Method and device for recommending a projection pose for converting a three-dimensional structure into a two-dimensional three-view drawing

Abstract

The application provides a three-dimensional structure map conversion two-dimensional three-view projection posture recommendation method and device, relates to the technical field of mechanical design, and the method comprises the following steps: firstly, three-dimensional part numerical model information to be recommended for projection posture is extracted; then, the three-dimensional part numerical model information is input into a trained deep learning model; and finally, based on the output three-dimensional part projection direction and posture corresponding transformation matrix, the posture of the three-dimensional part to be projected is automatically adjusted. By using the deep learning model, the transformation matrix corresponding to the recommended three-dimensional part projection direction and posture can be realized, so as to project a two-dimensional three-view, that is, the automatic judgment of the projection direction and the automatic posture transformation of the part numerical model are realized, the problem of low drawing efficiency is alleviated, the process of the designer judging the projection posture is reduced, and the design efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of mechanical design technology, and in particular to a method and apparatus for recommending projection postures for converting three-dimensional structural diagrams into two-dimensional three-view drawings in computer-aided design technology. Background Technology

[0002] In the field of mechanical design, when designing a complete solution, a three-dimensional model of the assembly, consisting of many parts, is usually created first using design software. After the three-dimensional model is completed, two-dimensional drawings need to be output for subsequent parts processing, manufacturing, and assembly.

[0003] In the process of generating 2D drawings, the first step is to break down the drawings according to the number of parts, with each part generally requiring a separate 2D drawing. Since the structure and shape of each part differ, in order to clearly represent the structure, shape, and dimensions of the parts through the 2D drawings, it is necessary to select the projection direction for each part individually. This mainly involves selecting the direction of the main view of the part, and then deciding whether to include side views and top views based on the characteristics and complexity of the part. Next, the projection orientation of each part needs to be adjusted so that the projected 2D view is aligned (parallel to the X-axis or Y-axis).

[0004] In the process of casting parts using existing design software, designers usually select the casting direction and view type, and then manually align the casting position of the parts. This relies too heavily on the designer's work experience and involves judging and adjusting the casting direction and position of a large number of parts one by one, resulting in low drawing efficiency. Summary of the Invention

[0005] The purpose of this invention is to provide a method and apparatus for recommending projection postures when converting three-dimensional structural diagrams into two-dimensional three-view diagrams, so as to alleviate the technical problems of excessive reliance on designers and low drawing efficiency in the prior art.

[0006] To achieve the above objectives, the technical solutions adopted in the embodiments of the present invention are as follows:

[0007] In a first aspect, embodiments of the present invention provide a method for recommending the projection posture of a three-dimensional structural diagram to a two-dimensional three-view drawing, comprising the following steps: extracting the three-dimensional part model information of the projection posture to be recommended; inputting the three-dimensional part model information into a trained deep learning model; and automatically correcting the posture of the three-dimensional part to be projected based on the output transformation matrix corresponding to the projection direction and posture of the three-dimensional structure.

[0008] In some possible implementations, the three-dimensional part model information for the recommended projection orientation includes: part structure information and part orientation information.

[0009] In some possible implementations, the training of the aforementioned deep learning model includes the following steps: establishing a historical parts database, wherein the input is 3D part model information; establishing a database of mapping relationships between the part model information and its corresponding 2D three-view projection direction and attitude transformation matrix; training the deep learning model based on the aforementioned historical parts database to learn the mapping relationship between the 3D part model information and the projection direction and attitude transformation matrix information; evaluating the correctness of the deep learning model and making corrections so as to recommend projection direction and projection attitude matrix using the corrected deep learning algorithm model.

[0010] In some possible implementations, the mapping relationship between the above-mentioned three-dimensional part digital model information and the projection direction and attitude transformation matrix information includes: the projection direction of the three-dimensional part and the transformation matrix corresponding to the attitude of the three-dimensional part in the projection state.

[0011] In some possible implementations, the projection direction of the three-dimensional part in the above mapping relationship is the normal vector corresponding to any two-dimensional view.

[0012] In some possible implementations, the pose transformation matrix of the three-dimensional part in the projection state includes the three-dimensional structure placement angle or rotation angle, offset or translation amount.

[0013] In some possible implementations, the deep learning model described above is used to simultaneously learn the mapping relationship between the projection direction of the three-dimensional part and the attitude transformation matrix of the three-dimensional part under the projection state.

[0014] In some possible implementations, the deep learning model synchronously outputs the projection direction of the three-dimensional part and the pose information of the three-dimensional structure in the projection state after learning.

[0015] Secondly, embodiments of the present invention provide a projection posture recommendation device for converting a three-dimensional structural diagram into a two-dimensional three-view drawing, comprising: an extraction module for extracting three-dimensional part model information of the projection posture to be recommended; an input module for inputting the three-dimensional part model information into a trained deep learning model; and a straightening module for automatically straightening the posture of the three-dimensional part to be projected based on the output transformation matrix corresponding to the projection direction and posture of the three-dimensional part.

[0016] Thirdly, embodiments of the present invention provide an electronic device, including a memory and a processor, wherein the memory stores a computer program that can run on the processor, and the processor executes the computer program to implement the steps of the method described in any of the first aspects above.

[0017] This invention provides a method and apparatus for recommending projection postures when converting three-dimensional structural diagrams into two-dimensional three-view drawings. The method includes: first, extracting the 3D part's digital model information for the recommended projection posture; then, inputting the 3D part's digital model information into a trained deep learning model; and finally, automatically correcting the posture of the 3D part to be projected based on the output transformation matrix corresponding to the projection direction and posture of the 3D part. By utilizing a deep learning model, the transformation matrix corresponding to the projection direction and posture of the 3D part can be recommended for projecting two-dimensional three-view drawings. This achieves automatic determination of the projection direction and automatic posture transformation of the part's digital model, alleviating the problem of low drawing efficiency, reducing the process of designers determining the projection posture, and improving design efficiency. Attached Figure Description

[0018] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0019] Figure 1 This is a flowchart illustrating a method for recommending projection postures in converting a three-dimensional structural diagram into two-dimensional three-view drawings, as provided in an embodiment of the present invention.

[0020] Figure 2 A specific deep learning algorithm network framework diagram is provided for an embodiment of the present invention;

[0021] Figure 3 A comparison of projection effects of a projection posture recommendation method for converting a three-dimensional structural diagram into a two-dimensional three-view drawing, provided in an embodiment of the present invention;

[0022] Figure 4 This is a schematic diagram of a projection posture device for converting a three-dimensional structural diagram into a two-dimensional three-view drawing, provided in an embodiment of the present invention.

[0023] Figure 5 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present invention. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.

[0025] Therefore, the following detailed description of the embodiments of the invention provided in the accompanying drawings is not intended to limit the scope of the claimed invention, but merely to illustrate selected embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the invention without inventive effort are within the scope of protection of the invention.

[0026] In the field of mechanical design, when designing a complete solution, a three-dimensional model of an assembly composed of many parts is usually created first using design software. After the three-dimensional model is completed, two-dimensional drawings need to be output (usually, 3D design software has its corresponding two-dimensional drawing output formats, such as DWG and DXF), for subsequent parts processing, manufacturing, and assembly. In generating two-dimensional drawings, the first step is to break down the drawings according to the number of parts; generally, each part requires a separate two-dimensional drawing. Since the structure and shape of each part differ, in order to clearly represent the structure, shape, and dimensions of the parts through two-dimensional drawings, the projection direction of each part needs to be selected individually, and the projection posture needs to be corrected. This mainly involves selecting the front view direction of the part, and then, based on the characteristics and complexity of the part, deciding whether side and top views are needed; finally, the drawing is performed with the correct posture.

[0027] In the process of casting parts using existing design software, designers usually choose the casting direction, casting posture, and view type. This relies excessively on the designer's work experience and involves judging the casting direction and casting posture of a large number of parts one by one, resulting in tedious work and low drawing efficiency.

[0028] Based on this, embodiments of the present invention provide a method and apparatus for recommending projection postures for converting three-dimensional structural diagrams into two-dimensional three-view diagrams, in order to alleviate the problem of low projection efficiency.

[0029] To facilitate understanding of this embodiment, a method for recommending projection postures for converting a three-dimensional structural diagram into two-dimensional three-view diagrams, as disclosed in this embodiment of the invention, will first be described in detail. (See [link to relevant documentation]). Figure 1 The diagram shows a method for recommending projection postures when converting a 3D structural diagram into a 2D three-view drawing. This method can be executed by an electronic device and mainly includes the following steps S110 to S120:

[0030] S110: Extract the 3D part model information of the projection posture to be recommended;

[0031] The 3D part model information for the recommended projection posture includes: part structure information and part posture information.

[0032] S120: Input the 3D part digital model information into the trained deep learning model;

[0033] S130: Based on the transformation matrix corresponding to the projection direction and posture of the output 3D part, automatically correct the posture of the 3D part to be projected.

[0034] This application provides a method for recommending the projection posture of a 3D structural diagram when converting it into three-view drawings. The method includes: first, extracting the 3D part's digital model information for the recommended projection posture; then, inputting the 3D part's digital model information into a trained deep learning model; and finally, automatically correcting the posture of the 3D part based on the output transformation matrix corresponding to the projection direction and posture of the 3D part. By utilizing a deep learning model, the method can recommend the transformation matrix corresponding to the projection direction and posture of the 3D part for projecting two-dimensional three-view drawings. This achieves automatic determination of the projection direction and automatic posture transformation of the part's digital model, alleviating the problem of low drawing efficiency, reducing the process of designers judging the projection posture, and improving design efficiency.

[0035] In one embodiment, training a deep learning model may include the following steps:

[0036] S201: Establish a historical parts database, where the input is 3D part digital model information;

[0037] S202: Establish a database that maps the part model information to its corresponding two-dimensional three-view projection direction and attitude transformation matrix;

[0038] The mapping relationship between the 3D part's digital model information and the projection direction and attitude transformation matrix information can include: the projection direction of the 3D part and the transformation matrix corresponding to the attitude of the 3D part under the projection state.

[0039] In this mapping relationship, the projection direction of the 3D part is the normal vector corresponding to any 2D view.

[0040] As a specific example, the transformation matrix may include: the orientation or rotation angle of the three-dimensional structure, the offset or translation amount.

[0041] S203: Train a deep learning model based on a historical parts database to learn the mapping relationship between the 3D parts digital model information and the projection direction and attitude transformation matrix information;

[0042] In one embodiment, the deep learning model is used to synchronously learn the mapping relationship between the projection direction of the 3D part and the pose transformation matrix of the 3D part under the projection state.

[0043] Furthermore, after learning, the deep learning model synchronously outputs the projection direction of the 3D part and the pose information of the 3D structure under the projection state.

[0044] S204: Evaluate the correctness of the deep learning model and make corrections so that the corrected deep learning model can be used to recommend projection directions and projection pose matrices.

[0045] In one embodiment, the projection directions of the two-dimensional three views of the three-dimensional part include: front view, side view, and top view; the attitude transformation matrix includes: a 3*3 rotation matrix or a 4*4 transformation matrix (rotation + translation); and the label samples used as input to the deep learning model include at least one of the attitude transformation matrices of the three-dimensional part digital model.

[0046] Typically, deep learning models include, but are not limited to, deep learning models, and can also be machine learning models. As a specific example, the deep learning model can be a convolutional neural network model, which includes: an input layer, convolutional layers, pooling layers, fully connected layers, and a vector output layer.

[0047] Figure 2 The diagram shows a specific deep learning algorithm network framework. After the 3D part's digital model information is input into the deep learning algorithm network, it undergoes multi-layer processing and finally outputs the normal vectors and pose transformation matrices of the three views. Each layer in the diagram consists of convolutional and pooling layers. The output of each layer represents the data dimension after the data from the previous layer has been processed by convolution, pooling, and activation functions. N×9, N×64, N×128, N×256, N×512, N×1024, N×512, N×256, N×128, and N×64 all represent the processed data dimensions. Layers connected by the "+" symbol in the diagram represent the addition and fusion of the preceding and following layers. N×64 to 1×64 layers undergo global pooling, then are fully connected to a 1×16 layer, and then fully connected again. The 1×16 layer is reshaped to 4×4, representing the pose transformation matrix, and the 1×9 layer is reshaped to 3×3, representing the normal vectors of the three views.

[0048] The deep learning model described above can automatically determine and adjust the projection posture, alleviating the problem of low projection efficiency, reducing the process for designers to determine the projection direction, and improving design efficiency.

[0049] As a specific example, this application provides a method for recommending projection poses when converting a three-dimensional structural diagram into three views, including:

[0050] 1. Establish historical dataset: The input samples are historical 3D part models, and the label samples are the pose transformation matrix of the historical 3D part models and the normal vectors corresponding to each projection direction (front view, side view, top view) in the 2D drawings.

[0051] 2. Construction of learning algorithm models: including but not limited to network frameworks and loss functions for machine learning, deep learning, and other algorithms.

[0052] 3. Training the learning algorithm model: Using the 3D part model from step 1 as input samples, and the projection direction (normal vector) and attitude transformation matrix as label samples, train the learning algorithm model from step 2 (e.g., supervised learning algorithm model). After training is complete, retain the trained network parameters.

[0053] 4. Application of learning algorithm model: When a new 3D part model is input, the corresponding attitude transformation matrix (4*4 matrix) and projection direction (normal vector) of the new 3D part are obtained after calculation by the network parameters in step 3.

[0054] 5. Automatic Projection: Based on the attitude transformation matrix and projection direction obtained in step 4, the 3D part model is projected to obtain the 2D projection of each view. A specific projection effect can be seen in [the image / description]. Figure 3 One is the projection structure without outputting the attitude transformation matrix (A - projection with uncorrected attitude), and the other is the projection result after adding the attitude transformation matrix (B - projection with corrected attitude).

[0055] This application embodiment constructs a dataset of historical 3D part digital models, their attitude transformation matrices, and projection directions, and designs an algorithm to learn the knowledge between the 3D digital models, attitude transformation matrices, and projection direction judgments, enabling an automatic projection process. This reduces the time designers spend judging projection attitudes and improves design efficiency. With a large number of historical design schemes, it allows for rapid migration of application scenarios and is applicable to the entire part design process across the industrial sector.

[0056] This invention provides a device for recommending projection postures when converting a three-dimensional structural diagram into three-view drawings. See [link to device]. Figure 4 The device includes:

[0057] Extraction module 410 is used to extract the 3D part model information of the projection posture to be recommended;

[0058] The input module 420 is used to input the 3D part digital model information into the trained deep learning model;

[0059] The alignment module 430 is used to automatically align the orientation of the 3D part to be projected based on the output transformation matrix corresponding to the projection direction and orientation of the 3D part.

[0060] The 3D part model information for the recommended projection posture includes: part structure information and part posture information.

[0061] In one embodiment, the apparatus for recommending the projection posture of the above-mentioned three-dimensional structural diagram to three-view drawing conversion may further include:

[0062] The training module is used to establish a historical parts database, with the input being 3D part model information; to establish a database of mapping relationships between the 3D part model information and its corresponding 2D three-view projection direction and attitude transformation matrix; to train a deep learning model based on the historical parts database, learning the mapping relationship between the 3D part model information and the projection direction and attitude transformation matrix information; and to evaluate and correct the correctness of the deep learning algorithm model, so as to recommend projection direction and projection attitude matrix using the corrected deep learning algorithm model.

[0063] This application provides a method and apparatus for recommending projection postures when converting a 3D structural diagram into a 2D three-view drawing. The method includes: first, extracting the 3D part's digital model information for the recommended projection posture; then, inputting the 3D part's digital model information into a trained deep learning model; and finally, automatically correcting the 3D part's posture based on the output transformation matrix corresponding to the projection direction and posture. By utilizing a deep learning model, the transformation matrix corresponding to the projection direction and posture of the 3D part can be recommended for projecting 2D three-view drawings. This achieves automatic determination of the projection direction and automatic posture transformation of the part's digital model, alleviating the problem of low drawing efficiency, reducing the process of designers determining the projection posture, and improving design efficiency.

[0064] The device for recommending projection postures for converting 3D structural diagrams to three-view drawings provided in this application embodiment can be specific hardware on a device or software or firmware installed on the device. The implementation principle and technical effects of the device provided in this application embodiment are the same as those in the foregoing method embodiments. For the sake of brevity, any parts not mentioned in the device embodiment can be referred to the corresponding content in the foregoing method embodiments. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can all be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here. The projection posture recommendation device for converting 3D structural diagrams to three-view drawings provided in this application embodiment has the same technical features as the projection posture recommendation method for converting 3D structural diagrams to three-view drawings provided in the foregoing embodiments, and therefore can solve the same technical problems and achieve the same technical effects.

[0065] This application also provides an electronic device, specifically, the electronic device includes a processor and a storage device; the storage device stores a computer program, and the computer program, when run by the processor, executes the method described in any of the above embodiments.

[0066] Figure 5This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. The electronic device 500 includes: a processor 50, a memory 51, a bus 52, and a communication interface 53. The processor 50, the communication interface 53, and the memory 51 are connected through the bus 52. The processor 50 is used to execute executable modules, such as computer programs, stored in the memory 51.

[0067] The memory 51 may include high-speed random access memory (RAM) or non-volatile memory, such as at least one disk storage device. Communication between this system network element and at least one other network element is achieved through at least one communication interface 53 (which can be wired or wireless), such as the Internet, wide area network, local area network, metropolitan area network, etc.

[0068] Bus 52 can be an ISA bus, PCI bus, or EISA bus, etc. The bus can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 5 The symbol is represented by a single double-headed arrow, but this does not mean that there is only one bus or one type of bus.

[0069] The memory 51 is used to store programs. After receiving an execution instruction, the processor 50 executes the programs. The method executed by the device for defining the flow process disclosed in any of the foregoing embodiments of the present invention can be applied to the processor 50 or implemented by the processor 50.

[0070] Processor 50 may be an integrated circuit chip with signal processing capabilities. In implementation, each step of the above method can be completed by the integrated logic circuitry in the hardware of processor 50 or by instructions in software form. Processor 50 can be a general-purpose processor, including a Central Processing Unit (CPU), a Network Processor (NP), etc.; it can also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field-Programmable Gate Array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components. It can implement or execute the methods, steps, and logic block diagrams disclosed in the embodiments of this invention. The general-purpose processor can be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this invention can be directly embodied in the execution of a hardware decoding processor, or executed by a combination of hardware and software modules in the decoding processor. The software modules can reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. The storage medium is located in memory 51. The processor 50 reads the information in memory 51 and, in conjunction with its hardware, completes the steps of the above method.

[0071] Corresponding to the above method, this application embodiment also provides a computer-readable storage medium storing machine-executable instructions. When the machine-executable instructions are called and run by a processor, the machine-executable instructions cause the processor to perform the steps of the above method.

[0072] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. The apparatus embodiments described above are merely illustrative. For example, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. Furthermore, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Additionally, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some communication interfaces; indirect couplings or communication connections between devices or units may be electrical, mechanical, or other forms.

[0073] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0074] In addition, the functional units in the embodiments provided in this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0075] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, electronic device, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0076] It should be noted that similar reference numerals and letters in the accompanying drawings indicate similar items. Therefore, once an item is defined in one accompanying drawing, it does not need to be further defined and explained in subsequent accompanying drawings. In addition, the terms "first," "second," "third," etc. are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0077] Finally, it should be noted that the above-described embodiments are merely specific implementations of the present invention, used to illustrate the technical solutions of the present invention, and not to limit them. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that any person skilled in the art can still modify or easily conceive of changes to the technical solutions described in the foregoing embodiments within the scope of the technology disclosed in the present invention, or make equivalent substitutions for some of the technical features; and these modifications, changes, or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be covered within the scope of protection of the present invention.

Claims

1. A method for recommending projection postures when converting a three-dimensional structural diagram into two-dimensional three-view diagrams, characterized in that, include: Extract the 3D part model information of the recommended projection posture; The 3D part digital model information is input into the trained deep learning model; wherein, the deep learning model synchronously learns the mapping relationship between the 3D part digital model information and the projection direction of the 3D part, and the posture transformation matrix of the 3D part in the projection state, and synchronously outputs the projection direction and posture information of the 3D part in the projection state; the transformation matrix corresponding to the posture of the 3D part in the projection state includes the 3D structure placement angle or rotation angle, offset or translation; the posture transformation matrix includes a 3*3 rotation matrix or a 4*4 transformation matrix; Based on the transformation matrix corresponding to the projection direction and posture of the output 3D part, the posture of the 3D part to be projected is automatically corrected.

2. The method for recommending projection postures for converting three-dimensional structural diagrams into two-dimensional three-view diagrams according to claim 1, characterized in that, The 3D part model information for the recommended projection posture includes: part structure information and part posture information.

3. The method for recommending projection postures for converting three-dimensional structural diagrams into two-dimensional three-view diagrams according to claim 1, characterized in that, The training of the deep learning model includes the following steps: Establish a historical parts database, with the input being 3D part digital model information; Establish a database that maps the part model information to its corresponding two-dimensional three-view projection direction and attitude transformation matrix; The deep learning model is trained based on the historical parts database to learn the mapping relationship between the three-dimensional parts digital model information and the projection direction and attitude transformation matrix information; The correctness of the trained deep learning model is evaluated and corrected so that the corrected deep learning model can be used to recommend projection direction and projection pose matrix.

4. The method for recommending projection postures for converting three-dimensional structural diagrams into two-dimensional three-view diagrams according to claim 3, characterized in that, In the mapping relationship, the projection direction of the three-dimensional part is the normal vector corresponding to any two-dimensional view.

5. A projection posture recommendation device for converting a three-dimensional structural diagram into two-dimensional three-view drawings, characterized in that, include: The extraction module is used to extract the 3D part model information of the projection posture to be recommended; An input module is used to input the 3D part digital model information into a trained deep learning model. The deep learning model synchronously learns the mapping relationship between the 3D part digital model information and the projection direction and pose transformation matrix of the 3D part under projection conditions, and synchronously outputs the projection direction and pose information of the 3D part under projection conditions. The transformation matrix corresponding to the pose of the 3D part under projection conditions includes the 3D structure placement angle or rotation angle, offset or translation. The pose transformation matrix includes a 3*3 rotation matrix or a 4*4 transformation matrix. The alignment module is used to automatically align the orientation of the 3D part to be projected based on the transformation matrix corresponding to the output 3D part projection direction and orientation.

6. An electronic device comprising a memory and a processor, wherein the memory stores a computer program executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method described in any one of claims 1 to 4.

Images