Method, device and computer program for generating space model

By extracting skeletal information and defining nodes and edges from drawing data, the method generates accurate 3D spatial models of complex structures, addressing the limitations of existing vectorization techniques and enabling user-friendly modeling of non-standard architectural features.

KR103044467B1Active Publication Date: 2026-11-06KT CORP
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
KR1020220138177
Authority / Receiving Office
KR · KR
Patent Type
Patents
Current Assignee / Owner
Filing Date
2022-10-25
Publication Date
2026-11-06
Estimated Expiration
2042-10-25

AI Technical Summary

Technical Problem

Existing vectorization methods for floor plans, especially those involving non-horizontal or vertical walls and circular structures, are costly and difficult to model accurately using CAD software, and drawing analysis techniques struggle to utilize classified spatial structures effectively for modeling.

Method used

A method and device for generating a spatial model by extracting skeletal information from drawing data, defining nodes and edges based on this information, and creating a spatial model using nodes and edges.

Benefits of technology

Enables ordinary users to create accurate three-dimensional spatial models of complex structures like houses with curved walls and multi-story buildings, facilitating the generation of highly detailed 3D models.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 112022112534756-PAT00001_ABST
    Figure 112022112534756-PAT00001_ABST
Patent Text Reader

Abstract

A method for generating a spatial model may include the steps of: receiving drawing data for a target space; extracting skeletal information for the target space from the drawing data; defining a plurality of nodes corresponding to corner points or intersection points for the target space based on the skeletal information; defining an edge connecting at least two adjacent nodes among the plurality of nodes; and generating a spatial model for the target space based on information about the nodes and information about the edges.
Need to check novelty before this filing date? Find Prior Art

Description

Technology Field

[0001] The present invention relates to a method, apparatus, and computer program for generating a spatial model. Background Technology

[0002] Vectorization of a drawing refers to the process of converting multiple walls (i.e., the walls forming a room) that constitute a floor plan into one or more vectors. While the most accurate method for this vectorization is for a professional to work on it while viewing the floor plan using CAD software, it is costly.

[0003] Meanwhile, drawing analysis technology is a technique that analyzes drawing images to classify multiple spatial structures included in the images into categories at the pixel level. However, the images corresponding to each spatial structure within each category classified by this drawing analysis technology are difficult to utilize for modeling.

[0004] Furthermore, vectorization using drawing analysis techniques becomes difficult to model when the structure consists of non-horizontal or vertical walls, such as diagonal walls or circular structures, as complexity increases. Prior art literature

[0005] Korean Registered Patent Publication No. 10-1190491 (Published Oct. 12, 2012) The problem to be solved

[0006] The present invention aims to solve the problems of the aforementioned prior art by extracting skeletal information about a target space from drawing data and defining nodes and edges based on the skeletal information to generate a spatial model.

[0007] However, the technical problems that this embodiment aims to solve are not limited to the technical problems described above, and other technical problems may exist. means of solving the problem

[0008] As a technical means for achieving the aforementioned technical problem, a method for generating a spatial model of the present invention may include: receiving drawing data for a target space; extracting skeletal information for the target space from the drawing data; defining a plurality of nodes corresponding to corner points or intersection points for the target space based on the skeletal information; defining an edge connecting at least two adjacent nodes among the plurality of nodes; and generating a spatial model for the target space based on information about the nodes and information about the edges.

[0009] The device for generating a spatial model of the present invention may include: an input unit for receiving drawing data for a target space; a skeletal information extraction unit for extracting skeletal information for the target space from the drawing data; a structure definition unit for defining a plurality of nodes corresponding to corner points or intersection points for the target space based on the skeletal information, and defining an edge connecting at least two adjacent nodes among the plurality of nodes; and a spatial model generation unit for generating a spatial model for the target space based on information about the nodes and information about the edges.

[0010] A computer program stored in a computer-readable recording medium comprising a sequence of instructions for generating a spatial model of the present invention may include, when executed by a computing device, a sequence of instructions for receiving drawing data for a target space, extracting skeletal information for the target space from the drawing data, defining a plurality of nodes corresponding to corner points or intersection points of the target space based on the skeletal information, defining an edge connecting at least two adjacent nodes among the plurality of nodes, and generating a spatial model for the target space based on information about the nodes and information about the edges.

[0011] The above-described means for solving the problem are merely exemplary and should not be interpreted as intended to limit the invention. In addition to the exemplary embodiments described above, additional embodiments described in the drawings and the detailed description of the invention may exist. Effects of the invention

[0012] According to any one of the means for solving the problem of the present invention described above, the present invention can generate a spatial model by extracting skeletal information about a target space from drawing data and defining nodes and edges based on the skeletal information.

[0013] Through this, the present invention can help ordinary people who are not experts to create a three-dimensional spatial model of a drawing.

[0014] In addition, the present invention can be utilized as a core technology for home twin technology that creates a 3D model identical to one's own house structure, and can generate highly accurate modeling for various types of drawings, such as houses with curved walls, multi-story structures, and single-family homes. Brief explanation of the drawing

[0015] FIG. 1 is a block diagram of a spatial model generation device according to one embodiment of the present invention. FIGS. 2a and 2b are drawings illustrating a method for predicting object information from drawing data based on a prediction model according to an embodiment of the present invention. FIGS. 3a and 3b are drawings for explaining a method of generating object information for doors and windows on object information for walls according to an embodiment of the present invention. FIGS. 4a to 4e are drawings for explaining a method of extracting skeletal information about a target space from object information (220) about a target space according to an embodiment of the present invention and defining it as nodes and edges. FIGS. 5a and 5b are drawings for explaining a method of generating vector data for a plurality of objects included in a target space based on information about nodes and edges. FIGS. 6a to 6c are drawings for explaining how to edit a room structure by editing nodes and edges for a target space. FIG. 7 is a diagram illustrating a method for generating a spatial model by applying height information to information about a target space according to an embodiment of the present invention. FIG. 8 is a diagram illustrating a method for creating nodes for each layer of stairs and connecting nodes with edges according to an embodiment of the present invention. FIG. 9 is a drawing of a spatial model generated based on drawing data according to one embodiment of the present invention. FIG. 10 is a flowchart illustrating a method for generating a spatial model according to an embodiment of the present invention. Specific details for implementing the invention

[0016] Embodiments of the present invention are described below in detail with reference to the attached drawings so that those skilled in the art can easily implement the invention. However, the present invention may be embodied in various different forms and is not limited to the embodiments described herein. Furthermore, in order to clearly explain the present invention in the drawings, parts unrelated to the explanation have been omitted, and similar parts throughout the specification are denoted by similar reference numerals.

[0017] Throughout the specification, when a part is described as being "connected" to another part, this includes not only cases where they are "directly connected," but also cases where they are "electrically connected" with other components interposed between them. Furthermore, when a part is described as "including" a certain component, this means that, unless specifically stated otherwise, it does not exclude other components but may include additional components.

[0018] In this specification, the term "part" includes a unit realized by hardware, a unit realized by software, and a unit realized using both. Additionally, one unit may be realized using two or more hardware, and two or more units may be realized by one hardware.

[0019] Some of the operations or functions described in this specification as being performed by a terminal or device may instead be performed by a server connected to said terminal or device. Likewise, some of the operations or functions described as being performed by a server may also be performed by a terminal or device connected to said server.

[0020] Hereinafter, specific details for implementing the present invention will be described with reference to the attached configuration diagram or process flowchart.

[0021] FIG. 1 is a block diagram of a spatial model generation device (100) according to one embodiment of the present invention.

[0022] Referring to FIG. 1, the spatial model generation device (100) may include an input unit (110), a prediction unit (120), an object information extraction unit (130), a skeleton information extraction unit (140), a structure definition unit (150), and a spatial model generation unit (160).

[0023] However, the spatial model generating device (100) illustrated in FIG. 1 is merely one embodiment of the present invention, and various modifications are possible based on the components illustrated in FIG. 1.

[0024] The input unit (110) can receive drawing data for the target space. For example, the drawing data may include information about the floor plan of the target space. As another example, the drawing data may further include drawing type information (basic type, extended type), number of floors information, exclusive area information, size information of the drawing data, and address information for the target space.

[0025] FIGS. 2a and 2b are drawings illustrating a method for predicting object information from drawing data (210) based on a prediction model according to an embodiment of the present invention.

[0026] Referring to FIGS. 2a and 2b, the prediction unit (120) of the spatial model generation device (100) can predict object information from drawing data based on the prediction model.

[0027] The prediction unit (120) can perform data annotation on the training drawing data and input the data annotated training drawing data into the prediction model to train the prediction model. Accordingly, the prediction unit (120) can predict object information (220) for a target space by inputting unidentified drawing data (210) into the trained prediction model. For example, the object information (220) for a target space may include object information (300) for a door included in the target space, object information for a window included in the target space, object information for a wall included in the target space, and object information for a room included in the target space.

[0028] FIGS. 3a and 3b are drawings for explaining a method of generating object information (300) for a door and a window on object information for a wall according to an embodiment of the present invention.

[0029] FIG. 3a is a diagram showing a method for redefining object information for a door when the line where object information for a wall and object information for a door intersect is a straight line according to an embodiment of the present invention, and FIG. 3b is a diagram showing a method for redefining object information for a door when the line where object information for a wall and object information for a door intersect is not a straight line according to an embodiment of the present invention.

[0030] Referring to FIGS. 3a and 3b, the object information extraction unit (130) can use a prediction model to extract object information for at least one of a door, window, and wall included in the target space from the drawing data (210).

[0031] For example, the object information extraction unit (130) can separate object information for each door, window, and wall based on drawing images corresponding to each door, window, and wall included in the drawing data (210), and derive object information for a room through the separated object information for the wall.

[0032] Object information about a room can be distinguished by different colors depending on the room type. For example, a bedroom can be indicated by the first color, a living room by the second color, a balcony by the third color, a bathroom by the fourth color, and so on.

[0033] Meanwhile, the object information extraction unit (130) can generate object information (300) for a door on object information for a wall, and redefine the object information (300) for a door according to the type of intersection line (320) between the object information for a wall and the object information (300) for a door.

[0034] Here, when redefining the object information (300) for the door, the object information (300) for the door can be expanded through an expansion operation to make it easier to extract the object information (300) for the door.

[0035] For example, when the intersection line (320) where the object information for the wall and the object information for the door (300) meet is a straight line, as shown in FIG. 3a, the length of the intersection line (320) where the object information for the wall and the object information for the door (300) meet is calculated, and if it is shorter than a preset threshold, the object information for the door (300) is removed, and if it is longer than the preset threshold, it is determined to be a door and the intersection line (320) can be redefined as the object information for the door (300).

[0036] For example, if the intersection line (321) where the object information for the wall and the object information for the door (301) expanded through the expansion operation meet is not a straight line, as shown in FIG. 3b, the object information for the door (301) can be separated into three lines connected to the intersection point corresponding to the object information for the door (301), and the longest line can be determined as the door and redefined as the object information for the door (301).

[0037] Meanwhile, the object information extraction unit (130) can generate object information for a window on object information for a wall, and redefine the object information for a window according to the type of intersection line between object information for a wall and object information for a window.

[0038] Here, when redefining object information for windows, the object information for windows can be expanded through an expansion operation to make it easier to extract the object information for windows.

[0039] For example, if the intersection line where object information about a wall and object information about a window meet is a straight line, the length of the intersection line can be calculated; if it is shorter than a preset threshold, the object information about the window is removed, and if it is greater than or equal to the preset threshold, it is determined to be a window and the intersection line can be redefined as object information about the window.

[0040] For example, if the line connecting the object information for the wall and the object information for the window is not a straight line, the object information for the window can be separated into three lines connected to the intersection point corresponding to the object information for the window, and the longest line can be determined as the window and redefined as the object information for the window.

[0041] Meanwhile, object information for doors and windows must be located on top of object information for walls to prevent the separation of skeletal information for walls and doors during the model creation process.

[0042] To prevent this phenomenon from occurring, if object information for doors and windows is not located on object information for walls, the object information for doors and windows can be projected onto object information for walls to correct it so that it is always located on object information for walls.

[0043] Meanwhile, object information regarding doors and windows can be distinguished by different colors depending on each type. For example, a single door can be indicated by the 5th color, a double door by the 6th color, and a window by the 7th color, and so on.

[0044] FIGS. 4a to 4d are drawings for explaining a method of extracting skeletal information (410) for a target space from object information for a target space according to an embodiment of the present invention and defining it as a node (421) and an edge (423).

[0045] FIG. 4a is a diagram in which skeletal information (410) for a target space is extracted from object information for a target space according to an embodiment of the present invention and defined as nodes (421) and edges (423).

[0046] Referring to FIG. 4a, the skeletal information extraction unit (140) can generate skeletal information for doors, windows, and walls included in the target space based on object information extracted from drawing data.

[0047] For example, the skeletal information extraction unit (140) can derive outline information constituting a wall from object information about a wall and use it to generate skeletal information about a wall, derive outline information constituting a door from object information about a door to generate skeletal information about a door, and derive outline information constituting a window from object information about a window to generate skeletal information about a window.

[0048] Referring to FIG. 4a, the structure definition unit (150) can define a plurality of nodes (421) corresponding to corner points or intersection points of the target space based on skeletal information (410) of the target space, and define an edge (423) connecting at least two adjacent nodes among the plurality of nodes (421).

[0049] FIG. 4b is a diagram illustrating a method of merging a plurality of nodes corresponding to an intersection included in skeletal information according to one embodiment of the present invention.

[0050] Referring to FIG. 4b, the structure definition unit (150) can merge at least two of the multiple nodes into one node based on the position between the two nodes.

[0051] For example, when generating skeletal information for a wall based on object information for a wall, at least two nodes (421a) corresponding to the intersections included in the skeletal information may occur due to the thickness of the skeletal information for the wall. In such a situation, if two or more nodes are located within a preset bounding box, the two or more corresponding nodes can be merged into one node to define a new node (421b).

[0052] Additionally, for example, as illustrated in FIG. 4c, among nodes with two connected nodes, the ratio of the distance between the node (421c) and the surrounding nodes to the sum of the distances between the node (421c) and the surrounding nodes is calculated, and if the ratio is less than 0.01, the node (421c) is determined to have the same directionality as the surrounding nodes and can be removed.

[0053] FIG. 4d is a diagram illustrating a method for correcting the position of a node according to an embodiment of the present invention.

[0054] Referring to FIG. 4d, the structure definition unit (150) may further include defining one of the plurality of nodes (421d) as a cross node (421e) and correcting the position of the cross node (421e) based on the positional relationship between the plurality of nodes (421d) connected to the cross node (421e).

[0055] For example, referring to FIG. 4d, as the angle between nodes approaches a right angle or horizontal, the data value generated during the creation of the spatial model becomes smaller, so the angle between multiple nodes connected to the intersection node (421e) can be corrected to be a right angle or horizontal.

[0056] FIG. 4e is a diagram in which outline information of a target space, which is a curve type according to one embodiment of the present invention, is defined as nodes and edges.

[0057] Referring to FIG. 4e, the structure definition unit (150) may further include defining one of the multiple nodes as a start node (421f), defining another of the multiple nodes connected to the start node (421f) as an end node (421g), classifying a curve type for an edge (423a) located between the start node (421f) and the end node (421g), and adding at least one node (421h) to the classified edge based on the curve type for the classified edge.

[0058] For example, a curve type can be classified as one of a first-order, second-order, or third-order function and defined as one of a polynomial type, a non-polynomial type, or a closed curve type through curve fitting.

[0059] Edges can be defined as polynomial types if the minimum value of the fitting error of the classified function is less than 0.01, calculated through the Mean Squared Error.

[0060] For example, in the case of a polynomial type, nodes and edges are represented as they are for a linear function, and for a 2nd or 3rd cubic function, the gradient can be calculated through derivative values ​​by converting to a coordinate system. Points where the gradient is 0, the start node, and the end node are designated as required nodes, and the number of nodes between the required nodes is defined and nodes can be placed at uniform intervals. The number of nodes between the required nodes can be derived through [Equation 1].

[0062] [Mathematical Formula 1]

[0063] Number of nodes = Length between nodes / Minimum length of skeletal information for the door

[0064] Here, the length between nodes refers to the length between essential nodes, and the minimum length of the skeletal information for the door refers to the minimum length between nodes of the skeletal information for the door.

[0066] Edges can be defined as non-polynomial types if the minimum value of the fitting error of the classified function is greater than 0.01, by calculating the fitting error value through the mean squared error.

[0067] For example, since the computational cost of finding the function for non-polynomial types is complex, the Ramer-Douglas-Peucker (RDP) algorithm can be used to detect nodes, define the number of nodes between nodes, and arrange nodes at uniform intervals, and the number of nodes between nodes can be derived through [Equation 2].

[0069] [Mathematical Formula 2]

[0070] Number of nodes = Length between nodes / (Minimum length of skeletal information for the door / 2)

[0072] If the start node and end node of an edge are the same point, it can be defined as a closed curve type.

[0073] For example, the closed curve type mainly occurs in columns and may require more precise node spacing compared to the polynomial and non-polynomial types. The number of nodes between nodes can be derived through [Equation 3].

[0075] [Mathematical Formula 3]

[0076] Number of nodes = Length between nodes / (Minimum length of skeletal information for the door * 2)

[0078] Meanwhile, the spatial model generation unit (160) can generate a spatial model for the target space based on information about the nodes and information about the edges.

[0079] FIGS. 5a and 5b are drawings for explaining a method of generating vector data for a plurality of objects included in a target space based on information about nodes and edges.

[0080] Referring to FIGS. 5a and 5b, the spatial model generation unit (160) can generate vector data for a plurality of objects included in the target space based on information about nodes and information about edges, and generate a spatial model for the target space based on the vector data for the plurality of objects.

[0081] Here, vector data refers to data suitable for implementing an analyzed drawing into a 3D model, and can be converted into one or more vector data based on information about nodes and edges for doors, windows, and walls.

[0082] Figure 5a is a drawing showing vector data for a wall generated based on information about nodes and information about edges.

[0083] Vector data (510) for the wall can be represented as a sequence of nodes.

[0084] For example, referring to Fig. 5a, the wall between nodes 9 and 10 can be represented by a sequence of nodes 9 and 10, and various wall thicknesses can be represented when implemented as a 3D model by assigning values ​​according to the thickness of the wall.

[0085] For example, data for the wall between node 9 and node 10 can be represented as 9-10, stroke-width=3.

[0086] In addition, if the data constitutes an exterior wall, a tag specifying that it is an exterior wall can be stored, and the exterior wall can be utilized as metadata when modeling front doors and windows.

[0087] For example, data about the wall between node 1 and node 2 can be represented as 1-2, stroke-width=5, outline.

[0088] Figure 5b is a drawing that generates vector data for doors and rooms based on information about nodes and information about edges.

[0089] Vector data (530) for a room can be represented as a sequence of nodes, and vector data (520) for a door can represent the position of the door on the edge of the wall.

[0090] Referring to Fig. 5b, vector data for a bathroom can be represented by a sequence of nodes 11, 12, 15, 14, and 11, and a room can be identified by a sequence of multiple nodes, allowing for editing of textures such as wallpaper, floor, and ceiling depending on the purpose of use of the room, and can be utilized as meta-information when arranging furniture in the room.

[0091] For example, vector data for a restroom can be represented as 11-12-15-14-11.

[0092] Referring to FIG. 5b, vector data (520) for the door can be represented on the edge connecting nodes 15 and 16, and the type of the door and the center coordinates of the door can be represented together.

[0093] For example, vector data (520) for a door on the edge connecting nodes 15 and 16 can be represented as 15-16, single door, [50, 200].

[0094] FIGS. 6a to 6c are drawings for explaining how to edit a room structure by editing nodes and edges for a target space.

[0095] Referring to FIGS. 6a to 6c, the shape and size of the room can be edited by changing the position of the node for the wall, and rooms can be added and removed by editing the node and edge for the wall.

[0096] FIG. 7 is a diagram illustrating a method for generating a spatial model by applying height information to skeletal information for a target space according to an embodiment of the present invention.

[0097] Referring to Fig. 7, height information can be applied to vector data for the target space to create a spatial model for the target space that can be represented at various heights.

[0098] For example, if the height of the living room is 3m and the height of the bathroom is 2m, and the default height of the space model for the target space is 2m, the vector data for the default height can be represented as default.

[0099] For example, vector data for the living room can be represented as 6-7-8-9-18-17-28-27-22-16-15-12-13-6, height=3, and vector data for the bathroom can be represented as 11-12-15-14-11 default.

[0100] In this way, by applying height information to vector data of the target space, it is possible to generate a spatial model with different height information applied to each room in cases where the height varies from room to room, such as in detached houses, pensions, or duplex structures.

[0101] FIG. 8 is a diagram illustrating a method of creating nodes (800) for each layer of stairs according to an embodiment of the present invention and connecting the nodes with edges.

[0102] Referring to FIG. 8, when the structure definition unit (150) creates a model of a multi-story structure or a multi-story building, it can represent vector data for each floor in three dimensions.

[0103] For example, referring to Fig. 8, vector data for the first-floor restroom can be expressed as 1, 11-12-15-14-11, and vector data for the second-floor restroom can be expressed as 2, 31-32-38-37-34-35-31.

[0104] In addition, nodes (800) for each layer of stairs can be created and connected with edges to represent vector data.

[0105] Through this, inter-floor movement can be achieved when a spatial model for the target space is created.

[0106] FIG. 9 is a drawing of a spatial model (900) generated based on drawing data according to one embodiment of the present invention.

[0107] As illustrated in FIG. 9, the spatial model generation unit (160) can generate a spatial model (900) for a target space based on information about nodes and information about edges.

[0108] In this way, by creating a spatial model (900) for the target space based on information about nodes and information about edges, the accuracy can be increased when reinterpreting into edges and nodes, and it can be divided into rooms and made easy to edit.

[0109] Meanwhile, a person skilled in the art will fully understand that the input unit (110), the prediction unit (120), the object information extraction unit (130), the skeletal information extraction unit (140), the structure definition unit (150), and the spatial model generation unit (160) may each be implemented separately, or that one or more of these may be implemented in an integrated manner.

[0110] FIG. 10 is a flowchart illustrating a method for generating a spatial model according to one embodiment of the present invention.

[0111] Referring to Fig. 10, drawing data for the target space can be received in step S1010.

[0112] In step S1020, skeletal information for the target space can be extracted from the drawing data.

[0113] In step S1030, multiple nodes corresponding to corner points or intersection points in the target space can be defined based on the skeletal information.

[0114] In step S1040, an edge connecting at least two adjacent nodes among multiple nodes can be defined.

[0115] In step S1050, a spatial model for the target space can be generated based on information about nodes and information about edges.

[0116] In the description above, steps S1010 to S1050 may be further divided into additional steps or combined into fewer steps, depending on an embodiment of the present invention. Additionally, some steps may be omitted as necessary, and the order between steps may be changed.

[0117] One embodiment of the present invention may also be implemented in the form of a recording medium comprising computer-executable instructions, such as program modules executed by a computer. A computer-readable medium may be any available medium accessible by a computer and includes both volatile and non-volatile media, and both removable and non-removable media. Additionally, a computer-readable medium may include all computer storage media. A computer storage medium includes both volatile and non-volatile, removable and non-removable media implemented by any method or technique for storing information, such as computer-readable instructions, data structures, program modules, or other data.

[0118] The foregoing description of the present invention is for illustrative purposes only, and those skilled in the art will understand that other specific forms can be easily modified without altering the technical spirit or essential features of the present invention. Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive. For example, each component described as a single unit may be implemented in a distributed manner, and components described as distributed may likewise be implemented in a combined form.

[0119] The scope of the present invention is defined by the claims set forth below rather than by the detailed description, and all modifications or variations derived from the meaning and scope of the claims and equivalent concepts thereof should be interpreted as being included within the scope of the present invention. Explanation of the symbols

[0120] 100: Spatial model generation device 110: Input section 120: Prediction section 130: Object Information Extraction Unit 140: Skeletal Information Extraction Unit 150: Structural Definition Section 160: Spatial Model Generation Section

Claims

Claim 1 A method for generating a spatial model comprising: receiving drawing data for a target space; extracting skeletal information for the target space from the drawing data; defining a plurality of nodes corresponding to corner points or intersection points for the target space based on the skeletal information; defining an edge connecting at least two adjacent nodes among the plurality of nodes; and generating a spatial model for the target space based on information about the nodes and information about the edges, and further comprising: extracting object information about an object included in the target space from the drawing data based on a prediction model; and generating the skeletal information based on the extracted object information. Claim 2 A method for generating a spatial model according to claim 1, further comprising the step of predicting object information from the drawing data based on the prediction model. Claim 3 A method for generating a spatial model according to claim 2, wherein the step of extracting object information further includes the step of extracting object information for at least one of a door, window, and wall included in the target space from the drawing data using the prediction model. Claim 4 A method for generating a spatial model according to claim 3, further comprising: a step of generating at least one of object information for a door or object information for a window on object information for a wall; and a step of redefining object information for a wall or object information for a window according to a type of intersection line between object information for a wall and object information for a door or a type of intersection line between object information for a wall and object information for a window. Claim 5 A method for generating a spatial model according to claim 1, further comprising the step of merging at least two of the plurality of nodes into one node based on the position between the two nodes. Claim 6 A method for generating a spatial model comprising: receiving drawing data for a target space; extracting skeletal information for the target space from the drawing data; defining a plurality of nodes corresponding to corner points or intersection points for the target space based on the skeletal information; defining an edge connecting at least two adjacent nodes among the plurality of nodes; and generating a spatial model for the target space based on information about the nodes and information about the edges, further comprising: defining one of the plurality of nodes as an intersection node; and correcting the position of the intersection node based on the positional relationship between the plurality of nodes connected to the intersection node. Claim 7 A method for generating a spatial model comprising: receiving drawing data for a target space; extracting skeletal information for the target space from the drawing data; defining a plurality of nodes corresponding to corner points or intersection points of the target space based on the skeletal information; defining an edge connecting at least two adjacent nodes among the plurality of nodes; and generating a spatial model for the target space based on information about the nodes and information about the edges, wherein the method further comprises: defining one of the plurality of nodes as a start node; defining another of the plurality of nodes connected to the start node as an end node; classifying a curve type for an edge located between the start node and the end node; and adding at least one node to the classified edge based on the curve type for the classified edge. Claim 8 A method for generating a spatial model according to claim 1, wherein when generating a model of a multi-story structure or a multi-story building, the method further comprises the step of generating nodes for the stairs of each floor; and the step of connecting the generated nodes with edges. Claim 9 A spatial model generating device comprising: an input unit receiving drawing data for a target space; a skeletal information extraction unit extracting skeletal information for the target space from the drawing data; a structure definition unit defining a plurality of nodes corresponding to corner points or intersection points for the target space based on the skeletal information and defining an edge connecting at least two adjacent nodes among the plurality of nodes; and a spatial model generating unit generating a spatial model for the target space based on information about the nodes and information about the edges, and further comprising an object information extraction unit extracting object information about an object included in the target space from the drawing data based on a prediction model, wherein the skeletal information extraction unit generates the skeletal information based on the extracted object information. Claim 10 A spatial model generating device according to claim 9, further comprising a prediction unit that predicts object information from the drawing data based on the prediction model. Claim 11 A spatial model generating device according to claim 10, wherein the object information extraction unit extracts object information for at least one of a door, window, and wall included in the target space from the drawing data using the prediction model. Claim 12 A spatial model generating device according to claim 11, wherein the object information extraction unit generates at least one of object information for the door or object information for the window on the object information for the wall, and the structure definition unit redefines the object information for the door or the object information for the window according to the type of intersection line between the object information for the wall and the object information for the door or the type of intersection line between the object information for the wall and the object information for the window. Claim 13 A spatial model generating device according to claim 9, wherein the structure definition unit comprises merging at least two of the plurality of nodes into one node based on the position between the two nodes. Claim 14 A spatial model generating device comprising: an input unit receiving drawing data for a target space; a skeletal information extraction unit extracting skeletal information for the target space from the drawing data; a structure definition unit defining a plurality of nodes corresponding to corner points or intersection points for the target space based on the skeletal information and defining an edge connecting at least two adjacent nodes among the plurality of nodes; and a spatial model generating unit generating a spatial model for the target space based on information about the nodes and information about the edges, wherein the structure definition unit further comprises defining one of the plurality of nodes as an intersection node and correcting the position of the intersection node based on the positional relationship between the plurality of nodes connected to the intersection node. Claim 15 A spatial model generating device comprising: an input unit receiving drawing data for a target space; a skeletal information extraction unit extracting skeletal information for the target space from the drawing data; a structure definition unit defining a plurality of nodes corresponding to corner points or intersection points for the target space based on the skeletal information, and defining an edge connecting at least two adjacent nodes among the plurality of nodes; and a spatial model generating unit generating a spatial model for the target space based on information about the nodes and information about the edges, wherein the structure definition unit further comprises defining one of the plurality of nodes as a start node, defining another node among the plurality of nodes connected to the start node as an end node, classifying a curve type for an edge located between the start node and the end node, and adding at least one node to the classified edge based on the curve type for the classified edge. Claim 16 In claim 9, the structure definition unit is a spatial model generation device that, when generating a model of a multi-story structure or a multi-story building, generates nodes for the stairs of each floor and connects the generated nodes with edges. Claim 17 A computer program stored on a computer-readable recording medium comprising a sequence of instructions for generating a spatial model, wherein, when the computer program is executed by a computing device, the computer program receives drawing data for a target space, extracts skeletal information for the target space from the drawing data, defines a plurality of nodes corresponding to corner points or intersection points for the target space based on the skeletal information, defines an edge connecting at least two adjacent nodes among the plurality of nodes, generates a spatial model for the target space based on information about the nodes and information about the edge, extracts object information about an object included in the target space from the drawing data based on a prediction model, and generates the skeletal information based on the extracted object information.

Citation Information

Patent Citations

  • Drawing display processor, drawing prefetch processing method and program, and recording medium of the program

    JP2003263348A

  • Map generation device and display method

    JP2014153824A

  • Method and program for modeling 3-dimension structure by 2-dimension floor plan

    KR1020160064784A

  • Method for building an indoor map and apparatus thereof

    KR1020180109410A