Three-dimensional model generation method, equipment and medium based on industrial parts genes

Abstract

The present application provides a method, device, and medium for generating a three-dimensional model based on industrial part genes, relating to the field of intelligent industrial manufacturing technology. The method comprises: obtaining sampling data of the industrial part to be generated, wherein the sampling data includes surface sampling data and edge sampling data; obtaining industrial part gene data of the industrial part to be generated, wherein the industrial part gene data corresponds to different phenotypes of the industrial part; obtaining a trained three-dimensional generation model, wherein the three-dimensional generation model is implemented based on a diffusion transformer, the three-dimensional generation model includes a surface denoising network and an edge denoising network, and the trained three-dimensional generation model is obtained by training surface data training samples, edge data training samples, and industrial part gene data training samples; and inputting the surface sampling data, edge sampling data, and industrial part gene data into the trained denoising model to generate a three-dimensional model corresponding to the industrial part to be generated through the trained denoising model.

Description

Technical Field

[0001] The present application relates to the field of intelligent industrial manufacturing technology, and in particular to a method, device and medium for generating a three-dimensional model based on industrial part genes. Background Art

[0002] A Computer-Aided Design (CAD) model is a three-dimensional model created using CAD software. It can contain information such as an object's geometry, dimensions, and materials, and can be visualized and edited. CAD models are widely used in the design and manufacturing processes of fields such as architecture, mechanics, electronics, and aviation, and are particularly widespread in the industrial sector. CAD models typically utilize a "sketch and extrude" representation method, which generates a limited range of three-dimensional (3D) shapes for industrial parts. This method is limited to generating sketches composed of three primitives: lines, arcs, and circles. It is also limited to performing extrusion operations and cannot achieve complex 3D shapes composed of complex curves and surfaces.

[0003] In related technologies, boundary representation (B-rep) is another CAD model description method that accurately expresses complex geometric shapes through parametric curves, surfaces, and face-edge-vertex topological relationships. It is particularly suitable for free-form surface modeling.

[0004] However, the existing CAD model generation method is limited to specific shape primitives, cannot meet the requirements of complex CAD model generation, and has low efficiency in CAD model generation. Summary of the Invention

[0005] The present application provides a three-dimensional model generation method, equipment and medium based on industrial part genes, which is used to solve the technical problems that the existing CAD model generation method is limited to specific shape primitives, cannot meet the complex CAD model generation requirements, and has low efficiency in CAD model generation.

[0006] In a first aspect, the present application provides a method for generating a three-dimensional model based on an industrial part gene, comprising:

[0007] Acquire sampling data of the industrial part to be generated, wherein the sampling data includes surface sampling data and edge sampling data;

[0008] Acquiring industrial part gene data of the industrial part to be generated, wherein the industrial part gene data corresponds to different phenotypes of the industrial part;

[0009] Obtaining a trained 3D generative model, wherein the 3D generative model is implemented based on a diffusion transformer, the 3D generative model includes a surface denoising network and an edge denoising network, and the trained 3D generative model is obtained by training surface data training samples, edge data training samples, and industrial part gene data training samples;

[0010] The surface sampling data, the edge sampling data and the industrial part gene data are input into a trained denoising model, so as to generate a three-dimensional model corresponding to the industrial part to be generated through the trained denoising model.

[0011] An embodiment of the present application provides a CAD model generation method based on industrial part DNA. For an industrial part to be generated, surface sampling data and edge sampling data are sampled, and corresponding industrial part DNA data is obtained. The surface sampling data, edge sampling data, and surface sampling data and edge sampling data are then input into a trained 3D generative model. The trained 3D generative model can then generate a CAD model corresponding to the industrial part to be generated. The 3D generative model here implements surface and edge denoising based on a Diffusion Transformer (DiT) architecture. The DiT architecture addresses the limitations of traditional diffusion models, efficiently models long-range dependencies, and captures richer global features. While improving the modeling capabilities of the denoising network, it also addresses the problem of difficult training convergence, providing support for accurate denoising of the 3D generative model. The structure of the 3D generative model treats vertex connections as edges, requiring only two denoising networks, surface and edge, to be designed. This avoids the design of vertex denoising networks, reduces computational effort, and improves generation efficiency. Based on this model architecture and combined with industrial part DNA data, it can guide the generation of industrial parts, meet complex CAD model generation requirements, and improve CAD model generation efficiency.

[0012] Optionally, before obtaining the trained three-dimensional generative model, the method further includes:

[0013] Obtain the original 3D generative model;

[0014] Obtaining surface data training samples, edge data training samples, and industrial parts gene data training samples;

[0015] The original three-dimensional generation model is trained according to the surface data training samples, the edge data training samples and the industrial parts gene data training samples to obtain the trained three-dimensional generation model.

[0016] Here, the embodiment of the present application pre-establishes a three-dimensional generation model and trains the model based on surface data training samples, edge data training samples and industrial parts gene data training samples to obtain a trained three-dimensional generation model that can achieve accurate denoising and CAD model generation. During the training process, the introduction of industrial parts DNA data can guide the generation of various industrial parts, solve the problem of part generation limitations, and improve generation accuracy. Through reliable training, the accuracy of the three-dimensional generation model can be improved, and the yield rate of the three-dimensional generation model generating the CAD model can be improved.

[0017] Optionally, obtaining the original three-dimensional generative model includes:

[0018] Constructing a surface denoising sub-model, wherein the surface denoising sub-model includes a surface encoder, a surface decoder and a surface denoising network, and the surface denoising network is composed of multiple diffusion transformers;

[0019] Constructing an edge denoising sub-model, wherein the edge denoising sub-model includes an edge encoder, an edge decoder, and an edge denoising network, and the edge denoising network is composed of multiple diffusion transformers;

[0020] Establishing an original three-dimensional generation model based on the surface denoising sub-model and the edge denoising sub-model;

[0021] The output of the surface denoising sub-model is connected to the input of the edge denoising sub-model.

[0022] Here, the embodiment of the present application pre-trains two encoders, two decoders and two denoising networks. Specifically, the surface encoder and the edge encoder are used to encode the input data into the latent variable space to improve processing efficiency. The surface denoising network and the edge denoising network are both composed of multiple diffusion transformers, which can improve the denoising performance of the surface denoising network and the edge denoising network. The embodiment of the present application also connects the output of the surface denoising sub-model to the input of the edge denoising sub-model to introduce the characteristics of the surface denoising network into the edge denoising network, thereby guiding the model to better remove edge noise, thereby improving the denoising effect and generation accuracy of the three-dimensional generation model.

[0023] Optionally, the training of the original three-dimensional generation model based on the surface data training samples, the edge data training samples, and the industrial parts gene data training samples to obtain the trained three-dimensional generation model includes:

[0024] Training the surface encoder and the surface decoder according to the surface data training sample to obtain a trained surface encoder and a trained surface decoder;

[0025] Training the edge encoder and the edge decoder according to the edge data training samples to obtain a trained edge encoder and a trained edge decoder;

[0026] Freezing the parameters of the trained surface encoder, the trained surface decoder, the trained edge encoder, and the trained edge decoder, and training the surface denoising network and the edge denoising network based on the surface data training samples, the edge data training samples, and the industrial parts gene data training samples to obtain a trained surface denoising network and a trained edge denoising network;

[0027] The trained three-dimensional generative model is obtained according to the trained surface encoder, the trained surface decoder, the trained edge encoder, the trained edge decoder, the trained face denoising network and the trained edge denoising network.

[0028] Here, in the embodiment of the present application, during the training process of the three-dimensional generation model, the surface encoder, decoder and edge encoder, decoder are first trained, and then in the subsequent training process, the data processing efficiency is improved. After the training of the surface encoder, decoder and edge encoder, decoder is completed, the parameters of the surface encoder, decoder and edge encoder, decoder are frozen respectively to train two sequential denoising models, the surface denoising network and the edge denoising network, to obtain the surface denoising network and the edge denoising network with good denoising effect, thereby realizing the training of the three-dimensional generation model with high training efficiency and strong reliability.

[0029] Optionally, the surface data training samples include a plurality of surface noise samples, surface potential feature samples corresponding to the surface noise samples, and corresponding surface denoising samples;

[0030] The edge data training samples include a plurality of edge noise samples, edge potential feature samples corresponding to the edge noise samples, and corresponding edge denoising samples;

[0031] The industrial parts gene data training samples include a plurality of industrial parts gene data samples.

[0032] Optionally, the sampled data is boundary representation data;

[0033] Accordingly, generating a three-dimensional model corresponding to the to-be-generated industrial part by using the trained denoising model includes:

[0034] Determining target boundary representation data of a tree structure corresponding to the number of samples according to an output result of the trained denoising model;

[0035] A three-dimensional model corresponding to the to-be-generated industrial part is generated according to the target boundary representation data.

[0036] Among them, in the model expression of the CAD model, the embodiment of the present application abandons the use of the "sketch and stretch" representation method that is difficult to express complex curves and surfaces, and uses B-rep to express. The geometry and topology of the B-rep model are unified into a hierarchical tree with a fixed graph topology, in which node features encode geometric information, and repeated nodes implicitly encode topological information. It has geometric accuracy and editing flexibility, and can ensure the generation efficiency of CAD models of industrial parts.

[0037] Optionally, the industrial part gene data includes at least one of shape, size, material, type, purpose, and assembly relationship.

[0038] Based on different division dimensions, the embodiment of the present application defines multiple DNA data including shape, size, material, type, purpose, and assembly relationship to guide the accurate, reliable and flexible generation of industrial parts.

[0039] In a second aspect, an embodiment of the present application provides a three-dimensional model generation device based on industrial part genes, comprising:

[0040] A sampling module, configured to obtain sampling data of the industrial part to be generated, wherein the sampling data includes surface sampling data and edge sampling data;

[0041] A first acquisition module is configured to acquire industrial part gene data of the industrial part to be generated, wherein the industrial part gene data corresponds to different phenotypes of the industrial part;

[0042] a second acquisition module, configured to acquire a trained 3D generative model, wherein the 3D generative model is implemented based on a diffusion transformer, comprises a surface denoising network and an edge denoising network, and is trained using surface data training samples, edge data training samples, and industrial part gene data training samples;

[0043] A generation module is used to input the surface sampling data, the edge sampling data and the industrial part gene data into a trained denoising model, so as to generate a three-dimensional model corresponding to the industrial part to be generated through the trained denoising model.

[0044] Optionally, before the second acquisition module is used to acquire the trained three-dimensional generative model, the apparatus further includes:

[0045] A model building module, used to obtain the original three-dimensional generative model;

[0046] A sample acquisition module is used to obtain surface data training samples, edge data training samples, and industrial parts gene data training samples;

[0047] The training module is used to train the original three-dimensional generation model according to the surface data training samples, the edge data training samples and the industrial parts gene data training samples to obtain the trained three-dimensional generation model.

[0048] Optionally, the model building module is used to:

[0049] Constructing a surface denoising sub-model, wherein the surface denoising sub-model includes a surface encoder, a surface decoder and a surface denoising network, and the surface denoising network is composed of multiple diffusion transformers;

[0050] Constructing an edge denoising sub-model, wherein the edge denoising sub-model includes an edge encoder, an edge decoder, and an edge denoising network, and the edge denoising network is composed of multiple diffusion transformers;

[0051] Establishing an original three-dimensional generation model based on the surface denoising sub-model and the edge denoising sub-model;

[0052] The output of the surface denoising sub-model is connected to the input of the edge denoising sub-model.

[0053] Optionally, the training module is used to:

[0054] Training the surface encoder and the surface decoder according to the surface data training sample to obtain a trained surface encoder and a trained surface decoder;

[0055] Training the edge encoder and the edge decoder according to the edge data training samples to obtain a trained edge encoder and a trained edge decoder;

[0056] Freezing the parameters of the trained surface encoder, the trained surface decoder, the trained edge encoder, and the trained edge decoder, and training the surface denoising network and the edge denoising network based on the surface data training samples, the edge data training samples, and the industrial parts gene data training samples to obtain a trained surface denoising network and a trained edge denoising network;

[0057] The trained three-dimensional generative model is obtained according to the trained surface encoder, the trained surface decoder, the trained edge encoder, the trained edge decoder, the trained face denoising network and the trained edge denoising network.

[0058] Optionally, the surface data training samples include a plurality of surface noise samples, surface potential feature samples corresponding to the surface noise samples, and corresponding surface denoising samples;

[0059] The edge data training samples include a plurality of edge noise samples, edge potential feature samples corresponding to the edge noise samples, and corresponding edge denoising samples;

[0060] The industrial parts gene data training samples include a plurality of industrial parts gene data samples.

[0061] The sampling data is boundary representation data;

[0062] Accordingly, the generation module is specifically used for:

[0063] Determining target boundary representation data of a tree structure corresponding to the number of samples according to an output result of the trained denoising model;

[0064] A three-dimensional model corresponding to the to-be-generated industrial part is generated according to the target boundary representation data.

[0065] Optionally, the industrial part gene data includes at least one of shape, size, material, type, purpose, and assembly relationship.

[0066] In a third aspect, the present application provides a device for generating a three-dimensional model based on industrial part genes, comprising: a memory, a processor;

[0067] The memory stores computer-executable instructions;

[0068] The processor executes the computer-executable instructions stored in the memory, so that the processor executes the above first aspect and / or various possible implementations of the first aspect.

[0069] In a fourth aspect, the present application provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed by a processor, they are used to implement the above first aspect and / or various possible implementations of the first aspect.

[0070] In a fifth aspect, the present application provides a computer program product, comprising a computer program, which, when executed by a processor, implements the above first aspect and / or various possible implementations of the first aspect.

[0071] The three-dimensional model generation method, device and medium based on industrial part genes provided in the present application, for the industrial part to be generated, samples its surface sampling data and edge sampling data, and obtains the corresponding industrial part DNA data, and inputs the above-mentioned surface sampling data and edge sampling data and the surface sampling data and edge sampling data into the trained three-dimensional generation model, so that the CAD model corresponding to the industrial part to be generated can be generated by the trained three-dimensional generation model. The three-dimensional generation model here implements surface denoising and edge denoising based on the DiT architecture. The DiT architecture solves the limitations of the traditional diffusion model, can efficiently model long-distance dependencies, and capture richer global features. While improving the modeling ability of the denoising network, it solves the problem of difficult training convergence, and provides support for accurate denoising of the three-dimensional generation model. The structure of the three-dimensional generation model treats the vertex connection as an edge, and only designs two surface and edge denoising networks, avoiding the design of vertex denoising networks, reducing the amount of calculation, and improving generation efficiency. On the basis of this model architecture, combined with industrial part DNA data, it can guide the generation of industrial parts, meet the requirements of complex CAD model generation, and improve the efficiency of CAD model generation. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0073] Figure 1 A schematic diagram showing a mapping of biological genes to industrial part genes provided in an embodiment of the present application;

[0074] Figure 2 A schematic diagram of the architecture of a 3D model generation system based on industrial part genes provided in an embodiment of the present application;

[0075] Figure 3 A schematic flow chart of a method for generating a three-dimensional model based on industrial part genes provided in an embodiment of the present application;

[0076] Figure 4 A schematic diagram of the structure of a three-dimensional generation model provided in an embodiment of the present application;

[0077] Figure 5 A schematic diagram of the structure of a three-dimensional model generation device based on industrial part genes provided in an embodiment of the present application;

[0078] Figure 6 A schematic structural diagram of a three-dimensional model generation device based on industrial part genes provided in an embodiment of the present application.

[0079] The above drawings illustrate specific embodiments of the present application, which will be described in more detail below. These drawings and the textual description are not intended to limit the scope of the present application in any way, but rather to illustrate the concepts of the present application to those skilled in the art by reference to specific embodiments. DETAILED DESCRIPTION

[0080] Exemplary embodiments are described in detail herein, with examples illustrated in the accompanying drawings. When the following description refers to the drawings, identical numerals in different drawings represent identical or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments are not intended to represent all embodiments consistent with this application.

[0081] First, let’s explain the terms involved in this application:

[0082] CAD models: CAD models are three-dimensional models created using computer-aided design (CAD) software. They contain information such as an object's geometry, dimensions, and materials, and can be visualized and edited. CAD models are widely used in the design and manufacturing processes of architecture, machinery, electronics, aviation, and other fields.

[0083] Sketch and Extrude: It is one of the main formats for describing CAD models. The workflow is to draw loops of two-dimensional (2D) curves as external and internal boundaries to create 2D contours; extrude 2D contours into 3D shapes; and add or subtract 3D shapes to build complex CAD models.

[0084] B-rep: One of the primary formats for describing CAD models, it is widely used in free-form surface modeling to express complex geometries. B-rep allows parametric curves and surfaces to replace the planes and linear edges used in mesh modeling. A B-rep consists of a set of interconnected faces, edges, and vertices. A face is the visible area of ​​a parametric surface, bounded by the closed loops formed by its adjacent edges, while an edge is the visible area of ​​a parametric curve, trimmed by the vertices that define its start and end points. Recording the adjacency relationships between adjacent edges and vertices enables the structure to provide a complete description of the final solid shape.

[0085] Currently, the "sketch and extrude" representation method is limited in its ability to generate 3D shapes for industrial parts. It is limited to sketches composed of three primitives: lines, arcs, and circles. It is also limited to performing extrusion operations, making it impossible to create complex 3D shapes composed of complex curves and surfaces. Another representation method, B-rep, is a predominant format for describing shapes and is widely used in free-form surface modeling to express complex geometries. Existing boundary-based CAD model generation methods are limited to specific 3D shape primitives (such as prisms), limiting the complexity of the results and the amount of training data available. Existing boundary-based CAD model generation methods suffer from slow inference speeds and struggle to generate well-structured geometries and correct topological relationships. When given constraints (such as categories), the generated models have a low yield rate. Comprehensively speaking, CAD model generation is limited to specific shape primitives, unable to meet the requirements for complex CAD model generation and resulting in low CAD model generation efficiency.

[0086] In order to solve the above problems, the embodiments of the present application provide a three-dimensional model generation method, equipment and medium based on industrial part genes. The method samples the surface sampling data and edge sampling data of the industrial part to be generated, and obtains the corresponding industrial part DNA data. The above-mentioned surface sampling data and edge sampling data and the surface sampling data and edge sampling data are input into the trained three-dimensional generation model. The CAD model corresponding to the industrial part to be generated can be generated through the trained three-dimensional generation model.

[0087] In order to efficiently generate CAD models of complex and diverse industrial parts based on three-dimensional generative models, the main issues that need to be considered in the embodiments of this application include: Problem 1, solving the expression problem of CAD models, mainly to study clearly what the input and output of the network should be: that is, what kind of expression method is used to represent the CAD model, and then encode this expression method as input, send it to the generative model, and decode the output of the model to obtain the final CAD model. Problem 2, studying an end-to-end network, the network needs to solve the encoding and decoding problems of the CAD model, that is, how to encode the 3D model file into an input signal that the network can understand, and how to decode and reconstruct the 3D model from the output of the network. How to design the network framework and backbone modules so that it can generate specific 3D models according to constraints. Problem 3, how to optimize the structure of the network so that it can efficiently generate CAD models, and improve various indicators such as the yield rate of the generated model.

[0088] Taking the above problems into consideration, an optional method of the three-dimensional generation model in the embodiment of the present application is that the three-dimensional generation model implements surface denoising and edge denoising based on the DiT architecture. The DiT architecture solves the limitations of the traditional diffusion model, can efficiently model long-distance dependencies, and capture richer global features. While improving the modeling ability of the denoising network, it solves the problem of difficult training convergence, and provides support for accurate denoising of the three-dimensional generation model. The structure of the three-dimensional generation model avoids the design of vertex denoising networks by treating the vertex connection as an edge, and only needs to design two denoising networks, surface and edge, thereby reducing the amount of calculation and improving generation efficiency. On the basis of this model architecture, combined with industrial parts DNA data, it can guide the generation of industrial parts, meet the requirements of complex CAD model generation, and improve the efficiency of CAD model generation.

[0089] Optionally, the embodiment of the present application is implemented based on the DNA of industrial parts. In the concept of digital genealogy of industrial parts, the basic attributes of the parts can be mapped to similar elements in biological DNA. For example, Figure 1 The embodiment of the present application provides a schematic diagram of mapping biological genes to industrial parts genes. It can be understood that: Figure 1 This is only a schematic representation of a feasible way of industrial parts DNA and does not affect the scope of protection of the embodiments of this application. Figure 1 Understand the evolution of industrial parts and their morphology, including shape, size, material, type, purpose, and fitting relationships.

[0090] Shape: The shape of a part is similar to the appearance of an organism. Different parts have different shapes, such as plate-shaped, T-shaped, block-shaped, etc.

[0091] Size: The dimensions of a part, including length, width, and height, are analogous to physical characteristics such as height and weight of an organism. Industrial parts are categorized by size into large, medium, and small.

[0092] Material: The material composition of a part, such as metal, plastic, or rubber. This corresponds to the concept of surface features in biological DNA, reflecting the inherent properties of the part. The specific surface features are not specifically limited in this embodiment and can be determined based on actual conditions. Examples are not provided in the figures.

[0093] Type: The classification of a part, meaning the specific category or lineage it belongs to, such as a gear, pipe, or bolt. Similar to species classification in biological classification, this DNA defines the type and structure of the part.

[0094] Purpose: The intended function or role of a part in a system, similar to a human occupation. This attribute determines the part's contribution to the overall system, such as transmission, fastening, clamping, and other functions.

[0095] Assembly relationships: This DNA is a unique property of a part, with no direct correspondence to biological entities. It describes the assembly relationship between a part and other parts in the system, specifically how they fit together and interact. Assembly relationships determine the compatibility and performance of parts within the overall assembly, including press fits, threaded connections, and clamped fits.

[0096] Together, these elements constitute the DNA of industrial parts, providing a comprehensive perspective on their characteristics, functions, and evolution within manufacturing systems. Just as biological DNA encapsulates the genetic information required for an organism's development and adaptation, the DNA of industrial parts enables the tracking of their historical evolution, modifications, and future potential, supporting optimization and customization in modern manufacturing. Furthermore, this DNA can further facilitate the generation of industrial parts under diverse constraints.

[0097] The embodiments of the present application draw on the concept of genes in the biological world and define the genes of industrial parts from different dimensions, corresponding to different phenotypes of industrial parts.

[0098] In the embodiment of the present application, industrial part DNA is used to guide the generation of industrial parts in a three-dimensional generation model of part generation, making the generated industrial part effects more explainable.

[0099] Optional, Figure 2 This is a schematic diagram of the architecture of a 3D model generation system based on industrial parts genes provided in an embodiment of the present application. Figure 1 In the above architecture, the above architecture includes at least one of a data acquisition device 201, a processing device 202 and a display device 203.

[0100] It is understood that the structure illustrated in the embodiments of this application does not constitute a specific limitation on the architecture of the 3D model generation system based on industrial part genes. In other feasible embodiments of this application, the above architecture may include more or fewer components than shown, or may combine or split certain components, or arrange the components differently. The specific configuration can be determined based on the actual application scenario and is not limited here. Figure 1 The components shown can be implemented in hardware, software, or a combination of software and hardware.

[0101] In a specific implementation process, the data acquisition device 201 may include an input / output interface and may also include a communication interface. The data acquisition device 201 may be connected to the processing device via the input / output interface or the communication interface.

[0102] The processing device 202 can sample the surface sampling data and edge sampling data of the industrial parts to be generated, and obtain the corresponding industrial part DNA data, and input the above surface sampling data and edge sampling data and the surface sampling data and edge sampling data into the trained three-dimensional generation model, so as to generate the CAD model corresponding to the industrial parts to be generated through the trained three-dimensional generation model.

[0103] Optionally, the processing device 202 is further configured to construct and train a three-dimensional generative model based on industrial part DNA.

[0104] The display device 203 may also be a touch screen display or a screen of a terminal device, which is used to receive user instructions while displaying the above-mentioned content to achieve interaction with the user.

[0105] It should be understood that the above-mentioned processing device can be implemented by a processor reading instructions in a memory and executing the instructions, or it can be implemented by a chip circuit.

[0106] In addition, the network architecture and business scenarios described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Ordinary technicians in this field can know that with the evolution of network architecture and the emergence of new business scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0107] The technical solution of this application is described in detail below with reference to specific embodiments:

[0108] Optionally, Figure 3 A flow chart of a method for generating a three-dimensional model based on industrial part genes provided in an embodiment of the present application. The execution subject of the embodiment of the present application may be Figure 2 The specific execution subject of the processing device 202 can be determined according to the actual application scenario. Figure 3 As shown, the method includes the following steps:

[0109] S301: Acquire sampling data of the industrial parts to be generated.

[0110] The sampling data includes surface sampling data and edge sampling data.

[0111] S302: Obtain industrial part gene data of the industrial part to be generated.

[0112] Among them, the industrial parts genetic data corresponds to different phenotypes of industrial parts.

[0113] Optionally, the industrial part gene data includes at least one of shape, size, material, type, purpose, and assembly relationship.

[0114] Based on different division dimensions, the embodiment of the present application defines multiple DNA data including shape, size, material, type, purpose, and assembly relationship to guide the accurate, reliable and flexible generation of industrial parts.

[0115] S303: Obtain a trained three-dimensional generative model.

[0116] Among them, the three-dimensional generation model is implemented based on the diffusion transformer. The three-dimensional generation model includes a surface denoising network and an edge denoising network. The trained three-dimensional generation model is obtained by training surface data training samples, edge data training samples and industrial parts gene data training samples.

[0117] S304: Inputting the surface sampling data, edge sampling data and industrial part gene data into the trained denoising model to generate a three-dimensional model corresponding to the industrial part to be generated through the trained denoising model.

[0118] Optionally, the sampling data is boundary representation data; accordingly, a three-dimensional model corresponding to the industrial part to be generated is generated through a trained denoising model, including: determining target boundary representation data of a tree structure corresponding to the sampling number based on the output result of the trained denoising model; and generating a three-dimensional model corresponding to the industrial part to be generated based on the target boundary representation data.

[0119] In one possible implementation, the embodiment of the present application abandons the "sketch and stretch" representation method that is difficult to express complex curves and surfaces in the model expression of the CAD model, and uses B-rep to express it. B-rep consists of geometric elements (face, edge, edge, vertex adjacency matrix) with pairwise topological relationships (face-edge, edge-vertex adjacency matrix). In order to uniformly express the two forms of data, topology and geometry, the geometry and topology of the B-rep model are unified into a hierarchical tree with a fixed graph topology, where node features encode geometric information, and repeated nodes (i.e., nodes with almost the same features during the generation process) implicitly encode topological information. The root node represents the entire CAD entity, so the tree has three levels: face, edge, and vertex (from top to bottom). The network input and output are such a tree-structured B-rep expression.

[0120] Among them, in the model expression of the CAD model, the embodiment of the present application abandons the use of the "sketch and stretch" representation method that is difficult to express complex curves and surfaces, and uses B-rep to express. The geometry and topology of the B-rep model are unified into a hierarchical tree with a fixed graph topology, in which node features encode geometric information, and repeated nodes implicitly encode topological information. It has geometric accuracy and editing flexibility, and can ensure the generation efficiency of CAD models of industrial parts.

[0121] An embodiment of the present application provides a CAD model generation method based on industrial part DNA. For the industrial part to be generated, its surface sampling data and edge sampling data are sampled, and the corresponding industrial part DNA data is obtained. The above-mentioned surface sampling data and edge sampling data and the surface sampling data and edge sampling data are input into a trained three-dimensional generation model. The CAD model corresponding to the industrial part to be generated can be generated by the trained three-dimensional generation model. The three-dimensional generation model here implements surface denoising and edge denoising based on the DiT architecture. The DiT architecture solves the limitations of the traditional diffusion model, can efficiently model long-distance dependencies, and capture richer global features. While improving the modeling ability of the denoising network, it solves the problem of difficult training convergence, and provides support for accurate denoising of the three-dimensional generation model. The structure of the three-dimensional generation model regards the vertex connection as an edge, and only two surface and edge denoising networks are designed, avoiding the design of the vertex denoising network, reducing the amount of calculation, and improving the generation efficiency. On the basis of this model architecture, combined with the industrial part DNA data, it can guide the generation of industrial parts, meet the complex CAD model generation requirements, and improve the efficiency of CAD model generation.

[0122] Optionally, before obtaining the trained three-dimensional generative model, it also includes: obtaining the original three-dimensional generative model; obtaining surface data training samples, edge data training samples and industrial parts gene data training samples; training the original three-dimensional generative model based on the surface data training samples, edge data training samples and industrial parts gene data training samples to obtain the trained three-dimensional generative model.

[0123] Here, the embodiment of the present application pre-establishes a three-dimensional generation model and trains the model based on surface data training samples, edge data training samples and industrial parts gene data training samples to obtain a trained three-dimensional generation model that can achieve accurate denoising and CAD model generation. During the training process, the introduction of industrial parts DNA data can guide the generation of various industrial parts, solve the problem of part generation limitations, and improve generation accuracy. Through reliable training, the accuracy of the three-dimensional generation model can be improved, and the yield rate of the three-dimensional generation model generating the CAD model can be improved.

[0124] Optionally, obtaining the original three-dimensional generative model includes: constructing a surface denoising sub-model, wherein the surface denoising sub-model includes a surface encoder, a surface decoder and a surface denoising network, and the surface denoising network is composed of multiple diffusion transformers; constructing an edge denoising sub-model, wherein the edge denoising sub-model includes an edge encoder, an edge decoder and an edge denoising network, and the edge denoising network is composed of multiple diffusion transformers; establishing the original three-dimensional generative model based on the surface denoising sub-model and the edge denoising sub-model; wherein the output of the surface denoising sub-model is connected to the input of the edge denoising sub-model.

[0125] Here, the embodiment of the present application pre-trains two encoders, two decoders and two denoising networks. Specifically, the surface encoder and the edge encoder are used to encode the input data into the latent variable space to improve processing efficiency. The surface denoising network and the edge denoising network are both composed of multiple diffusion transformers, which can improve the denoising performance of the surface denoising network and the edge denoising network. The embodiment of the present application also connects the output of the surface denoising sub-model to the input of the edge denoising sub-model to introduce the characteristics of the surface denoising network into the edge denoising network, thereby guiding the model to better remove edge noise, thereby improving the denoising effect and generation accuracy of the three-dimensional generation model.

[0126] Demonstratively, Figure 4 A schematic diagram of the structure of a three-dimensional generation model provided in an embodiment of the present application is shown in FIG. Figure 4 As shown, the three-dimensional generation model provided in the embodiment of the present application is a pipeline model.

[0127] like Figure 4 As shown in the figure, the encoder consists of two downsampling layers (DownBlock), a residual layer (ResNetBlock), an intermediate processing layer (MidBlock), and a convolutional layer (Global Self-Attention and Convolution, GSC). The decoder consists of two upsampling layers (UpBlock), a ResNetBlock, a MidBlock, and a GSC. The MidBlock module is a Self-Attention module.

[0128] based on Figure 4 In the three-dimensional generative model, an optional method of the embodiment of the present application is: first, two encoders are trained separately using the edges and faces of the boundary expression, so as to train the encoder and decoder. The encoder is responsible for encoding the input data into the latent variable space, where the model processes the input data more efficiently, and the decoder is responsible for decoding the network output, thereby reconstructing the B-rep expression of the tree structure, which can be further converted into a CAD entity.

[0129] Optionally, the encoder can be a variational autoencoder (VAE). VAEs generate continuous and diverse data with stability and reliability, and their latent space is semantically meaningful, improving the interpretability of 3D generative models. VAEs also enable end-to-end training and can be combined with other models to meet encoding and decoding requirements, improving the processing efficiency of 3D generative models for industrial parts.

[0130] Optionally, the original three-dimensional generative model is trained according to the surface data training samples, the edge data training samples and the industrial parts gene data training samples to obtain a trained three-dimensional generative model, including: training the surface encoder and the surface decoder according to the surface data training samples to obtain a trained surface encoder and a trained surface decoder; training the edge encoder and the edge decoder according to the edge data training samples to obtain a trained edge encoder and a trained edge decoder; freezing the parameters of the trained surface encoder, the trained surface decoder, the trained edge encoder and the trained edge decoder, and training the surface denoising network and the edge denoising network according to the surface data training samples, the edge data training samples and the industrial parts gene data training samples to obtain a trained surface denoising network and a trained edge denoising network; obtaining the trained three-dimensional generative model according to the trained surface encoder, the trained surface decoder, the trained edge encoder, the trained edge decoder, the trained surface denoising network and the trained edge denoising network.

[0131] Here, in the embodiment of the present application, during the training process of the three-dimensional generation model, the surface encoder, decoder and edge encoder, decoder are first trained, and then in the subsequent training process, the data processing efficiency is improved. After the training of the surface encoder, decoder and edge encoder, decoder is completed, the parameters of the surface encoder, decoder and edge encoder, decoder are frozen to train two sequential denoising models, the surface denoising network and the edge denoising network, to obtain the surface denoising network and the edge denoising network with good denoising effect, thereby realizing the training of the three-dimensional generation model with high training efficiency and strong reliability.

[0132] Optionally, the surface data training samples include multiple surface noise samples, surface potential feature samples corresponding to the surface noise samples, and corresponding surface denoising samples; the edge data training samples include multiple surface noise samples, edge potential feature samples corresponding to the edge noise samples, and corresponding edge denoising samples; the industrial part gene data training samples include multiple industrial part gene data samples.

[0133] Optionally, the surface noise samples and the surface noise samples can be obtained by adding noise based on the surface denoised samples and the edge denoised samples.

[0134] based on Figure 4 The 3D generative model in

[15] needs to be trained for denoising. To avoid the instability of training caused by directly generating a complete tree-like B-rep, a sequential denoising method is used, first denoising the faces and then the edges.

[0135] In order to improve the reasoning speed, the embodiment of the present application proposes to connect two points and regard them as an edge, thereby avoiding denoising the vertices and reducing the amount of calculation.

[0136] In one possible implementation, based on Figure 4 The training process of the 3D generative model is as follows:

[0137] Training VAE: The edges and faces of the boundary expression correspond to the training of two different VAEs, but the network structure of the two VAEs is the same, only the edge or face data input is used during training. Taking the edge VAE training as an example: First, the input data is the shape feature in the CAD B-rep , which is a one-dimensional (1D) array of 3D points sampled along the parameterized curve, and an encoder is used to obtain the potential features in the latent space. , and then decoded by a decoder to get the output The encoder is trained with a Mean Squared Error (MSE) reconstruction loss and a Kullback-Leibler Divergence (KL) regularization term.

[0138] The loss function of the encoder is:

[0139]

[0140] in, represents the loss function of the encoder, represents the KL regularized image, represents the MSE reconstruction loss, Represents the mean and variance of the encoder samples.

[0141] Training the surface and edge denoising networks: After training the edge and face VAE models, the parameters of these models are frozen to train the sequential denoising models (i.e., the face denoising network and the edge denoising network). To more efficiently model long-range dependencies and capture richer global features, the DiT architecture replaces the UNet network as the core module of the denoising network. Compared to the common transformer-based diffusion model, the DiT architecture improves the modeling capability of the denoising network while addressing the problem of poor training convergence through designs such as the cross-attention block (which concatenates the embeddings of t and c into a sequence and adds an additional multi-head cross-attention layer after the self-attention block) and the adaptive layer normalization block (which replaces standard layer normalization in the transformer block with adaptive layer normalization). Optionally, the denoising network has eight DiT blocks, each with 24 attention layers and 12 heads. The output of the denoising network is passed through the corresponding decoder to reconstruct the denoised edges or faces.

[0142] During training, input sampled face features , the potential features are obtained through the face encoder , then , time step t and category condition c (if there is a category, it is input, if not, it is not input, corresponding to conditional generation and unconditional generation respectively) are embedded and input into the face denoising network The edge denoising network and the face denoising network have essentially the same structure, differing in that the edge denoising network incorporates the denoised face feature F as input, thereby guiding the network to better remove noise from edges. The category condition c here is the industrial part DNA in the present embodiment.

[0143] The loss functions of the surface and edge denoising networks are:

[0144]

[0145]

[0146] in, is the loss function of the face denoising network, is the loss function of the edge denoising network, and Represent the potential features of clean nodes respectively and The parameter t represents the time step, c represents the category label, is a constant hyperparameter, , refers to the real noise added by the forward process at time step t, which is the supervisory signal for model training.

[0147] Based on the above embodiments, the embodiments of the present application can achieve the following technical effects:

[0148] The CAD model is expressed using a boundary representation (B-rep), and the geometry and topology are unified into a hierarchical tree with a fixed graph topology. By training a denoising model, a clean tree structure is restored from a noisy tree structure, thereby mastering the ability to generate CAD models. This data expression method solves the problem that the "sketch and stretch" expression method has difficulty in generating 3D shapes with complex curves or surfaces.

[0149] In terms of the design of the denoising network, sequential denoising avoids the difficulty of generating a complete model at one time, and by connecting two vertices as an edge, there is no need to design a separate vertex denoising network, which reduces the amount of calculation and improves the inference speed.

[0150] By introducing the DiT architecture as the core module of the denoising network, and directly modeling global dependencies through self-attention and cross-attention mechanisms, the network's expressive power is improved, addressing the difficulty of training diffusion models based on transformer structures. This makes the generated CAD model's geometry and topology more consistent with industrial reality, resulting in a higher yield rate. Furthermore, by incorporating the DNA of industrial parts, the model becomes more interpretable.

[0151] Figure 5 A schematic diagram of the structure of a three-dimensional model generation device based on industrial part genes provided in an embodiment of the present application is shown in FIG. Figure 5 As shown, the apparatus of this embodiment of the present application includes: a sampling module 501, a first acquisition module 502, a second acquisition module 503, and a generation module 504. The apparatus for generating a 3D model based on an industrial part gene can be the aforementioned processing device itself, or a chip or integrated circuit that implements the functions of the processing device. It should be noted that the division of sampling module 501, first acquisition module 502, second acquisition module 503, and generation module 504 is merely a logical functional division; physically, the two can be integrated or independent.

[0152] The sampling module is used to obtain sampling data of the industrial part to be generated, wherein the sampling data includes surface sampling data and edge sampling data;

[0153] A first acquisition module is used to acquire industrial part gene data of the industrial part to be generated, wherein the industrial part gene data corresponds to different phenotypes of the industrial part;

[0154] A second acquisition module is configured to acquire a trained 3D generative model, wherein the 3D generative model is implemented based on a diffusion transformer and includes a surface denoising network and an edge denoising network. The trained 3D generative model is obtained by training surface data training samples, edge data training samples, and industrial parts gene data training samples;

[0155] The generation module is used to input the surface sampling data, edge sampling data and industrial part gene data into the trained denoising model to generate a three-dimensional model corresponding to the industrial part to be generated through the trained denoising model.

[0156] Optionally, before the second acquisition module is used to acquire the trained three-dimensional generative model, the apparatus further includes:

[0157] A model building module, used to obtain the original three-dimensional generative model;

[0158] A sample acquisition module is used to obtain surface data training samples, edge data training samples, and industrial parts gene data training samples;

[0159] The training module is used to train the original three-dimensional generation model according to the surface data training samples, the edge data training samples and the industrial parts gene data training samples to obtain a trained three-dimensional generation model.

[0160] Optionally, the model building module is used to:

[0161] Constructing a surface denoising sub-model, wherein the surface denoising sub-model includes a surface encoder, a surface decoder, and a surface denoising network, and the surface denoising network is composed of multiple diffusion transformers;

[0162] Construct an edge denoising sub-model, where the edge denoising sub-model includes an edge encoder, an edge decoder, and an edge denoising network, and the edge denoising network is composed of multiple diffusion transformers;

[0163] According to the surface denoising sub-model and the edge denoising sub-model, the original 3D generative model is established;

[0164] The output of the face denoising sub-model is connected to the input of the edge denoising sub-model.

[0165] Optionally, the training module is used to:

[0166] Training the surface encoder and the surface decoder according to the surface data training samples to obtain a trained surface encoder and a trained surface decoder;

[0167] Training the edge encoder and the edge decoder according to the edge data training samples to obtain a trained edge encoder and a trained edge decoder;

[0168] Freeze the parameters of the trained surface encoder, the trained surface decoder, the trained edge encoder, and the trained edge decoder, and train the face denoising network and the edge denoising network based on face data training samples, edge data training samples, and industrial parts gene data training samples to obtain a trained face denoising network and a trained edge denoising network;

[0169] A trained 3D generative model is obtained according to the trained surface encoder, the trained surface decoder, the trained edge encoder, the trained edge decoder, the trained face denoising network and the trained edge denoising network.

[0170] Optionally, the surface data training samples include a plurality of surface noise samples, surface potential feature samples corresponding to the surface noise samples, and corresponding surface denoising samples;

[0171] The edge data training samples include multiple edge noise samples, edge potential feature samples corresponding to the edge noise samples, and corresponding edge denoising samples;

[0172] The industrial parts gene data training samples include multiple industrial parts gene data samples.

[0173] The sampling data is the boundary representation data;

[0174] Accordingly, the generation module is specifically used to:

[0175] According to the output results of the trained denoising model, the target boundary representation data of the tree structure corresponding to the sampling number is determined;

[0176] According to the target boundary representation data, a three-dimensional model corresponding to the industrial part to be generated is generated.

[0177] Optionally, the industrial part gene data includes at least one of shape, size, material, type, purpose, and assembly relationship.

[0178] refer to Figure 6 , which shows a schematic structural diagram of a 3D model generation device 600 based on industrial part genes suitable for implementing embodiments of the present disclosure. This 3D model generation device 600 based on industrial part genes can be a terminal device or a server. Terminal devices include, but are not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, personal digital assistants (PDAs), tablet computers (Android devices), portable media players (PMPs), and in-vehicle terminals (e.g., in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. Figure 6 The three-dimensional model generation device based on industrial part genes shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present disclosure.

[0179] like Figure 6 As shown, a 3D model generation device 600 based on an industrial part gene may include a processing device (e.g., a central processing unit, a graphics processing unit, etc.) 601, which can perform various appropriate actions and processes based on programs stored in a read-only memory (ROM) 602 or programs loaded from a storage device 608 into a random access memory (RAM) 603. RAM 603 also stores various programs and data required for the operation of the 3D model generation device 600 based on an industrial part gene. Processing device 601, ROM 602, and RAM 603 are interconnected via a bus 604. An input / output (I / O) interface 605 is also connected to bus 604.

[0180] Typically, the following devices can be connected to the I / O interface 605: input devices 606 including, for example, a touch screen, a touchpad, a keyboard, a mouse, a camera, a microphone, an accelerometer, a gyroscope, etc.; output devices 607 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 608 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 609. The communication device 609 can allow the three-dimensional model generation device 600 based on industrial part genes to communicate with other devices wirelessly or by wire to exchange data. Although Figure 6 The three-dimensional model generating apparatus 600 based on the industrial part gene is shown with various devices, but it should be understood that it is not required to implement or possess all the devices shown, and more or fewer devices may be implemented or possessed instead.

[0181] In particular, according to an embodiment of the present disclosure, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present disclosure includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network via the communication device 609, or installed from the storage device 608, or installed from the ROM 602. When the computer program is executed by the processing device 601, the above-mentioned functions defined in the method of the embodiment of the present disclosure are performed.

[0182] It should be noted that the computer-readable medium described above in the present disclosure may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. Computer-readable storage media may include, for example, but not limited to, electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, devices, or components, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection having one or more conductors, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such a propagated data signal may take a variety of forms, including, but not limited to, electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium may also be any computer-readable medium other than a computer-readable storage medium that can transmit, propagate, or transport a program for use by or in connection with an instruction execution system, apparatus, or device. Program code embodied on a computer-readable medium may be transmitted using any suitable medium, including but not limited to wire, optical cable, RF (radio frequency), or any suitable combination thereof.

[0183] The computer-readable medium may be included in the three-dimensional model generation device based on industrial part genes, or may exist independently without being assembled into the three-dimensional model generation device based on industrial part genes.

[0184] The above-mentioned computer-readable medium carries one or more programs. When the above-mentioned one or more programs are executed by the three-dimensional model generation device based on industrial part genes, the three-dimensional model generation device based on industrial part genes executes the method shown in the above-mentioned embodiment.

[0185] Computer program code for performing the operations of the present disclosure may be written in one or more programming languages, or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a stand-alone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving a remote computer, the remote computer may be connected to the user's computer via any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0186] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present disclosure. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the module, program segment, or a part of code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order than that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram and / or flowchart, and the combination of the boxes in the block diagram and / or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0187] The units involved in the embodiments described in this disclosure may be implemented in software or hardware. In some cases, the name of a unit does not limit the unit itself. For example, the first acquisition unit may also be described as a "unit for acquiring at least two Internet Protocol addresses."

[0188] The functions described above herein may be performed, at least in part, by one or more hardware logic components. For example, and without limitation, exemplary types of hardware logic components that may be used include: field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), systems on chips (SOCs), complex programmable logic devices (CPLDs), and the like.

[0189] In the context of the present disclosure, a machine-readable medium may be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium may be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium may include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing. More specific examples of machine-readable storage media may include an electrical connection based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), optical fibers, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0190] The three-dimensional model generation device based on industrial parts genes in the embodiment of the present application can be used to execute the technical solutions in the above-mentioned method embodiments of the present application. Its implementation principles and technical effects are similar and will not be repeated here.

[0191] An embodiment of the present application also provides a computer-readable storage medium, which stores computer-executable instructions. When the computer-executable instructions are executed by a processor, they are used to implement any of the above-mentioned methods for generating a three-dimensional model based on industrial part genes.

[0192] An embodiment of the present application also provides a computer program product, including a computer program, which, when executed by a processor, is used to implement any of the above-mentioned methods for generating a three-dimensional model based on industrial part genes.

[0193] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interface, device or unit, which can be electrical, mechanical or other forms.

[0194] In addition, the functional units in the various embodiments of the present application may be integrated into a single processing unit, or each unit may exist physically separately, or two or more units may be integrated into a single unit. The aforementioned integrated units may be implemented in the form of hardware or software functional units.

[0195] Those skilled in the art will readily appreciate other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. This application is intended to cover any variations, uses, or adaptations of the present invention that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The description and examples are to be considered as exemplary only.

[0196] It will be understood that the present application is not limited to the exact construction that has been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof.

Claims

1. A three-dimensional model generation method based on industrial parts genes, characterized in that: include: Acquire sampling data of the industrial part to be generated, wherein the sampling data includes surface sampling data and edge sampling data; Obtaining industrial part gene data of the industrial part to be generated, wherein the industrial part gene data corresponds to different phenotypes of the industrial part; the industrial part gene data includes at least one of shape, size, material, type, purpose, and assembly relationship; Obtaining a trained 3D generative model, wherein the 3D generative model is implemented based on a diffusion transformer, the 3D generative model includes a surface denoising sub-model and an edge denoising sub-model, the surface denoising sub-model includes a surface encoder, a surface decoder, and a surface denoising network, the edge denoising sub-model includes an edge encoder, an edge decoder, and an edge denoising network, the output of the surface denoising sub-model is connected to the input of the edge denoising sub-model, and the surface denoising network and the edge denoising network are both composed of multiple diffusion transformers; the trained 3D generative model is obtained by training surface data training samples, edge data training samples, and industrial part gene data training samples; Inputting the surface sampling data, the edge sampling data and the industrial part gene data into a trained denoising model, so as to generate a three-dimensional model corresponding to the industrial part to be generated through the trained denoising model; The sampling data is boundary representation data; Accordingly, generating a three-dimensional model corresponding to the to-be-generated industrial part by using the trained denoising model includes: Determining target boundary representation data of a tree structure corresponding to the number of samples according to an output result of the trained denoising model; A three-dimensional model corresponding to the to-be-generated industrial part is generated according to the target boundary representation data.

2. The method according to claim 1, characterized in that Before obtaining the trained three-dimensional generative model, the method further includes: Obtain the original 3D generative model; Obtaining surface data training samples, edge data training samples, and industrial parts gene data training samples; The original three-dimensional generation model is trained according to the surface data training samples, the edge data training samples and the industrial parts gene data training samples to obtain the trained three-dimensional generation model.

3. The method according to claim 2, characterized in that The obtaining of the original three-dimensional generative model includes: Constructing a surface denoising sub-model, wherein the surface denoising sub-model includes a surface encoder, a surface decoder and a surface denoising network, and the surface denoising network is composed of multiple diffusion transformers; Constructing an edge denoising sub-model, wherein the edge denoising sub-model includes an edge encoder, an edge decoder, and an edge denoising network, and the edge denoising network is composed of multiple diffusion transformers; Establishing an original three-dimensional generation model based on the surface denoising sub-model and the edge denoising sub-model; The output of the surface denoising sub-model is connected to the input of the edge denoising sub-model.

4. The method according to claim 3, characterized in that The training of the original three-dimensional generation model based on the surface data training samples, the edge data training samples, and the industrial parts gene data training samples to obtain the trained three-dimensional generation model includes: Training the surface encoder and the surface decoder according to the surface data training sample to obtain a trained surface encoder and a trained surface decoder; Training the edge encoder and the edge decoder according to the edge data training samples to obtain a trained edge encoder and a trained edge decoder; Freezing the parameters of the trained surface encoder, the trained surface decoder, the trained edge encoder, and the trained edge decoder, and training the surface denoising network and the edge denoising network based on the surface data training samples, the edge data training samples, and the industrial parts gene data training samples to obtain a trained surface denoising network and a trained edge denoising network; The trained three-dimensional generative model is obtained according to the trained surface encoder, the trained surface decoder, the trained edge encoder, the trained edge decoder, the trained face denoising network and the trained edge denoising network.

5. The method according to claim 4, characterized in that The surface data training samples include a plurality of surface noise samples, surface potential feature samples corresponding to the surface noise samples, and corresponding surface denoising samples; The edge data training samples include a plurality of edge noise samples, edge potential feature samples corresponding to the edge noise samples, and corresponding edge denoising samples; The industrial parts gene data training samples include a plurality of industrial parts gene data samples.

6. A 3D model generation device based on industrial parts genes, characterized in that: include: Memory, processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory, so that the processor performs the method according to any one of claims 1 to 5.

7. A computer-readable storage medium, characterized in that The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 5 when executed by a processor.

8. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 5 when executed by a processor.

Images