维线画图的表面。
其中,所述表面为二维线画图中处于同一平面的边围成的区域,为凸面或凹面。
302:若所述二维线画图包含凹面,则根据预设策略,在二维线画图上添加辅助顶点,以及,辅助边,获取扩展的二维线画图。
其中,所述辅助顶点为除所述二维线画图的原始顶点之外的新的顶点,所述辅助边为辅助顶点与二维线画图的原始顶点之间的连线,或者,为二维线画图的两个原始顶点之间的连线,或者,为两个辅助顶点之间的连线;所述扩展的二维线画图为包含所述辅助顶点和辅助边的二维线画图,所述扩展的二维线画图的表面均为凸面,所述凸面为所述二维线画图的表面,所述凸面的所有顶点均在所述凸面的凸壳上。
优选的,本发明实施例中可以通过下述两种方式对所述二维线画图添加辅助边,获取扩展的二维线画图,下面分别对这两种方式进行说明:
(1)分别在所述二维线画图的每个凹面上,增加第一类辅助边,其中,所述第一类辅助边平行于所述凹面的任一边,所述第一类辅助边的一个端点为所述凹面的凹点,所述第一类辅助边的另一端点为辅助顶点,所述辅助顶点为与所述凹面的边与所述第一类辅助线的交点;
遍历所述二维线画图的凸面;
若存在第一凸面的不相邻的两个边上,分别包含所述辅助顶点,则在所述第一凸面上增加第二类辅助边,其中,所述第二类辅助边的两个端点分别为所述第一凸面的不相邻的两个边上的辅助顶点。
例如,如图4所示,二维线画图4a的凹面有:abcdefgha,ijklmnopi,在凹面abcdefgha,以凹点c为端点作一条平行于凹面的边ha的平行线cG,其中,cG与凹面的边hg相交,则cG属于第一类辅助线,同理,遍历所有凹面的凹点增加第一类辅助边:cG、cE、dC、dA、nF、nH、mB、mJ;此时,在二维线画图4a的凸面hgjih存在第一类辅助边的端点C、D、G、H,在凸面gfkjg上存在第一类辅助边的端点A、B,在凸面hapih上存在第一类辅助边的端点E、F,分别将同一凸面上的所述端点进行连接,获取第二类辅助边:GH、CD、AB、EF,得到二维线画图4a的扩展线画图4b。
(2)若存在第一凹面,所述第一凹面包含共线边,则在所述第一凹面上增加第三类辅助边,其中,所述第三类辅助边为所述第一凹面的共线边之间的连线。
其中,所述共线边为同一平面上在同一直线的不同边。
例如,如图5所示,二维线画图5a中包含凹面acfgjknoa,bdehilmpb,其中,凹面acfgjknoa包含共线边on和gf,将共线边on和gf连接,则连接线ng为第三类辅助边,同理,在凹面bdehilmpb上添加第三类辅助边mh。
303:根据所述扩展的二维线画图,获取所述二维线画图的分割面。
其中,所述分割面二维线画图上的一条回路,通过对二维线画图的切割而形成,分割面既包括二维线画图的原始边,也包括一些在二维线画图表面上添加的边(即辅助边)。
示例性,所述根据所述扩展的二维线画图,获取所述二维线画图的分割面,具体包括:
遍历所述扩展的二维线画图中的边;
若存在第一边的两顶点中至少一个顶点的度不小于4,则将所述第一边归入第一边集合;其中,所述顶点的度为与所述顶点连接的边的个数;
对所述第一边集合中的每一个边,分别采用深度优先搜索算法获取从所述边的一端点到所述边的另一端点的回路;
若搜索到的回路与预设准则不矛盾,则将所述回路确定为所述二维线画图的分割面的回路,获取分割面;其中,所述预设准则包含以下5个准则:
(1)若回路包含一个度为3的3D凸点,则所述回路不是分割面的回路;其中,所述3D凸点为一个顶点,且所述凸点所在的面均为凸面。
(2)若回路经过两个相邻面f1、f2,且f1、f2共同拥有两个或两个以上不在同一直线上的边,则所述回路不是分割面的回路;
(3)若回路的边自相交,则所述回路不是分割面的回路;
(4)若回路包含有一个弦,且该弦在二维线画图的表面上,则所述回路不是分割面的回路;其中,所述弦为连接回路的上两个不相邻的点,且在回路内部的线段。
(5)若回路包含有两个不共线的边,且两个边是同一个面f的边、以及,所述回路围城的区域和f之间有重叠区域,则所述回路不是分割面的回路。
例如,如图6所示,获取二维线画图6a的分割面的过程为:顶点d的度等于4,选择hd这条线段,搜索从端点d到端点h的回路,首先从d点出发搜索:dc,dk,di;发现dc与准则1至5都不矛盾,所以从dc继续往下做深度优先搜索,得到:dcb,dci,dcg;发现dcb中的b是一个3d凸顶点,与准则1矛盾,则从dcb不再往下做深度优先搜索;dcg和dci与准则1至5不矛盾,则继续做深度优选搜索,从dci往下搜索到dcih和dcig:发现dcih中出现了弦di在回路hdcih之内,与准则4矛盾,故回路hdcih不是分割面;发现dcig出现了弦gc在回路hdcig之内,与准则4矛盾,故回路hdcih 不是分割面;从dcg往下搜索到dcgh和dcgf,发现dcgf中的f点为3D凸点,与准则1矛盾,停止继续向下搜索;发现dcgh与准则1至5不矛盾,则确定hdcgh为分割面。同理,用深度优先搜索二维线画图6a中与准则1至5不矛盾的回路,最后获取二维线画图6a的分割面为jklmj和hdcih。
进一步的,当所述扩展的二维线画图包含第一类辅助边和第二类辅助边时,步骤303还包括:
在获取的所述二维线画图的分割面中,若同一个凹点上出现至少两个分割面,选择包含边的数量最小的分割面或者周长最短的分割面为所述凹点所在的分割面。
例如,如图4所示,采用上述方法获取的二维线画图的分割面为:EcnFE,cnHFC,dABmd,CdmDC,其中,凹点c上出现两个分割面:EcnFE,cnHFC,则删除分割面cnHFC,保留周长最短的分割面EcnFE,同理,保留凹点d所在的dABmd分割面。
304:根据所述分割面对所述二维线画图进行分割,获取所述二维线画图的多个子线画图。
示例性的,所述根据所述分割面对所述二维线画图分割,获取所述二维线画图的多个子线画图,具体包括:
在所述扩展的二维线画图中,将与所述分割面有公共边的表面归入第一面集合;
选取所述第一面集合中的第一表面归入第二面集合,并将所述第一面集合中的所述第一表面删除,其中,所述第一表面为所述第一面集合中的任意一个表面;
在所述第一面集合中,若存在第二表面,与所述第二面集合中的表面有公共边,且所述公共边与所述分割面的一个顶点相连,且所述公共边不属于所述分割面的边,则将所述第二面从所述第一面集合中删除,归入第二面集合;
获取与所述分割面的顶点相连但不属于所述分割面的边的边的集合;
在所述边的集合中,若存在第一边,与所述第一边相连的所述扩展的二维线画图的两个表面与所述第二面集合中的第一表面属于同一面集合,则将所述第一边归入第一边集合,否则,归入第二边集合;
根据所述第一面集合,第二面集合,第一边集合,第二边集合对所述二维线画图进行分割,获取所述二维线画图的多个子线画图。
例如,如图6所示,二维线画图6a的分割面为:jmlkj,dhgcd;以分割面jmlkj为例进行说明:首先找到图6a中与jmlkj有公共边的面,即jmlkj的相邻面:ajkdhea,abcdklmja,sjkts,klotk,mlonm,sjmns,将这6个面归入第一面集合;取出第一面集合中的第一面ajkdhea放入第二面集合,则第一面集合中只剩下{abcdklmja,sjkts,klotk,mlonm,sjmns};发现第一面集合中的面abcdklmja和第一面ajkdhea有公共边:aj和kd,且这两条边与分割面jmlkj的顶点j和k相连,但是aj和kd不是分割面jmlkj的边,则将面abcdklmja从第一面集合中删除,归入第二面集合;而第一面集合中的面klotk,mlonm,sjmns,ajkdhea虽然和第一面abcdklmja有公共边,但是这些公共边也是分割面jmlkj的边,所以将面klotk,mlonm,sjmnsa,jkdhea保留在第一面集合中;最后第一面集合为{klotk,mlonm,sjmnsa,jkdhea}和第二面集合为{ajkdhea,abcdklmja},将第一面集合中的面,二维线画图中与第一面集合中的面相连接的面,以及分割面组成的线画图作为一个子线画图;将第二面集合中的面,二维线画图中与第二面集合中的面相连接的面,以及分割面组成的线画图作为一个子线画图;同理,对于分割面dhgcd采用同样的方法进行分割,则得到图6中6b所示的二维线画图的三个子线画图。
其中,需要说明的是,在获取二维线画图的过程中,步骤302中添加辅助顶点和辅助线的两种方法可以单独使用,也可以混合使用,若添加辅助顶点和辅助线的方法混合使用,本发明实施例中,还可以通过下述方法将二维线画图划分为多个子线画图:
若存在第一凹面,所述第一凹面包含共线边,则在所述第一凹 面上增加第三类辅助边,获取第一扩展的二维线画图,其中,所述第三类辅助边为所述第一凹面的共线边之间的连线;
根据第一扩展的二维线画图,获取所述二维线画图的分割面,
根据所述分割面将所述二维线画图分割为多个子线画图;
在所述子线画图中,若存在子线画图中包含凹面,则分别在所述子线画图的每个凹面上,增加第一类辅助边,其中,所述第一类所述第一类辅助边平行于所述凹面的任一边,所述第一类辅助边的一个端点为所述凹面的凹点,所述第一类辅助边的另一端点为辅助顶点,所述辅助顶点为与所述凹面的边与所述第一类辅助线的交点;
同时,遍历所述子线画图的凸面;若存在第一凸面的不相邻的两个边上,分别包含所述辅助顶点,则在所述第一凸面上增加第二类辅助边,获取扩展的子线画图,其中,所述第二类辅助边的两个端点分别为所述第一凸面的不相邻的两个边上的辅助顶点;
根据扩展的子线画图,获取子线画图的分割面;
根据所述子线画图的分割面再将所述子线画图分割为多个子线画图。
例如,如图7所示,二维线画图7a的凹面utspolkgfedau包含共线边ad和gk,凹面bvwrqnmjhiycbv包含共线边bc和hj,则添加第三类辅助边,dg和ch,获取第一扩展的二维线画图7b;同时,根据步骤303所述的方法获取分割面cdghc,且根据步骤304所述的方法获取二维线画图的两个子线画图7b1和7b2;对子线画图7b2包含凹面autspolka和bvwrqnmjb,则对子线画图7b2添加第一类辅助边sE,sF,pC,pA,rF,rH,qB,qD,添加第二类辅助边EF,GH,CD,AB,得到扩展的子线画图7d;根据7d,采用步骤303所述的获取的方法,获取扩展的子线画图的分割面FEsrF和qpABq;利用分割面FEsrF和qpABq,以及,步骤304所述的方法获取子线画图的多个子线画图:7d1,7d2,7d3;则最终获取到二维线画图7a的子线画图为:7d1,7d2,7d3和7b1。
由上可知,本发明实施例提供的构建三维物体的方法,可以分 割出更加简单的、符合视觉感知规律的子线画图,从而可以高效准确地完成三维物体的重建;避免了现有技术中处理的线画图较为复杂或非常复杂时,系统因解空间的急剧膨胀而陷入局部最优,导致三维重构不准确的缺陷。
下面结合图8描述本发明实施例提供的一种计算机辅助设计系统80,如图8所示,计算机辅助设计系统80包括:
获取模块801,用于获取二维线画图。
其中,二维线画图为由直线线段组合而成的平面图,可以用一个包含顶点和边的二维线画图表示;边和顶点组成二维线画图的面,组成的二维线画图的面可以为凸面或凹面;所述凸面指面上的所有顶点均在该面的凸壳上的面,凸面上的所有顶点均为凸点,凹面指面上的至少一个顶点不在该面的凸壳上的面,凹面上不在所述凹面的凸壳上的点为凹点,在所述凹面的凸壳上的点为凸点;例如,如图2所示,为本发明实施例提供的凸面和凹面的结构示意图,如图2所示,2a为凸面,2b为凹面,顶点A为凸点,顶点B为凹点。优选的,本发明实施例中,用户或者设计者可以使用通过电脑笔、鼠标、手指(在一个触摸屏上)等绘图工具在计算机或者移动终端的屏幕上绘出包含边和顶点的二维线画图。
分割模块802,用于将所述二维线画图分割为多个子线画图。
其中,所述子线画图为所述二维线画图分割后的子图,也为由直线线段组合而成的平面图,一个给定的二维线画图可以分割为若干个子线画图。
三维重构模块803,用于分别对每个子线画图进行三维重构。
重组模块804,用于将所述子线画图的三维重构结果进行重组得到三维物体。
优选的,所述三维重构模块803,具体用于:
根据最小化目标函数
Figure PCTCN2014088311-appb-000007
计算出子线画图中所有顶点的深度值;其中φi(z1,z2,Λ,
Figure PCTCN2014088311-appb-000008
)的建立使用了MSDA(Minimum Standard Deviation of Angles,角的最小标准偏差)、面 的平面性、线的平行性、等距性、角的正交性五条规则,它们的解释如下:MSDA规则要求三维物体在所有顶点上的角的标准偏差最小;面的平面性规则要求在一个面上的所有顶点在三维空间上共平面;线的平行性规则要求在线画图上平行的一对直线在三维空间中也平行;等距性规则要求三维空间中物体的边的比率与它们在线画图上的比率相同;角的正交性要求,如果一个由三条线形成的角在线画图上满足正交的角条件,它在三维空间中的三个角正交。Nv为子线画图的顶点的个数,zi(i=1,2,ΛNv)为第i个顶点的深度座标值,c为规则函数的个数,
Figure PCTCN2014088311-appb-000009
为一组给定的权值,这组权值是通过实验得到,例如,可以使用100,1,20,15,20分别对应以上五个规则。
可以理解的是,上述示例只是为了帮助理解本发明的技术方案,而不应该被视为对本发明技术方案的具体限制。的定义可以使用文献[K.Shoji,K.Kato.And F.Toyama.3-D interpretation of single line drawings based on entropy minimization principle.IEEE Conference on Computer Vision and Pattern Recognition(CVPR),2001]、[y.Leclerc and M.Fischler.An Optimization-Based Approach to the Interpretation of Single Line.Drawings as 3D Wire Frames.International Journal of Computer Vision(IJCV),9(2):113-136,1992]、[H.Lipson and M.Shpitaln.optimization based reconstruction of a 3d object from a single freehand line drawing.Computer-Aided Design,28(7):651-663,1996]、[A.Turner,D.Chapman,and A.Penn.Sketching space.Computer and Graphics,24:869-879,2000]、[J.Liu,L.Cao,Z.Li and X.Tang.Plane-based optimization for 3D object reconstruction from single line drawings.IEEE Transactions on Pattern Analysis and Machine Intelligence(TPAMI),30(2):315-327,2008]中提到的规则函数,比如夹角的标准方差等函数,Nv为子线画图的顶点的个数,zi(i=1,2,ΛNv)为第i个顶点的深度座标值,
Figure PCTCN2014088311-appb-000010
可以是文献[K.Shoji,K.Kato.And F.Toyama.3-D interpretation of single line drawings based on  entropy minimization principle.IEEE Conference on Computer Vision and Pattern Recognition(CVPR),2001]、[y.Leclerc and M.Fischler.An Optimization-Based Approach to the Interpretation of Single Line.Drawings as 3D Wire Frames.International Journal of Computer Vision(IJCV),9(2):113-136,1992]、[H.Lipson and M.Shpitaln.optimization based reconstruction of a 3d object from a single freehand line drawing.Computer-Aided Design,28(7):651-663,1996]、[A.Turner,D.Chapman,and A.Penn.Sketching space.Computer and Graphics,24:869-879,2000]、[J.Liu,L.Cao,Z.Li and X.Tang.Plane-based optimization for 3D object reconstruction from single line drawings.IEEE Transactions on Pattern Analysis and Machine Intelligence(TPAMI),30(2):315-327,2008]等中提到的权值,c为规则函数的个数。
进一步的,所述重组模块804,具体用于:
通过原始二维线画图的得到各个三维重构结果的结合部位信息,同时通过比较结合部位的尺寸重新调整需要结合的两个部分的大小,把所有子物体结合为整体之后,再根据最小化目标函数
Figure PCTCN2014088311-appb-000011
调整重组的三维物体的大小;其中Nv为重组的三维物体的顶点的个数。
可以理解的是,三维重构结果之间的结合可以是点点出结合,也可以是点线处结合,例如需要结合的两个三维重构结果可能结合部位是第一个子线画图的三维重构结果的一个点和第二个子线画图的三维重构结果的一个点;也可能是第一个子线画图的三维重构结果的一个点和第二个子线画图的三维重构结果的一条线的某个位置;也可能是第一个子线画图的三维重构结果的一个点和第二个子线画图的三维重构结果的一个面上的某个位置;也可能是第一个子线画图的三维重构结果的一条线和第二个子线画图的三维重构结果 的一条线的某个位置;也可能是第一个子线画图的三维重构结果的一条线和第二个子线画图的三维重构结果的一个面的某个位置;也可以第一个子线画图的三维重构结果的一个面和第二个子线画图的三维重构结果的一个面的结合。这种结合可以是两个子物体的真正拼接在一起,也可以是从一个子物体中挖除另外一个子物体或者第一子物体的一部分,例如当按照拼接规则实施之后,发现第二子物体实际在第一子物体内部(即内部空间重合),则应该从第一子物体中挖除掉第二子物体与其在空间上重叠的部分。
不难发现,在计算机辅助设计系统利用二维线画图构建三维物体的基本过程中,若分割后的子线画图比较复杂,则会造成子线画图的三维重建计算复杂且不够准确,从而使得重组后的三维物体不够准确。
为此,本发明实施例中,需要找到更符合人们思维方式的,简单的,更易于进行三维重构的子线画图,以使得三维物体的重构高效准确;优选的,本发明实施例中,子线画图应该不包含凹面且子线画图应该包含可能少的面。
下面结合图9,对本发明实时提供的一种分割模块802的进行描述,如图9所示,包括:
提取单元8021,用于提取所述二维线画图的表面。
其中,所述表面为所述二维线画图中处于同一平面的边围成的区域,为凸面或凹面。
添加单元8022,用于若所述二维线画图包含凹面,则根据预设策略,在二维线画图上添加辅助顶点,以及,辅助边,获取扩展的二维线画图。
其中,所述辅助顶点为除所述二维线画图的原始顶点之外的新的顶点,所述辅助边为辅助顶点与二维线画图的原始顶点之间的连线,或者,为二维线画图的两个原始顶点之间的连线,或者,为两个辅助顶点之间的连线;所述扩展的二维线画图为包含所述辅助顶点和辅助边的二维线画图,所述扩展的二维线画图的表面均为凸面, 所述凸面为所述二维线画图的表面,所述凸面的所有顶点均在所述凸面的凸壳上。
分割面获取单元8023,用于根据所述扩展的二维线画图,获取所述二维线画图的分割面。
其中,所述分割面二维线画图上的一条回路,通过对二维线画图的切割而形成,分割面既包括二维线画图的原始边,也包括一些在二维线画图表面上添加的边(即辅助边)。
子图获取单元8024,用于根据所述分割面对所述二维线画图进行分割,获取所述二维线画图的多个子线画图。
进一步的,所述添加单元8022,具体用于下述两种情况中:
(1)分别在所述二维线画图的每个凹面上,增加第一类辅助边,其中,所述第一类辅助边平行于所述凹面的任一边,所述第一类辅助边的一个端点为所述凹面的凹点,所述第一类辅助边的另一端点为辅助顶点,所述辅助顶点为与所述凹面的边与所述第一类辅助线的交点;
遍历所述二维线画图的凸面;
若存在第一凸面的不相邻的两个边上,分别包含所述辅助顶点,则在所述第一凸面上增加第二类辅助边,其中,所述第二类辅助边的两个端点分别为所述第一凸面的不相邻的两个边上的辅助顶点。
例如,如图4所示,二维线画图4a的凹面有:abcdefgha,ijklmnopi,在凹面abcdefgha,以凹点c为端点作一条平行于凹面的边ha的平行线cG,其中,cG与凹面的边hg相交,则cG属于第一类辅助线,同理,遍历所有凹面的凹点增加第一类辅助边:cG、cE、dC、dA、nF、nH、mB、mJ;此时,在二维线画图4a的凸面hgjih存在第一类辅助边的端点C、D、G、H,在凸面gfkjg上存在第一类辅助边的端点A、B,在凸面hapih上存在第一类辅助边的端点E、F,分别将同一凸面上的所述端点进行连接,获取第二类辅助边:GH、CD、AB、EF,得到二维线画图4a的扩展线画图4b。
(2)若存在第一凹面,所述第一凹面包含共线边,则在所述第 一凹面上增加第三类辅助边,其中,所述第三类辅助边为所述第一凹面的共线边之间的连线。
其中,所述共线边为同一平面上在同一直线的不同边。
例如,如图5所示,二维线画图5a中包含凹面acfgjknoa,bdehilmpb,其中,凹面acfgjknoa包含共线边on和gf,将共线边on和gf连接,则连接线ng为第三类辅助边,同理,在凹面bdehilmpb上添加第三类辅助边mh。
进一步的,所述分割面获取单元8023,具体用于:
遍历所述扩展的二维线画图中的边;
若存在第一边的两顶点中至少一个顶点的度不小于4,则将所述第一边归入第一边集合;其中,所述顶点的度为与所述顶点连接的边的个数;
对所述第一边集合中的每一个边,分别采用深度优先搜索算法获取从所述边的一端点到所述边的另一端点的回路;
若搜索到的回路与预设准则不矛盾,则将所述回路确定为所述二维线画图的分割面的回路,获取分割面;其中,所述预设准则包含以下5个准则:
(1)若回路包含一个度为3的3D凸点,则所述回路不是分割面的回路;其中,所述3D凸点为一个顶点,且所述凸点所在的面均为凸面。
(2)若回路经过两个相邻面f1、f2,且f1、f2共同拥有两个或两个以上不在同一直线上的边,则所述回路不是分割面的回路;
(3)若回路的边自相交,则所述回路不是分割面的回路;
(4)若回路包含有一个弦,且该弦在二维线画图的表面上,则所述回路不是分割面的回路;其中,所述弦为连接回路的上两个不相邻的点,且在回路内部的线段。
(5)若回路包含有两个不共线的边,且两个边是同一个面f的边、以及,所述回路围城的区域和f之间有重叠区域,则所述回路不是分割面的回路。
例如,如图6所示,获取二维线画图6a的分割面的过程为:顶点d的度等于4,选择hd这条线段,搜索从端点d到端点h的回路,首先从d点出发搜索:dc,dk,di;发现dc与准则1至5都不矛盾,所以从dc继续往下做深度优先搜索,得到:dcb,dci,dcg;发现dcb中的b是一个3d凸顶点,与准则1矛盾,则从dcb不再往下做深度优先搜索;dcg和dci与准则1至5不矛盾,则继续做深度优选搜索,从dci往下搜索到dcih和dcig:发现dcih中出现了弦di在回路hdcih之内,与准则4矛盾,故回路hdcih不是分割面;发现dcig出现了弦gc在回路hdcig之内,与准则4矛盾,故回路hdcih不是分割面;从dcg往下搜索到dcgh和dcgf,发现dcgf中的f点为3D凸点,与准则1矛盾,停止继续向下搜索;发现dcgh与准则1至5不矛盾,则确定hdcgh为分割面。同理,用深度优先搜索二维线画图6a中与准则1至5不矛盾的回路,最后获取二维线画图6a的分割面为jklmj和hdcih。
进一步的,当所述扩展的二维线画图包含第一类辅助边和第二类辅助边时,所述分割面获取单元8023,还用于:
在获取的所述二维线画图的分割面中,若同一个凹点上出现至少两个分割面,选择包含边的数量最小的分割面或者周长最短的分割面为所述凹点所在的分割面。
例如,如图4所示,采用上述方法获取的二维线画图的分割面为:EcnFE,cnHFC,dABmd,CdmDC,其中,凹点c上出现两个分割面:EcnFE,cnHFC,则删除分割面cnHFC,保留周长最短的分割面EcnFE,同理,保留凹点d所在的dABmd分割面。
进一步的,所述子图获取单元8024,具体用于:
在所述扩展的二维线画图中,将与所述分割面有公共边的表面归入第一面集合;
选取所述第一面集合中的第一表面归入第二面集合,并将所述第一面集合中的所述第一表面删除,其中,所述第一表面为所述第一面集合中的任意一个表面;
在所述第一面集合中,若存在第二表面,与所述第二面集合中的表面有公共边,且所述公共边与所述分割面的一个顶点相连,且所述公共边不属于所述分割面的边,则将所述第二面从所述第一面集合中删除,归入第二面集合;
获取与所述分割面的顶点相连但不属于所述分割面的边的边的集合;
在所述边的集合中,若存在第一边,与所述第一边相连的所述扩展的二维线画图的两个表面与所述第二面集合中的第一表面属于同一面集合,则将所述第一边归入第一边集合,否则,归入第二边集合;
根据所述第一面集合,第二面集合,第一边集合,第二边集合对所述二维线画图进行分割,获取所述二维线画图的多个子线画图。
例如,如图6所示,二维线画图6a的分割面为:jmlkj,dhgcd;以分割面jmlkj为例进行说明:首先找到图6a中与jmlkj有公共边的面,即jmlkj的相邻面:ajkdhea,abcdklmja,sjkts,klotk,mlonm,sjmns,将这6个面归入第一面集合;取出第一面集合中的第一面ajkdhea放入第二面集合,则第一面集合中只剩下{abcdklmja,sjkts,klotk,mlonm,sjmns};发现第一面集合中的面abcdklmja和第一面ajkdhea有公共边:aj和kd,且这两条边与分割面jmlkj的顶点j和k相连,但是aj和kd不是分割面jmlkj的边,则将面abcdklmja从第一面集合中删除,归入第二面集合;而第一面集合中的面klotk,mlonm,sjmns,ajkdhea虽然和第一面abcdklmja有公共边,但是这些公共边也是分割面jmlkj的边,所以将面klotk,mlonm,sjmnsa,jkdhea保留在第一面集合中;最后第一面集合为{klotk,mlonm,sjmnsa,jkdhea}和第二面集合为{ajkdhea,abcdklmja},将第一面集合中的面,二维线画图中与第一面集合中的面相连接的面,以及分割面组成的线画图作为一个子线画图;将第二面集合中的面,二维线画图中与第二面集合中的面相连接的面,以及分割面组成的线画图作为一个子线画图;同理,对于分割面dhgcd采用同样的方法进 行分割,则得到图6中6b所示的二维线画图的三个子线画图。
需要说明的是,本实施例中的获取模块可以为计算机辅助设计系统中单独设立的处理器,也可以为集成在计算机辅助设计系统中的某一个处理器,此外,也可以以程序代码的形式存储于基站的存储器中,由基站的某一个处理器调用并执行以上获取模块的功能;分割模块与获取模块相似,且可以与获取模块集成在一起,也可以独立实现;三维重构模块与获取模块和分割模块相似,且可以与获取模块和分割模块集成在一起,也可以独立实现;重组模块与获取模块、分割模块、三维重构模块相似,且可以与获取模块、分割模块、三维重构模块集成在一起,也可以独立实现。这里所述的处理器可以是一个中央处理器(Central Processing Unit,CPU),或者是特定集成电路(Application Specific Integrated Circuit,ASIC),或者是被配置成实施本发明实施例的一个或多个集成电路。
由上可知,本发明实施例提供一种构建三维物体的方法及计算机辅助设计系统,获取二维线画图;将所述二维线画图分割为多个子线画图;分别对每个子线画图进行三维重构;将所述子线画图的三维重构结果进行重组得到三维物体,其中,所述将所述二维线画图分割为多个子线画图,具体包括:提取所述二维线画图的表面,其中,所述表面为所述二维线画图中处于同一平面的边围成的区域,为凸面或凹面;若所述二维线画图包含凹面,则根据预设策略,在二维线画图上添加辅助顶点,以及,辅助边,获取扩展的二维线画图;根据所述扩展的二维线画图,获取所述二维线画图的分割面;根据所述分割面对所述二维线画图进行分割,获取所述二维线画图的多个子线画图。如此,对二维线画图进行处理,分割出更加简单的、符合视觉感知规律的子线画图,高效准确地完成三维物体的重建;避免了现有技术中处理的线画图较为复杂或非常复杂时,系统因解空间的急剧膨胀而陷入局部最优,导致三维重构不准确的缺陷。
在本申请所提供的几个实施例中,应该理解到,所揭露的系统,装置和方法,可以通过其它的方式实现。例如,以上所描述的装置 实施例仅仅是示意性的,例如,所述单元的划分,仅仅为一种逻辑功能划分,实际实现时可以有另外的划分方式,例如多个单元或组件可以结合或者可以集成到另一个系统,或一些特征可以忽略,或不执行。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的,作为单元显示的部件可以是或者也可以不是物理单元,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外,在本发明各个实施例中的各功能单元可以集成在一个处理单元中,也可以是各个单元单独物理包括,也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现,也可以采用硬件加软件功能单元的形式实现。
最后应说明的是:以上实施例仅用以说明本发明的技术方案,而非对其限制;尽管参照前述实施例对本发明进行了详细的说明,本领域的普通技术人员应当理解:其依然可以对前述各实施例所记载的技术方案进行修改,或者对其中部分技术特征进行等同替换;而这些修改或者替换,并不使相应技术方案的本质脱离本发明各实施例技术方案的精神和范围。

Claims (12)

  1. 一种构建三维物体的方法,包括:获取二维线画图;将所述二维线画图分割为多个子线画图;分别对每个子线画图进行三维重构;将所述子线画图的三维重构结果进行重组得到三维物体,其特征在于,所述将所述二维线画图分割为多个子线画图,具体包括:
    提取所述二维线画图的表面,其中,所述表面为所述二维线画图中处于同一平面的边围成的区域,为凹面或凸面;
    若所述二维线画图包含凹面,则根据预设策略,在二维线画图上添加辅助顶点,以及,辅助边,获取扩展的二维线画图;
    根据所述扩展的二维线画图,获取所述二维线画图的分割面;
    根据所述分割面对所述二维线画图进行分割,获取所述二维线画图的多个子线画图。
  2. 根据权利要求1所述的方法,其特征在于,所述根据预设策略,在二维线画图上添加辅助顶点,以及,辅助边,获取扩展的二维线画图,包括:
    分别在所述二维线画图的每个凹面上,增加第一类辅助边,其中,所述第一类辅助边平行于所述凹面的任一边,所述第一类辅助边的一个端点为所述凹面的凹点,所述第一类辅助边的另一端点为辅助顶点,所述辅助顶点为与所述凹面的边与所述第一类辅助线的交点;
    遍历所述二维线画图的凸面;
    若存在第一凸面的不相邻的两个边上,分别包含所述辅助顶点,则在所述第一凸面上增加第二类辅助边,其中,所述第二类辅助边的两个端点分别为所述第一凸面的不相邻的两个边上的辅助顶点。
  3. 根据权利要求1所述的方法,其特征在于,所述根据预设策略,在二维线画图上添加辅助顶点,以及,辅助边,获取扩展的二维线画图,包括:
    若存在第一凹面,所述第一凹面包含共线边,则在所述第一凹面上增加第三类辅助边,其中,所述第三类辅助边为所述第一凹面的共线边之间的连线。
  4. 根据权利要求2或3所述的方法,其特征在于,所述根据所述扩展的二维线画图,获取所述二维线画图的分割面,包括:
    遍历所述扩展的二维线画图中的边;
    若存在第一边的两顶点中至少一个顶点的度不小于4,则将所述第一边归入第一边集合;
    对所述第一边集合中的每一个边,分别采用深度优先搜索算法获取从所述边的一端点到所述边的另一端点的回路;
    若搜索到的回路与预设准则不矛盾,则将所述回路确定为所述二维线画图的分割面的回路,获取分割面;其中,所述预设准则包括:若回路包含一个度为3的3D凸点,则所述回路不是分割面的回路;若回路经过两个相邻面f1、f2,且f1、f2共同拥有两个或两个以上不在同一直线上的边,则所述回路不是分割面的回路;若回路的边自相交,则所述回路不是分割面的回路;若回路包含有一个弦,且该弦在二维线画图的表面上,则所述回路不是分割面的回路;若回路包含有两个不共线的边,且两个边是同一个面f的边、以及,所述回路围城的区域和f之间有重叠区域,则所述回路不是分割面的回路。
  5. 根据权利要求4所述的方法,其特征在于,所述方法还包括:
    在获取的所述二维线画图的分割面中,若同一个凹点上出现至少两个分割面,则选择包含边的数量最小的分割面或者周长最短的分割面为所述凹点所在的分割面。
  6. 根据权利要求1-5任一项所述的方法,其特征在于,所述根据所述分割面对所述二维线画图进行分割,获取所述二维线画图的多个子线画图,包括:
    在所述扩展的二维线画图中,将与所述分割面有公共边的表面归入第一面集合;
    选取所述第一面集合中的第一表面归入第二面集合,并将所述第一面集合中的所述第一表面删除,其中,所述第一表面为所述第一面集合中的任意一个表面;
    在所述第一面集合中,若存在第二表面,与所述第二面集合中的 表面有公共边,且所述公共边与所述分割面的一个顶点相连,且所述公共边不属于所述分割面的边,则将所述第二面从所述第一面集合中删除,归入第二面集合;
    获取与所述分割面的顶点相连但不属于所述分割面的边的边的集合;
    在所述边的集合中,若存在第一边,与所述第一边相连的所述扩展的二维线画图的两个表面与所述第二面集合中的第一表面属于同一面集合,则将所述第一边归入第一边集合,否则,归入第二边集合;
    根据所述第一面集合,第二面集合,第一边集合,第二边集合对所述二维线画图进行分割,获取所述二维线画图的多个子线画图。
  7. 一种计算机辅助设计系统,包括:获取模块,用于获取二维线画图;分割模块,用于将所述二维线画图分割为多个子线画图;三维重构模块,用于分别对每个子线画图进行三维重构;重组模块,用于将所述子线画图的三维重构结果进行重组得到三维物体,其特征在于,所述分割模块包括:
    凹面获取单元,用于提取所述二维线画图的表面,其中,所述表面为所述二维线画图中处于同一平面的边围成的区域,为凸面或凹面;
    添加单元,用于若所述二维线画图包含凹面,则根据预设策略,在二维线画图上添加辅助顶点,以及,辅助边,获取扩展的二维线画图;
    分割面获取单元,用于根据所述扩展的二维线画图,获取所述二维线画图的分割面;
    子图获取单元,根据所述分割面对所述二维线画图进行分割,获取所述二维线画图的多个子线画图。
  8. 根据权利要求7所述的计算机辅助设计系统,其特征在于,所述添加单元,具体用于:
    分别在所述二维线画图的每个凹面上,增加第一类辅助边,其中,所述第一类辅助边平行于所述凹面的任一边,所述第一类辅助边的一 个端点为所述凹面的凹点,所述第一类辅助边的另一端点为辅助顶点,所述辅助顶点为与所述凹面的边与所述第一类辅助线的交点;
    遍历所述二维线画图的凸面;
    若存在第一凸面的不相邻的两个边上,分别包含所述辅助顶点,则在所述第一凸面上增加第二类辅助边,其中,所述第二类辅助边的两个端点分别为所述第一凸面的不相邻的两个边上的辅助顶点。
  9. 根据权利要求7所述的计算机辅助设计系统,其特征在于,所述添加单元,具体用于:
    若存在第一凹面,所述第一凹面包含共线边,则在所述第一凹面上增加第三类辅助边,其中,所述第三类辅助边为所述第一凹面的共线边之间的连线。
  10. 根据权利要求8或9所述计算机辅助设计系统,其特征在于,所述分割面获取单元,具体用于:
    遍历所述扩展的二维线画图中的边;
    若存在第一边的两顶点中至少一个顶点的度不小于4,则将所述第一边归入第一边集合;
    对所述第一边集合中的每一个边,分别采用深度优先搜索算法获取从所述边的一端点到所述边的另一端点的回路;
    若搜索到的回路与预设准则不矛盾,则将所述回路确定为所述二维线画图的分割面的回路,获取分割面;其中,所述预设准则包括:若回路包含一个度为3的3D凸点,则所述回路不是分割面的回路;若回路经过两个相邻面f1、f2,且f1、f2共同拥有两个或两个以上不在同一直线上的边,则所述回路不是分割面的回路;若回路的边自相交,则所述回路不是分割面的回路;若回路包含有一个弦,且该弦在二维线画图的表面上,则所述回路不是分割面的回路;若回路包含有两个不共线的边,且两个边是同一个面f的边、以及,所述回路围城的区域和f之间有重叠区域,则所述回路不是分割面的回路。
  11. 根据权利要求10所述的计算机辅助设计系统,其特征在于,所述分割面获取单元,还用于:
    在获取的所述二维线画图的分割面中,若同一个凹点上出现至少两个分割面,则选择包含边的数量最小的分割面或者周长最短的分割面为所述凹点所在的分割面。
  12. 根据权利要求7-11任一项所述的计算机辅助设计系统,其特征在于,所述子图获取单元,具体用于:
    在所述扩展的二维线画图中,将与所述分割面有公共边的表面归入第一面集合;
    选取所述第一面集合中的第一表面归入第二面集合,并将所述第一面集合中的所述第一表面删除,其中,所述第一表面为所述第一面集合中的任意一个表面;
    在所述第一面集合中,若存在第二表面,与所述第二面集合中的表面有公共边,且所述公共边与所述分割面的一个顶点相连,且所述公共边不属于所述分割面的边,则将所述第二面从所述第一面集合中删除,归入第二面集合;
    获取与所述分割面的顶点相连但不属于所述分割面的边的边的集合;
    在所述边的集合中,若存在第一边,与所述第一边相连的所述扩展的二维线画图的两个表面与所述第二面集合中的第一表面属于同一面集合,则将所述第一边归入第一边集合,否则,归入第二边集合;
    根据所述第一面集合,第二面集合,第一边集合,第二边集合对所述二维线画图进行分割,获取所述二维线画图的多个子线画图。
PCT/CN2014/088311 2014-05-16 2014-10-10 一种构建三维物体的方法及计算机辅助设计系统 WO2015172504A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201410209597.0A CN105096378B (zh) 2014-05-16 2014-05-16 一种构建三维物体的方法及计算机辅助设计系统
CN201410209597.0 2014-05-16

Publications (1)

Publication Number Publication Date
WO2015172504A1 true WO2015172504A1 (zh) 2015-11-19

Family

ID=54479252

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2014/088311 WO2015172504A1 (zh) 2014-05-16 2014-10-10 一种构建三维物体的方法及计算机辅助设计系统

Country Status (2)

Country Link
CN (1) CN105096378B (zh)
WO (1) WO2015172504A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108496127A (zh) * 2016-01-22 2018-09-04 高通股份有限公司 集中于对象的有效三维重构

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109493376B (zh) * 2017-09-13 2022-02-22 腾讯科技(深圳)有限公司 图像处理方法和装置、存储介质及电子装置

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101799937A (zh) * 2010-03-23 2010-08-11 南京大学 一种采用草图创建三维模型的方法
CN102855667A (zh) * 2011-06-30 2013-01-02 中国科学院深圳先进技术研究院 三维物体的计算机辅助设计系统及方法
CN103473811A (zh) * 2013-08-20 2013-12-25 浙江工业大学 基于二维手绘线画图的三维实体模型便捷生成方法

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4855939A (en) * 1987-09-11 1989-08-08 International Business Machines Corp. 3D Dimensioning in computer aided drafting
US20020067354A1 (en) * 2000-10-02 2002-06-06 Beom-Soo Oh Method and system for reconstructing 3D objects from free-hand line drawing
US20080297503A1 (en) * 2007-05-30 2008-12-04 John Dickinson System and method for reconstructing a 3D solid model from a 2D line drawing

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101799937A (zh) * 2010-03-23 2010-08-11 南京大学 一种采用草图创建三维模型的方法
CN102855667A (zh) * 2011-06-30 2013-01-02 中国科学院深圳先进技术研究院 三维物体的计算机辅助设计系统及方法
CN103473811A (zh) * 2013-08-20 2013-12-25 浙江工业大学 基于二维手绘线画图的三维实体模型便捷生成方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
LIU, JIANZHUANG ET AL.: "Decomposition of Complex Line Drawings with Hidden Lines for 3D Planar-faced Manifold Object Reconstruction", IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, vol. 33, no. 1, 31 January 2011 (2011-01-31), pages 4, XP011373487 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108496127A (zh) * 2016-01-22 2018-09-04 高通股份有限公司 集中于对象的有效三维重构

Also Published As

Publication number Publication date
CN105096378A (zh) 2015-11-25
CN105096378B (zh) 2018-04-10

Similar Documents

Publication Publication Date Title
Chen et al. 3-sweep: Extracting editable objects from a single photo
KR102040988B1 (ko) 복수의 3d 곡선들로부터의 표면 생성
JP5445199B2 (ja) 三次元モデル分割装置及び三次元モデル分割方法
Chang et al. 3D shape engineering and design parameterization
KR102055955B1 (ko) 강체 운동들에 의해 변환된 기하학적 엘리먼트들
CN101937579B (zh) 一种利用透视草图创建三维曲面模型的方法
US9898860B2 (en) Method, apparatus and terminal for reconstructing three-dimensional object
US20080040080A1 (en) System and Method for Identifying Original Design Intents Using 3D Scan Data
Tierny et al. Interactive quadrangulation with Reeb atlases and connectivity textures
Beccari et al. A fast interactive reverse-engineering system
WO2012071688A1 (zh) 基干感知信息的三维模型形狀分析方法
KR20130004066A (ko) 기하학적 3차원 모델링된 오브젝트를 디자인하는 방법
TWI425442B (zh) Method of Reconstructing Three - dimensional Housing Model on Aeronautical Mapping System
CN107680154B (zh) 基于视图的体素几何参数提取方法
Li et al. Geometric structure simplification of 3D building models
Gao et al. Structured volume decomposition via generalized sweeping
WO2015172504A1 (zh) 一种构建三维物体的方法及计算机辅助设计系统
Nieser et al. Patch layout from feature graphs
Chong et al. Automatic mesh-healing technique for model repair and finite element model generation
Knott et al. Towards mesh-based deep learning for semantic segmentation in photogrammetry
CN114329666A (zh) 用于解析并导入三维管材刀路的方法、装置、处理器及其计算机可读存储介质
CN102855667A (zh) 三维物体的计算机辅助设计系统及方法
Delamé et al. From a medial surface to a mesh
Liu et al. Sketch simplification guided by complex agglomeration
US10832454B1 (en) Edit propagation on raster sketches

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 14891890

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 14891890

Country of ref document: EP

Kind code of ref document: A1