Traceable mapping 2D-3D through design method and device

Abstract

The application discloses a 2D-3D through design method and device with traceable mapping, and relates to the technical field of vehicle industrial design. The method comprises the following steps: constructing a parameterized engineering constraint set, obtaining a reference graph, a draft and a user instruction, obtaining standard input draft data through visual rendering, interactive editing and constraint mapping. An initial material is constructed, multi-view joint generation and local update are performed, double checking and iterative correction are completed, a 2D fine graph is obtained, a high-precision 3D model is finally generated through multi-view generation, geometric reconstruction and parameterized fitting, and the final result is combined with constraint traceable records. The application can stably output a front view / side view 2D design graph meeting rigid engineering constraints based on a traceable, verifiable and iterative engineering delivery chain containing a draft.

Description

Technical Field

[0001] This invention relates to the field of vehicle industrial design technology, and in particular to a traceable mapping 2D-3D integrated design method and apparatus. Background Technology

[0002] The design of the front end of rail transit vehicles is a typical example of a task with a strong coupling between aesthetic styling and engineering constraints. In the conceptual stage, it must not only express style, brand identity, and a sense of technology, but also simultaneously meet a series of rigid engineering constraints, such as: key dimensions of the vehicle's cross-section (width, height, etc.); the relationship between the side view outline boundary and key lengths of the front end; the positions of key components (such as lights, windows, couplers, wipers, and components related to the driver's cab view) and their relative constraints; and geometric consistency between the front and side views (consistent scale and component alignment for the same object). In the engineering process, if the 2D stage cannot achieve "acceptable satisfaction of engineering constraints" and "consistent front / side view..." Figure 1 If the design is flawed, subsequent 3D modeling, structural aerodynamic verification, and engineering design will undergo frequent rework, leading to extended design cycles, increased costs, and enhanced delivery uncertainty. In recent years, the text-to-image / image-to-image capabilities of general generative AI have developed rapidly, and the industry generally adopts a method of prompting words + reference images + partial redrawing to achieve rapid concept drawing and style exploration. However, in industrial design delivery scenarios such as rail transit train heads with strong engineering constraints and multi-view consistency, general solutions generally have structural deficiencies and are unacceptable. Therefore, there is an urgent need in this field for a collaborative solution that can directly connect 2D engineering-constrained multi-view design drawings to "3D editable parametric model delivery" and perform constraint traceability mapping. Summary of the Invention

[0003] To address the technical problems of existing technologies in general generative artificial intelligence used for the design of rail transit vehicle front shapes—namely, the lack of parametric representation oriented towards engineering constraints and a closed-loop mechanism of "automatic verification + reverse correction"—this invention provides a traceable mapping 2D-3D integrated design method and apparatus. The technical solution is as follows:

[0004] On the one hand, a traceable mapping 2D-3D through-design method is provided, which is implemented by a traceable mapping 2D-3D through-design device. The method includes: S1: Construct a project configuration file that includes engineering constraints, and create a parameterized set of engineering constraints based on the engineering constraints. The parameterized set of engineering constraints includes parameterized engineering constraints, a subset of cross-section constraints, a subset of side-view constraints, a subset of component constraints, and a subset of consistency constraints. The initial engineering constraints include cross-section width and height data, track gauge data, side-view key length segment data, and component baseline data. S2: Obtain the preferred reference image set, preliminary line drawing, and user instructions. Based on the parameterized engineering constraint set, perform two-dimensional or three-dimensional visualization rendering. Collect user interactive editing instructions. Through constraint mapping processing, obtain constraint line drawing. Merge the preliminary line drawing to obtain standard input line drawing data. The preliminary line drawing includes outlines, baselines, and coarse positioning of key components. The standard input line drawing data expresses the target component as geometric primitives. The geometric primitives include boxes, key points, or outlines. S3: Based on the user instructions and the standard input line drawing data, construct the initial generated material, and after multi-view joint generation processing, generate a joint multi-view 2D detailed image, and perform local updates to obtain the updated generated material; S4: Perform engineering constraint verification and multi-view consistency verification on the updated generated materials, generate correction instructions, perform iterative engineering constraint verification, and obtain compliant 2D detailed drawings, final verification report and correction log; S5: Based on the compliant 2D detailed map, a parametric 3D model is obtained through multi-view generation processing, geometric reconstruction, and parametric fitting processing. Then, 3D constraint verification and accuracy improvement are performed to obtain a high-precision parametric 3D model. The parametric fitting processing includes extracting feature lines and cross-sectional curves from the mesh, fitting and generating parametric surfaces, and establishing editable parameters. The editable parameters include cross-sectional parameters, curvature control lines, and key feature points. S6: Acquire environmental photos, use an environment fusion rendering algorithm to obtain multi-view renderings, then bind constraint version numbers, and after mapping relationship establishment processing, obtain the final deliverable. The final deliverable includes the high-precision parametric 3D model, the multi-view renderings, and constraint traceability mapping records. The constraint traceability mapping records include the correspondence between constraint IDs and 3D parameters.

[0005] Preferably, the construction of S1 includes a project configuration file for engineering constraints, and a parameterized set of engineering constraints is created based on the engineering constraints. The parameterized set of engineering constraints includes parameterized engineering constraints, a subset of section constraints, a subset of side-view constraints, a subset of component constraints, and a subset of consistency constraints. The initial engineering constraints include section width and height data, track gauge data, side-view key length segment data, and component baseline data, including: S11: The project information and engineering constraints are structured to obtain the project configuration file. The project information includes the project background, project objectives, project scope, project details, project deliverables, and set of engineering constraints. S12: Based on the engineering constraints, after versioning and state activation processing, an initial constraint set is obtained; S13: Using a parametric modeling method, the initial constraint set is parametrically processed to obtain parametric engineering constraints. The parametric processing includes version management and dependency handling. S14: Based on the parameterized engineering constraints, and combined with the preset cross-sectional geometric parameter thresholds, the cross-sectional constraints are limited to obtain a cross-sectional constraint subset, which includes cross-sectional width, cross-sectional height and track gauge. S15: Based on the parameterized engineering constraints and combined with the preset side view contour threshold, the side view constraints are limited to obtain a subset of side view constraints, which includes the vehicle front side view contour boundary and the proportional relationship of key length segments. S16: Based on the parameterized engineering constraints and combined with the preset key component offset range, the component class constraints are limited to obtain a component constraint subset, which includes one or more of the following: headlight position, window position, hook position, wiper position, and their allowable offset range. S17: Based on the parameterized engineering constraints, the consistency class constraints are limited by the view alignment rules to obtain a consistency constraint subset, which includes the shared scale and key point alignment relationship between front view and side view. S18: Merge the parameterized engineering constraints, the section constraint subset, the side view constraint subset, the component constraint subset, and the consistency constraint subset to obtain a parameterized engineering constraint set.

[0006] Preferably, in step S2, the preferred reference drawing set, preliminary line drawing, and user instructions are acquired. Based on the parameterized engineering constraint set, two-dimensional or three-dimensional visualization rendering is performed. User interactive editing instructions are collected, and constraint line drawings are obtained through constraint mapping processing. The preliminary line drawings are then merged to obtain standard input line drawing data. The preliminary line drawings include outlines, baselines, and coarse positioning of key components. The standard input line drawing data expresses the target components as geometric primitives, including frames, key points, or outlines, including: S21: Collect inspiration images through the image import interface to obtain initial reference image data. The collection includes filtering and optimization of the imported inspiration images. S22: Based on the initial reference map data, a preferred reference map set is obtained through labeling and grouping operations; S23: Based on the parameterized set of engineering constraints and the initial reference diagram data, a visualization rendering engine is used to perform two-dimensional or three-dimensional visualization rendering to obtain an orthogonal constraint view, which is used to display the set of engineering constraints in an orthogonal coordinate system. S24: Through the orthogonal constraint view, user instructions are collected and interactive editing instructions are extracted to obtain the edited engineering constraint data. The user's interactive editing instructions include instructions for adjusting the cross-sectional width and height, track gauge, and component baseline. S25: The edited engineering constraint data is updated synchronously across multiple views to obtain an updated constraint set. The synchronous update of multiple views includes the synchronous update of the side views. S26: Extract mapping requirements from user instructions, and select a mapping method using a preset mapping strategy library based on the project configuration file and mapping requirements to obtain a target mapping method. The target mapping method includes at least one of constraint layer method, key point / box annotation method, contour boundary mask method, or multi-channel input method. The multi-channel input method includes multi-channel encoding of the updated constraint set to obtain multi-channel constraint line art. The multi-channel encoding includes independently encoding and integrating the line art channel, constraint channel, and component channel. S27: Based on the updated constraint set and the target mapping method, constraint line art is obtained after constraint mapping processing; S28: Obtain preliminary line art, input the constraint line art and preliminary line art into the line art fusion model, perform data fusion processing, and obtain standard input line art data. The preliminary line art includes outlines, baselines, and coarse positioning of key components. The standard input line art data expresses the target components as geometric primitives, including boxes, key points, or outlines.

[0007] Preferably, step S3 involves constructing initial generated material based on the user instructions and the standard input line drawing data, performing multi-view joint generation processing to generate a jointly generated multi-view 2D detailed image, and then performing local updates to obtain updated generated material, including: S31: Obtain prompt words from the user instructions, extract style, material and detail descriptions from the prompt words, and merge the preferred reference image set, the standard input line art data and the prompt words to obtain the initial generated material; S32: Extract controllable displacement instructions for components from the prompt words, lock the target geometric primitives based on the standard input line art data, perform displacement operations on the target geometric primitives to obtain updated input line art data, and add it to the initial generated material; S33: Extract the generation requirement instruction from the user instruction, and based on the initial generation material, adopt the first generation scheme or the second generation scheme, and obtain the jointly generated multi-view 2D detailed map through multi-view joint generation processing. The first generation scheme is that the front view and the other view images share the same constraint code. When the generation requirement instruction includes key point mapping, the second generation scheme includes first generating the main view image, and then generating other view images with the key points and scale of the main view as constraints. The two views share the constraint version and key point mapping. The jointly generated multi-view 2D detailed map includes the front view and the side view. S34: Extract local editing instructions from the user instructions, and perform local updates on the jointly generated multi-view 2D fine image based on the selection area and local prompts to obtain the updated generated material. The feature of the local update is to maintain the stability of the non-selection area content.

[0008] Preferably, step S4 involves performing engineering constraint verification and multi-view consistency verification on the updated generated materials, generating correction instructions, performing iterative engineering constraint verification, and obtaining a compliant 2D detailed image, a final verification report, and a correction log, including: S41: Using the parameterized set of engineering constraints, perform engineering constraint verification and multi-view consistency verification on the updated generated material to obtain the engineering constraint verification result. The engineering constraint verification includes verification of size, outline, component area and reference height. The multi-view consistency verification includes scale consistency verification, key component alignment consistency verification and outline key point correspondence consistency verification. S42: Based on the engineering constraint verification results, after comprehensive judgment, a verification report is obtained and the repair area is marked on the line drawing. The verification report includes a qualified or unqualified judgment, a list of violations, the deviation amount, and a suggested correction direction. S43: Based on the verification report that is determined to be unqualified, and in conjunction with the repair area, a correction instruction is obtained. The correction instruction includes adjusting the constraint code, generating a correction prompt word, and triggering local regeneration for the unrepaired area. S44: The correction instruction and the updated generated material are verified through iterative engineering constraints. When the number of verification iterations reaches a preset threshold and the new round of verification report is still unqualified, the process is terminated to obtain output information containing the reasons why the requirements cannot be met. S45: Based on the qualified verification report, after final confirmation, a compliant 2D detailed drawing, a final verification report, and a correction log are obtained. The compliant 2D detailed drawing includes a front view and a side view.

[0009] Preferably, in step S5, based on the compliant 2D detailed image, a parametric 3D model is obtained through multi-view generation processing, geometric reconstruction, and parametric fitting processing. Then, 3D constraint verification and accuracy improvement are performed to obtain a high-precision parametric 3D model. The parametric fitting processing includes extracting feature lines and cross-sectional curves from the mesh, fitting and generating parametric surfaces, and establishing editable parameters. The editable parameters include cross-sectional parameters, curvature control lines, and key feature points, including: S51: Based on the compliant 2D detailed image, a multi-view condition input is obtained through multi-view generation processing, wherein the multi-view condition input includes images from multiple different viewpoints; S52: Input the multi-view conditions into the geometric reconstruction model to obtain an initial 3D mesh. The geometric reconstruction model has a basic model prior for achieving zero-sample shape restoration. S53: The initial 3D mesh is refined using a surface refinement algorithm to obtain a refined 3D mesh. The refinement process includes normal-based refinement and local refinement. S54: Based on the refined 3D mesh, a parametric 3D model is obtained through parametric fitting. The parametric fitting process includes extracting feature lines and cross-sectional curves from the mesh, fitting and generating parametric surfaces, and establishing editable parameters, including cross-sectional parameters, curvature control lines, and key feature points. S55: Using the engineering constraint set, perform 3D constraint verification on the parametric 3D model to obtain the verification result. The 3D constraint verification includes size verification and component position verification. S56: Based on the unqualified verification results, after parametric adjustment processing, an updated parametric 3D model is obtained. The parametric adjustment processing includes adjusting the editable parameters or returning to perform surface refinement processing. S57: Input the updated parametric 3D model into the accuracy enhancement module to obtain a high-precision parametric 3D model. The accuracy enhancement module includes an algorithm for obtaining usable shapes based on base film priors and a fine-tuning model based on a small number of samples.

[0010] Preferably, in step S6, acquiring environmental photos employs an environment fusion rendering algorithm to obtain multi-view renderings. Constraint version numbers are then bound, and after mapping relationship establishment, the final deliverable is obtained. The final deliverable includes the high-precision parametric 3D model, the multi-view renderings, and constraint traceability mapping records. The constraint traceability mapping records include the correspondence between constraint IDs and 3D parameters, including: S61: Acquire environmental photos, and use an environmental fusion rendering algorithm to fuse the high-precision parametric 3D model with the environmental photos to obtain a fused environmental review image. The fused environmental review image is used to display the design effect in a real environment. S62: Based on the fusion environment review diagram, after perspective expansion and consistency maintenance processing, a multi-view effect diagram is obtained, which includes other perspective effects generated under the premise of maintaining engineering constraints and perspective consistency. S63: Using a data association module, constraint version numbers are bound to each multi-view rendering to obtain a versioned design drawing. The constraint version number is used to identify the version of the constraint set followed by the versioned design drawing. S64: Using an identifier allocation strategy, a unique constraint ID is assigned to each constraint in the updated constraint set to obtain an identifiable constraint set. The constraint ID is used to uniquely index a single constraint. S65: Based on the identified constraint set, standard input line drawing data, and each multi-view rendering, a constraint traceability mapping record is obtained after mapping relationship establishment processing. The constraint traceability mapping record includes the correspondence between constraint IDs and 3D parameters, including cross-sectional width and height parameters and component position parameters. S66: Merge the high-precision parametric 3D model, the multi-view renderings, and the constraint traceability mapping records to obtain the final deliverable.

[0011] On the other hand, a traceable mapping 2D-3D through-design apparatus is provided, which is applied to the traceable mapping 2D-3D through-design method. The apparatus includes: Engineering Constraints Module: Used to build a project configuration file that includes engineering constraints, and to create a parameterized set of engineering constraints based on the engineering constraints. The parameterized set of engineering constraints includes parameterized engineering constraints, a subset of cross-section constraints, a subset of side-view constraints, a subset of component constraints, and a subset of consistency constraints. The initial engineering constraints include cross-section width and height data, track gauge data, key length segment data for side views, and component baseline data. Line art data module: used to acquire the preferred reference drawing set, preliminary line art and user instructions, perform two-dimensional or three-dimensional visualization rendering based on the parameterized engineering constraint set, collect user interactive editing instructions, obtain constraint line art through constraint mapping processing, and merge the preliminary line art to obtain standard input line art data. The preliminary line art includes outlines, baselines and coarse positioning of key components. The standard input line art data expresses the target components as geometric primitives, including boxes, key points or outlines. Material generation module: Based on the user instructions and the standard input line drawing data, it constructs initial generated materials, performs multi-view joint generation processing to generate a joint multi-view 2D detailed image, and performs local updates to obtain the updated generated materials; Engineering constraint verification module: Used to perform engineering constraint verification and multi-view consistency verification on the updated generated materials, generate correction instructions, perform iterative engineering constraint verification, and obtain compliant 2D detailed map, final verification report and correction log; 3D Model Module: Based on the compliant 2D detailed image, a parametric 3D model is obtained through multi-view generation, geometric reconstruction, and parametric fitting. Then, 3D constraint verification and accuracy improvement are performed to obtain a high-precision parametric 3D model. The parametric fitting process includes extracting feature lines and cross-sectional curves from the mesh, fitting and generating parametric surfaces, and establishing editable parameters, including cross-sectional parameters, curvature control lines, and key feature points. The constraint traceability mapping module is used to acquire environmental photos, use an environmental fusion rendering algorithm to obtain multi-view renderings, bind constraint version numbers, and obtain the final deliverable after establishing a mapping relationship. The final deliverable includes the high-precision parametric 3D model, the multi-view renderings, and constraint traceability mapping records. The constraint traceability mapping records include the correspondence between constraint IDs and 3D parameters.

[0012] On the other hand, a traceable mapping 2D-3D through-design device is provided, the traceable mapping 2D-3D through-design device comprising: a processor; a memory, the memory storing computer-readable instructions, which, when executed by the processor, implement the method described in any of the above-described traceable mapping 2D-3D through-design methods.

[0013] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following: Based on a traceable, verifiable, and iterative engineering delivery chain that includes line drawings, we can stably output front / side view 2D design drawings that meet rigid engineering constraints. This provides core delivery capabilities in the early concept stages of industrial design for rail transit vehicle front ends. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1This is a flowchart of a traceable mapping 2D-3D through-design method provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of a line drawing and a 2D drawing provided in an embodiment of the present invention; Figure 3 This is a block diagram of a traceable mapping 2D-3D through-design device provided in an embodiment of the present invention; Figure 4 This is a structural schematic diagram of a traceable mapping 2D-3D through-design device provided in an embodiment of the present invention. Detailed Implementation

[0016] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0017] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0018] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, their intended meanings are consistent. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, their intended meanings are consistent.

[0019] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0020] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0021] This invention provides a traceable mapping 2D-3D through-design method, which can be implemented by a traceable mapping 2D-3D through-design device, which can be a terminal or a server. Figure 1 The flowchart shown is a traceable mapping 2D-3D through-design method. The processing flow of this method may include the following steps:

[0022] Preferably, a project configuration file including engineering constraints is constructed, and a parameterized set of engineering constraints is created based on the engineering constraints. The parameterized set of engineering constraints includes parameterized engineering constraints, a subset of section constraints, a subset of side-view constraints, a subset of component constraints, and a subset of consistency constraints. The initial engineering constraints include section width and height data, track gauge data, side-view key length segment data, and component baseline data, including: The project information and engineering constraints are structured to obtain a project configuration file. The project information includes project background, project objectives, project scope, project details, project deliverables, and a set of engineering constraints. Based on the aforementioned engineering constraints, an initial constraint set is obtained after versioning and state activation processing; A parametric modeling method is used to parametrically process the initial constraint set to obtain parametric engineering constraints. The parametric processing includes version management and dependency handling. Based on the parameterized engineering constraints, and combined with preset cross-sectional geometric parameter thresholds, the cross-sectional constraints are limited to obtain a cross-sectional constraint subset, which includes cross-sectional width, cross-sectional height, and track gauge. Based on the parameterized engineering constraints, and combined with the preset side view contour threshold, the side view constraints are limited to obtain a subset of side view constraints, which includes the vehicle front side view contour boundary and the proportional relationship of key length segments. Based on the parameterized engineering constraints, and combined with the preset offset range of key components, the component class constraints are limited to obtain a component constraint subset. The component constraint subset includes one or more of the following: headlight position, window position, hook position, wiper position, and allowable offset range. Based on the parameterized engineering constraints, the consistency constraints are limited by the view alignment rules to obtain a consistency constraint subset, which includes the shared scale and key point alignment relationship between front and side views. The parameterized engineering constraints, the section constraint subset, the side view constraint subset, the component constraint subset, and the consistency constraint subset are combined to obtain the parameterized engineering constraint set.

[0023] In some embodiments, parameterized templates transform engineering constraints from text / tables into computable, interactive, and synchronized objects.

[0024] Preferably, a preferred reference drawing set, preliminary line drawings, and user instructions are obtained. Based on the parameterized engineering constraint set, two-dimensional or three-dimensional visualization rendering is performed. User interactive editing instructions are collected, and constraint line drawings are obtained through constraint mapping processing. The preliminary line drawings are then merged to obtain standard input line drawing data. The preliminary line drawings include outlines, baselines, and coarse positioning of key components. The standard input line drawing data expresses the target components as geometric primitives, including frames, key points, or outlines, including: Inspiration images are collected through the image import interface to obtain initial reference image data. The collection includes filtering and optimization of the imported inspiration images. Based on the initial reference map data, a preferred set of reference maps is obtained through labeling and grouping operations; Based on the parameterized set of engineering constraints and the initial reference diagram data, a visualization rendering engine is used to perform two-dimensional or three-dimensional visualization rendering to obtain an orthogonal constraint view, which is used to display the set of engineering constraints in an orthogonal coordinate system. User commands are collected through the orthogonal constraint view, and interactive editing commands are extracted to obtain the edited engineering constraint data. The user's interactive editing commands include commands to adjust the cross-sectional width and height, track gauge, and component baseline. The edited engineering constraint data is then updated synchronously across multiple views to obtain an updated constraint set. The synchronous update of the multiple views includes the synchronous update of the side views. The mapping requirements are extracted from the user instructions. Based on the project configuration file and mapping requirements, a preset mapping strategy library is used to select a mapping method to obtain the target mapping method. The target mapping method includes at least one of the following: constraint layer method, key point / box annotation method, contour boundary mask method, or multi-channel input method. The multi-channel input method includes multi-channel encoding of the updated constraint set to obtain multi-channel constraint line art. The multi-channel encoding includes independently encoding and integrating the line art channel, constraint channel, and component channel. Based on the updated constraint set and the target mapping method, constraint line art is obtained after constraint mapping processing; Obtain the preliminary line drawing, input the constraint line drawing and the preliminary line drawing into the line drawing fusion model, perform data fusion processing, and obtain standard input line drawing data. The preliminary line drawing includes outlines, baselines, and coarse positioning of key components. The standard input line drawing data expresses the target components as geometric primitives, including boxes, key points, or outlines.

[0025] In some embodiments, such as Figure 2As shown, instead of using prompts to generate constraints, the constraints are first applied to the geometric framework, forming a constraint line drawing input X. Common implementations include at least one of the following: constraint layer method: drawing cross-sectional width and height, track gauge, and component baselines as geometric auxiliary lines / frames; key point / frame annotation method: creating position frames or key points for components such as headlights; contour boundary mask method: limiting the allowed area for generating the contour; multi-channel input method: encoding the line drawing channel, constraint channel, and component channel separately before inputting. The effect is that the model sees and adheres to the constraints, rather than requiring humans to guess whether the model follows them.

[0026] It should be noted that the line art uses layered vector graphics (SVG / AI) or pixel art (PNG, transparent background, 256×256 / 512×512 resolution) to ensure that the geometric constraint lines are separated from the background; The basic layers of the line art include the overall outline of the car (mask), the width and height frame of the body section, and the wheelbase baseline; draw the baselines / position frames for the headlights, grille, and wheels; draw the geometric auxiliary lines such as the body centerline, ground baseline, and upper and lower boundary lines of the windows. Constraint lines use high-contrast colors (such as red #FF0000), with a line width of 2px, and no blurring / aliasing to ensure that the geometric features of the model are clearly identifiable.

[0027] It should be further explained that the core framework is Stable Diffusion (SD) + ControlNet. Specifically, ControlNet supports constraint encoding of geometric line drawings, key points, and masks.

[0028] Preferably, OpenCV is used for layered line art processing.

[0029] Preferably, based on the user instructions and the standard input line drawing data, initial generated material is constructed, and after multi-view joint generation processing, a joint generated multi-view 2D detailed image is generated, and local updates are performed to obtain updated generated material, including: The user command is used to obtain prompt words. Style, material and detail descriptions are extracted from the prompt words. The preferred reference image set, the standard input line art data and the prompt words are merged to obtain the initial generated material. Extract controllable displacement instructions for components from the prompt words, lock the target geometric primitives based on the standard input line art data, perform displacement operations on the target geometric primitives to obtain updated input line art data, and add it to the initial generated material; The generation requirement instruction is extracted from the user instruction. Based on the initial generation material, a first generation scheme or a second generation scheme is adopted. After multi-view joint generation processing, a joint generated multi-view 2D detailed map is obtained. The first generation scheme is that the front view and the other view images share the same constraint code. When the generation requirement instruction includes key point mapping, the second generation scheme includes first generating a main view image, and then generating other view images with the key points and scale of the main view as constraints. The two views share the constraint version and key point mapping. The joint generated multi-view 2D detailed map includes a front view and a side view. Local editing instructions are extracted from the user instructions, and the jointly generated multi-view 2D fine image is locally updated based on the selection area and local prompts to obtain the updated generated material. The feature of the local update is to maintain the stability of the non-selected area content.

[0030] In some embodiments, the user instruction example is: "Generate a front view of the 2025 model car with a minimalist tech style, a body material of matte carbon fiber, and the left front headlight in the front view is shifted to the right; generate front / side views with key point mapping alignment; generate a rear window that becomes circular, while the rest of the area remains unchanged."

[0031] Information Extraction: Style / Material / Detail: Minimalist tech style, matte carbon fiber, continuous LED grille; Component displacement instructions: Front view left front headlight → move 10px to the right; Front / Side view + key point mapping; Window → circle, non-selection area stable. Constraint layer: Standard input line art (front / side view geometry / component / contour constraints).

[0032] Based on the coordinate annotations of the standard line art, the reference frame of the left front headlight is located; a displacement operation is performed. Constraint configuration: ControlNet loads line art constraint weights, ensuring the model only adheres to the front view geometry / component constraints; output: a detailed 2D front view conforming to dimensional specifications. Key point mapping: According to industrial design proportions (front / side view projection rules), front view key points are mapped to their corresponding positions in the side view; ControlNet loads dual-branch constraints (line art constraints + key point constraints); the SAM model is used to identify the wheel hub area in the side view, generating a precise wheel hub mask (only the wheel hub is selected, the rest of the area is unselected); ControlNet locks the features of the unselected area (weight 0.95), allowing only the selected area to change. A fusion algorithm is used to ensure a natural transition between the selected and unselected areas, without any stitching artifacts.

[0033] Preferably, the updated generated materials undergo engineering constraint verification and multi-view consistency verification, generating correction instructions, performing iterative engineering constraint verification, and obtaining a compliant 2D detailed drawing, a final verification report, and a correction log, including: Using the parameterized set of engineering constraints, the updated generated material is subjected to engineering constraint verification and multi-view consistency verification to obtain the engineering constraint verification results. The engineering constraint verification includes verification of size, outline, component area and reference height. The multi-view consistency verification includes scale consistency verification, key component alignment consistency verification and outline key point correspondence consistency verification. Based on the engineering constraint verification results, a verification report is obtained after comprehensive judgment, and the repair area is marked on the line drawing. The verification report includes a qualified or unqualified judgment, a list of violations, the deviation amount, and a suggested correction direction. Based on the verification report that is deemed unqualified, and in conjunction with the repair area, a correction instruction is obtained. The correction instruction includes adjusting the constraint code, generating a correction prompt word, and triggering local regeneration for the unrepaired area. The correction instructions and the updated generated materials are verified through iterative engineering constraints. When the number of verification iterations reaches a preset threshold and the new round of verification report is still unqualified, the process is terminated to obtain output information containing the reasons why the requirements cannot be met. Based on the verification report that is deemed qualified, after final confirmation, a compliant 2D detailed drawing, a final verification report, and a correction log are obtained. The compliant 2D detailed drawing includes a front view and a side view.

[0034] In some embodiments, the 2D detailed image is optimized, and the core contour is extracted using Canny edge detection (threshold 100 / 200), preserving the edge lines of the car body, headlights, and windows. The edge image is converted into a black and white binary image (white edges, black background) to eliminate color interference. The generated image is vectorized (pixels → vector geometric primitives). The processed pixel image is converted into a vector image with geometric parameters, the core of which is "contour fitting + parameter annotation". The binary edge image is converted into an SVG vector path, automatically recognizing geometric primitives such as straight lines, arcs, and Bézier curves to obtain a vectorized generated image containing all key geometric parameter annotations. The geometric parameters of the standard constraint line drawing are read, and the origin of the constraint line drawing is aligned with the vectorized generated image for precise vector-level comparison, comparing parameters one by one according to two dimensions: "engineering constraint verification + multi-view consistency verification".

[0035] Preferably, based on the compliant 2D detailed image, a parametric 3D model is obtained through multi-view generation processing, geometric reconstruction, and parametric fitting processing. Then, 3D constraint verification and accuracy improvement are performed to obtain a high-precision parametric 3D model. The parametric fitting processing includes extracting feature lines and cross-sectional curves from the mesh, fitting and generating parametric surfaces, and establishing editable parameters. The editable parameters include cross-sectional parameters, curvature control lines, and key feature points, including: Based on the compliant 2D detailed image, a multi-view conditional input is obtained through multi-view generation processing, which includes images from multiple different viewpoints. The multi-view conditions are input into the geometric reconstruction model to obtain an initial 3D mesh. The geometric reconstruction model has a basic model prior for achieving zero-sample shape restoration. A surface refinement algorithm is used to refine the initial 3D mesh to obtain a refined 3D mesh. The refinement process includes normal-based refinement and local refinement. Based on the refined 3D mesh, a parametric 3D model is obtained through parametric fitting. The parametric fitting process includes extracting feature lines and cross-sectional curves from the mesh, fitting and generating parametric surfaces, and establishing editable parameters, including cross-sectional parameters, curvature control lines, and key feature points. The parameterized 3D model is subjected to 3D constraint verification using the engineering constraint set to obtain the verification result. The 3D constraint verification includes size verification and component position verification. Based on the unqualified verification results, an updated parametric 3D model is obtained through parametric adjustment processing. The parametric adjustment processing includes adjusting the editable parameters or returning to perform surface refinement processing. The updated parametric 3D model is input into the accuracy enhancement module to obtain a high-precision parametric 3D model. The accuracy enhancement module includes an algorithm for obtaining usable shapes based on base film priors and a fine-tuning model based on a small number of samples.

[0036] In some embodiments, Instant Neural Graphics Primitives (Instant-NGP) are used to implement multi-view image-to-3D scene / object reconstruction models. Alternatively, OpenMVS or single-image-to-3D shape models can be used to generate 3D meshes.

[0037] It should be noted that Poisson surface reconstruction is used to reconstruct and refine the 3D mesh from the point cloud. High-precision mesh refinement can be achieved using tools such as Open3D and PCL. Based on algorithms such as Laplacian smoothing and normal smoothing provided by the Open3D open-source library, the 3D mesh can be smoothed and optimized, and relevant parameters (such as the number of iterations and constraint angles) can be adjusted according to engineering requirements.

[0038] Preferably, environmental photos are acquired, and an environmental fusion rendering algorithm is used to obtain multi-view renderings. Then, constraint version numbers are bound, and after mapping relationship establishment processing, the final deliverable is obtained. The final deliverable includes the high-precision parametric 3D model, the multi-view renderings, and constraint traceability mapping records. The constraint traceability mapping records include the correspondence between constraint IDs and 3D parameters, including: An environmental photograph is acquired, and an environmental fusion rendering algorithm is used to fuse the high-precision parametric 3D model with the environmental photograph to obtain a fused environmental review image. The fused environmental review image is used to display the design effect in a real environment. Based on the fusion environment review diagram, after perspective expansion and consistency maintenance processing, a multi-view effect diagram is obtained. The multi-view effect diagram includes other perspective effects generated under the premise of maintaining engineering constraints and perspective consistency. A data association module is used to bind constraint version numbers to each multi-view rendering to obtain a versioned design drawing. The constraint version number is used to identify the version of the constraint set followed by the versioned design drawing. An identifier allocation strategy is adopted to assign a unique constraint ID to each constraint in the updated constraint set, resulting in an identifiable constraint set. The constraint ID is used to uniquely index a single constraint. Based on the identified constraint set, standard input line drawing data, and each multi-view rendering, a constraint traceability mapping record is obtained after mapping relationship establishment. The constraint traceability mapping record includes the correspondence between constraint IDs and 3D parameters, including cross-sectional width and height parameters and component position parameters. The high-precision parametric 3D model, the multi-view renderings, and the constraint traceability mapping records are merged to obtain the final deliverable.

[0039] In some embodiments, the environment fusion rendering employs a camera pose estimation model and a physically based rendering model; the lighting estimation employs an ambient light field model.

[0040] The above is an introduction to the method embodiments. The following describes the solution described in this application through device embodiments.

[0041] Figure 3 This is a block diagram illustrating a traceable mapping 2D-3D through-design apparatus according to an exemplary embodiment, the apparatus being used in a traceable mapping 2D-3D through-design method. (Refer to...) Figure 3 The device includes an engineering constraint module 301, a line drawing data module 302, a material generation module 303, an engineering constraint verification module 304, a 3D model module 305, and a constraint traceability mapping module 306.

[0042] Engineering constraint module 301: used to construct a project configuration file including engineering constraints, and to create a parameterized set of engineering constraints based on the engineering constraints. The parameterized set of engineering constraints includes parameterized engineering constraints, a subset of cross-section constraints, a subset of side-view constraints, a subset of component constraints, and a subset of consistency constraints. The initial engineering constraints include cross-section width and height data, track gauge data, side-view key length segment data, and component baseline data. Line drawing data module 302: used to acquire a set of preferred reference drawings, preliminary line drawings and user instructions, perform two-dimensional or three-dimensional visualization rendering based on the parameterized set of engineering constraints, collect user interactive editing instructions, obtain constraint line drawings through constraint mapping processing, and merge the preliminary line drawings to obtain standard input line drawing data. The preliminary line drawings include outlines, baselines and coarse positioning of key components. The standard input line drawing data expresses the target components as geometric primitives, including frames, key points or outlines. Material generation module 303: Based on the user instructions and the standard input line drawing data, it constructs initial generated materials, generates a joint multi-view 2D detailed image through multi-view joint generation processing, and performs local updates to obtain updated generated materials; Engineering constraint verification module 304: Used to perform engineering constraint verification and multi-view consistency verification on the updated generated materials, generate correction instructions, perform iterative engineering constraint verification, and obtain compliant 2D detailed drawings, final verification reports and correction logs; 3D Model Module 305: Based on the compliant 2D detailed image, it generates a parametric 3D model through multi-view generation, geometric reconstruction, and parametric fitting, and then performs 3D constraint verification and accuracy improvement to obtain a high-precision parametric 3D model. The parametric fitting process includes extracting feature lines and cross-sectional curves from the mesh, fitting and generating parametric surfaces, and establishing editable parameters, including cross-sectional parameters, curvature control lines, and key feature points. Constraint tracing mapping module 306: used to acquire environmental photos, use environmental fusion rendering algorithm to obtain multi-view effect images, then bind constraint version numbers, and after mapping relationship establishment processing, obtain the final delivery results. The final delivery results include the high-precision parametric 3D model, the multi-view effect images and constraint tracing mapping records. The constraint tracing mapping records include the correspondence between constraint IDs and 3D parameters.

[0043] A traceable mapping 2D-3D through-design device, the traceable mapping 2D-3D through-design device comprising: a processor; a memory storing computer-readable instructions, wherein when the computer-readable instructions are executed by the processor, the method described in any one of the above-described traceable mapping 2D-3D through-design methods is implemented.

[0044] Figure 4 This is a structural schematic diagram of a traceable mapping 2D-3D through-design device provided in an embodiment of the present invention, as shown below. Figure 4 As shown, the traceable mapping 2D-3D through-design equipment may include the above-mentioned Figure 3 The illustrated traceable mapping 2D-3D through-design device 410 may optionally include a first processor 2001.

[0045] Optionally, the traceable mapping 2D-3D through-design device 410 may also include a memory 2002 and a transceiver 2003.

[0046] The first processor 2001, memory 2002, and transceiver 2003 can be connected via a communication bus.

[0047] The following is combined with Figure 4 A detailed description of each component of the traceable mapping 2D-3D through-design device 410 is provided below: The first processor 2001 is the control center of the traceable mapping 2D-3D through-design device 410. It can be a single processor or a collective term for multiple processing elements. For example, the first processor 2001 can be one or more central processing units (CPUs), application-specific integrated circuits (ASICs), or one or more integrated circuits configured to implement embodiments of the present invention, such as one or more digital signal processors (DSPs), or one or more field-programmable gate arrays (FPGAs).

[0048] Optionally, the first processor 2001 can perform various functions of the traceable mapping 2D-3D through-design device 410 by running or executing software programs stored in memory 2002 and calling data stored in memory 2002.

[0049] In a specific implementation, as one example, the first processor 2001 may include one or more CPUs, for example... Figure 4 CPU0 and CPU1 are shown in the diagram.

[0050] In a specific implementation, as one example, the traceable mapping 2D-3D through-design device 410 may also include multiple processors, for example... Figure 4The first processor 2001 and the second processor 2004 are shown in the diagram. Each of these processors can be a single-core processor or a multi-core processor. Here, a processor can refer to one or more devices, circuits, and / or processing cores used to process data (such as computer program instructions).

[0051] The memory 2002 is used to store the software program that executes the present invention, and is controlled by the first processor 2001 to execute it. The specific implementation method can be referred to the above method embodiment, and will not be repeated here.

[0052] Optionally, the memory 2002 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. The memory 2002 may be integrated with the first processor 2001 or may exist independently, and may be connected via the interface circuit of the traceable 2D-3D through-design device 410. Figure 4 (Not shown in the image) is coupled to the first processor 2001, and this embodiment of the invention does not specifically limit this.

[0053] The transceiver 2003 is used to communicate with network devices or with terminal devices.

[0054] Alternatively, transceiver 2003 may include a receiver and a transmitter. Figure 4 (Not shown separately). The receiver is used to implement the receiving function, and the transmitter is used to implement the transmitting function.

[0055] Optionally, the transceiver 2003 can be integrated with the first processor 2001 or exist independently, and can be connected to the interface circuit of the traceable 2D-3D through-design device 410. Figure 4 (Not shown in the image) is coupled to the first processor 2001, and this embodiment of the invention does not specifically limit this.

[0056] It should be noted that, Figure 4 The structure of the traceable mapping 2D-3D through-design device 410 shown does not constitute a limitation on the router. The actual knowledge structure identification device may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0057] Furthermore, the technical effects of the traceable mapping 2D-3D through-design device 410 can be referred to the technical effects of the traceable mapping 2D-3D through-design method described in the above method embodiments, and will not be repeated here.

[0058] It should be understood that the first processor 2001 in the embodiments of the present invention may be a central processing unit (CPU), or it may be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc.

[0059] It should also be understood that the memory in the embodiments of the present invention can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate synchronous DRAM (DDR SDRAM), enhanced synchronous DRAM (ESDRAM), synchronous linked DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0060] The above embodiments can be implemented, in whole or in part, by software, hardware (such as circuits), firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of the present invention are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. A semiconductor medium can be a solid-state drive.

[0061] It should be understood that the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. A and B can be singular or plural. Additionally, the character " / " in this article generally indicates an "or" relationship between the preceding and following related objects, but it can also represent an "and / or" relationship. Please refer to the context for a more accurate understanding.

[0062] In this invention, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of a single item or a plurality of items. For example, at least one of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, and c can be a single item or multiple items.

[0063] It should be understood that, in various embodiments of the present invention, the order of the above-mentioned process numbers does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

[0064] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0065] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the devices, apparatuses, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0066] In the several embodiments provided by this invention, it should be understood that the disclosed devices, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0067] 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.

[0068] In addition, the functional units in the various embodiments of the present invention 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.

[0069] 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, or the part that contributes to the prior art, or a part 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, server, 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.

[0070] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A traceable mapping 2D-3D through-design method, characterized in that, The method includes: S1: Construct a project configuration file that includes engineering constraints, and create a parameterized set of engineering constraints based on the engineering constraints. The parameterized set of engineering constraints includes parameterized engineering constraints, a subset of cross-section constraints, a subset of side-view constraints, a subset of component constraints, and a subset of consistency constraints. The initial engineering constraints include cross-section width and height data, track gauge data, side-view key length segment data, and component baseline data. S2: Obtain the preferred reference image set, preliminary line drawing, and user instructions. Based on the parameterized engineering constraint set, perform two-dimensional or three-dimensional visualization rendering. Collect user interactive editing instructions. Through constraint mapping processing, obtain constraint line drawing. Merge the preliminary line drawing to obtain standard input line drawing data. The preliminary line drawing includes outlines, baselines, and coarse positioning of key components. The standard input line drawing data expresses the target component as geometric primitives. The geometric primitives include boxes, key points, or outlines. S3: Based on the user instructions and the standard input line drawing data, construct the initial generated material, and after multi-view joint generation processing, generate a joint multi-view 2D detailed image, and perform local updates to obtain the updated generated material; S4: Perform engineering constraint verification and multi-view consistency verification on the updated generated materials, generate correction instructions, perform iterative engineering constraint verification, and obtain compliant 2D detailed drawings, final verification report and correction log; S5: Based on the compliant 2D detailed map, a parametric 3D model is obtained through multi-view generation processing, geometric reconstruction, and parametric fitting processing. Then, 3D constraint verification and accuracy improvement are performed to obtain a high-precision parametric 3D model. The parametric fitting processing includes extracting feature lines and cross-sectional curves from the mesh, fitting and generating parametric surfaces, and establishing editable parameters. The editable parameters include cross-sectional parameters, curvature control lines, and key feature points. S6: Acquire environmental photos, use an environmental fusion rendering algorithm to obtain multi-view renderings, then bind constraint version numbers, and after mapping relationship establishment processing, obtain the final deliverable. The final deliverable includes the high-precision parametric 3D model, the multi-view renderings, and constraint traceability mapping records. The constraint traceability mapping records include the correspondence between constraint IDs and 3D parameters.

2. The traceable mapping 2D-3D through-design method according to claim 1, characterized in that, The construction of S1 includes a project configuration file for engineering constraints, and the creation of a parameterized set of engineering constraints based on these constraints. This parameterized set of engineering constraints includes parameterized engineering constraints, a subset of section constraints, a subset of side-view constraints, a subset of component constraints, and a subset of consistency constraints. The initial engineering constraints include section width and height data, track gauge data, side-view critical length segment data, and component baseline data, including: S11: The project information and engineering constraints are structured to obtain a project configuration file. The project information includes project background, project objectives, project scope, project details, project deliverables, and a set of engineering constraints. S12: Based on the engineering constraints, after versioning and state activation processing, an initial constraint set is obtained; S13: Using a parametric modeling method, the initial constraint set is parametrically processed to obtain parametric engineering constraints. The parametric processing includes version management and dependency handling. S14: Based on the parameterized engineering constraints, and combined with the preset cross-sectional geometric parameter thresholds, the cross-sectional constraints are limited to obtain a cross-sectional constraint subset, which includes cross-sectional width, cross-sectional height and track gauge. S15: Based on the parameterized engineering constraints and combined with the preset side view contour threshold, the side view constraints are limited to obtain a subset of side view constraints. The subset of side view constraints includes the boundary of the front side view contour and the proportional relationship of key length segments. S16: Based on the parameterized engineering constraints and combined with the preset key component offset range, the component class constraints are limited to obtain a component constraint subset. The component constraint subset includes one or more of the following: headlight position, window position, hook position, wiper position, and their allowable offset range. S17: Based on the parameterized engineering constraints, the view alignment rules are used to limit the consistency constraints to obtain a subset of consistency constraints. The subset of consistency constraints includes the shared scale and key point alignment relationship between front and side views. S18: Merge the parameterized engineering constraints, the section constraint subset, the side view constraint subset, the component constraint subset, and the consistency constraint subset to obtain a parameterized engineering constraint set.

3. The traceable mapping 2D-3D through-design method according to claim 1, characterized in that, The S2 process involves acquiring a preferred reference image set, preliminary line drawings, and user instructions. Based on the parameterized engineering constraint set, it performs two-dimensional or three-dimensional visualization rendering, collects user interactive editing instructions, and obtains constraint line drawings through constraint mapping. These are then merged with the preliminary line drawings to obtain standard input line drawing data. The preliminary line drawings include outlines, baselines, and coarse positioning of key components. The standard input line drawing data represents the target components as geometric primitives, including frames, key points, or outlines. S21: Collect inspiration images through the image import interface to obtain initial reference image data. The collection includes filtering and optimization of the imported inspiration images. S22: Based on the initial reference map data, a preferred reference map set is obtained through labeling and grouping operations; S23: Based on the parameterized set of engineering constraints and the initial reference diagram data, a visualization rendering engine is used to perform two-dimensional or three-dimensional visualization rendering to obtain an orthogonal constraint view, which is used to display the set of engineering constraints in an orthogonal coordinate system. S24: Through the orthogonal constraint view, user instructions are collected and interactive editing instructions are extracted to obtain the edited engineering constraint data. The user's interactive editing instructions include instructions for adjusting the cross-sectional width and height, track gauge, and component baseline. S25: The edited engineering constraint data is updated synchronously across multiple views to obtain an updated constraint set. The synchronous update of multiple views includes the synchronous update of the side views. S26: Extract mapping requirements from user instructions, and select a mapping method using a preset mapping strategy library based on the project configuration file and mapping requirements to obtain a target mapping method. The target mapping method includes at least one of constraint layer method, key point / box annotation method, contour boundary mask method, or multi-channel input method. The multi-channel input method includes multi-channel encoding of the updated constraint set to obtain multi-channel constraint line art. The multi-channel encoding includes independently encoding and integrating the line art channel, constraint channel, and component channel. S27: Based on the updated constraint set and the target mapping method, constraint line art is obtained after constraint mapping processing; S28: Obtain preliminary line art, input the constraint line art and preliminary line art into the line art fusion model, perform data fusion processing, and obtain standard input line art data. The preliminary line art includes outlines, baselines, and coarse positioning of key components. The standard input line art data expresses the target components as geometric primitives, including boxes, key points, or outlines.

4. The traceable mapping 2D-3D through-design method according to claim 1, characterized in that, S3, based on the user instructions and the standard input line drawing data, constructs initial generated material, performs multi-view joint generation processing to generate a joint multi-view 2D detailed image, and performs local updates to obtain updated generated material, including: S31: Obtain prompt words from the user instructions, extract style, material and detail descriptions from the prompt words, and merge the preferred reference image set, the standard input line art data and the prompt words to obtain the initial generated material; S32: Extract controllable displacement instructions for components from the prompt words, lock the target geometric primitives based on the standard input line art data, perform displacement operations on the target geometric primitives to obtain updated input line art data, and add it to the initial generated material; S33: Extract the generation requirement instruction from the user instruction, and based on the initial generation material, adopt the first generation scheme or the second generation scheme, and after multi-view joint generation processing, obtain a jointly generated multi-view 2D detailed map. The first generation scheme is that the front view and other view images share the same constraint code. When the generation requirement instruction includes key point mapping, the second generation scheme includes first generating the main view image, and then using the key points and scale of the main view as constraints to generate other view images. The two views share the constraint version and key point mapping. The jointly generated multi-view 2D detailed map includes the front view and the side view. S34: Extract local editing instructions from the user instructions, and perform local updates on the jointly generated multi-view 2D fine image based on the selection area and local prompts to obtain the updated generated material. The feature of the local update is to maintain the stability of the non-selection area content.

5. The traceable mapping 2D-3D through-design method according to claim 1, characterized in that, S4 performs engineering constraint verification and multi-view consistency verification on the updated generated materials, generates correction instructions, performs iterative engineering constraint verification, and obtains a compliant 2D detailed image, a final verification report, and a correction log, including: S41: Using the parameterized set of engineering constraints, perform engineering constraint verification and multi-view consistency verification on the updated generated material to obtain the engineering constraint verification result. The engineering constraint verification includes verification of size, outline, component area and reference height. The multi-view consistency verification includes scale consistency verification, key component alignment consistency verification and outline key point correspondence consistency verification. S42: Based on the engineering constraint verification results, after comprehensive judgment, a verification report is obtained and the repair area is marked on the line drawing. The verification report includes a qualified or unqualified judgment, a list of violations, the deviation amount, and a suggested correction direction. S43: Based on the verification report that is determined to be unqualified, and in conjunction with the repair area, a correction instruction is obtained. The correction instruction includes adjusting the constraint code, generating a correction prompt word, and triggering local regeneration for the unrepaired area. S44: The correction instruction and the updated generated material are verified through iterative engineering constraints. When the number of verification iterations reaches a preset threshold and the new round of verification report is still unqualified, the process is terminated to obtain output information containing the reasons why the requirements cannot be met. S45: Based on the qualified verification report, after final confirmation, a compliant 2D detailed drawing, a final verification report, and a correction log are obtained. The compliant 2D detailed drawing includes a front view and a side view.

6. The traceable mapping 2D-3D through-design method according to claim 1, characterized in that, S5, based on the compliant 2D detailed image, undergoes multi-view generation processing, geometric reconstruction, and parametric fitting processing to obtain a parametric 3D model. This is followed by 3D constraint verification and accuracy enhancement to obtain a high-precision parametric 3D model. The parametric fitting process includes extracting feature lines and cross-sectional curves from the mesh, fitting and generating parametric surfaces, and establishing editable parameters. These editable parameters include cross-sectional parameters, curvature control lines, and key feature points, including: S51: Based on the compliant 2D detailed image, a multi-view condition input is obtained through multi-view generation processing, wherein the multi-view condition input includes images from multiple different viewpoints; S52: Input the multi-view conditions into the geometric reconstruction model to obtain an initial 3D mesh. The geometric reconstruction model has a basic model prior for achieving zero-sample shape restoration. S53: The initial 3D mesh is refined using a surface refinement algorithm to obtain a refined 3D mesh. The refinement process includes normal-based refinement and local refinement. S54: Based on the refined 3D mesh, a parametric 3D model is obtained through parametric fitting. The parametric fitting process includes extracting feature lines and cross-sectional curves from the mesh, fitting and generating parametric surfaces, and establishing editable parameters, including cross-sectional parameters, curvature control lines, and key feature points. S55: Using the engineering constraint set, perform 3D constraint verification on the parametric 3D model to obtain the verification result. The 3D constraint verification includes size verification and component position verification. S56: Based on the unqualified verification results, after parametric adjustment processing, an updated parametric 3D model is obtained. The parametric adjustment processing includes adjusting the editable parameters or returning to perform surface refinement processing. S57: Input the updated parametric 3D model into the accuracy enhancement module to obtain a high-precision parametric 3D model. The accuracy enhancement module includes an algorithm for obtaining usable shapes based on base film priors and a fine-tuning model based on a small number of samples.

7. The traceable mapping 2D-3D through-design method according to claim 1, characterized in that, The S6 process acquires environmental photos using an environment fusion rendering algorithm to obtain multi-view renderings. Constraint version numbers are then bound, and a mapping relationship is established to obtain the final deliverable. The final deliverable includes the high-precision parametric 3D model, the multi-view renderings, and a constraint traceability mapping record. The constraint traceability mapping record includes the correspondence between constraint IDs and 3D parameters, including: S61: Acquire environmental photos, and use an environmental fusion rendering algorithm to fuse the high-precision parametric 3D model with the environmental photos to obtain a fused environmental review image. The fused environmental review image is used to display the design effect in a real environment. S62: Based on the fusion environment review diagram, after perspective expansion and consistency maintenance processing, a multi-view effect diagram is obtained, which includes other perspective effects generated under the premise of maintaining engineering constraints and perspective consistency. S63: Using a data association module, constraint version numbers are bound to each multi-view rendering to obtain a versioned design drawing. The constraint version number is used to identify the version of the constraint set followed by the versioned design drawing. S64: Using an identifier allocation strategy, assign a unique constraint ID to each constraint in the updated constraint set to obtain an identifiable constraint set. The constraint ID is used to uniquely index a single constraint. S65: Based on the identified constraint set, standard input line drawing data, and each multi-view rendering, a constraint traceability mapping record is obtained after mapping relationship establishment processing. The constraint traceability mapping record includes the correspondence between constraint IDs and 3D parameters, including cross-sectional width and height parameters and component position parameters. S66: Merge the high-precision parametric 3D model, the multi-view renderings, and the constraint traceability mapping records to obtain the final deliverable.

8. A traceable mapping 2D-3D through-design apparatus, wherein the traceable mapping 2D-3D through-design apparatus is used to implement the traceable mapping 2D-3D through-design method as described in any one of claims 1-7, characterized in that, The device includes: Engineering Constraints Module: Used to build a project configuration file that includes engineering constraints, and to create a parameterized set of engineering constraints based on the engineering constraints. The parameterized set of engineering constraints includes parameterized engineering constraints, a subset of cross-section constraints, a subset of side-view constraints, a subset of component constraints, and a subset of consistency constraints. The initial engineering constraints include cross-section width and height data, track gauge data, key length segment data for side views, and component baseline data. Line art data module: used to acquire the preferred reference drawing set, preliminary line art and user instructions, perform two-dimensional or three-dimensional visualization rendering based on the parameterized engineering constraint set, collect user interactive editing instructions, obtain constraint line art through constraint mapping processing, and merge the preliminary line art to obtain standard input line art data. The preliminary line art includes outlines, baselines and coarse positioning of key components. The standard input line art data expresses the target components as geometric primitives, including boxes, key points or outlines. Material generation module: Based on the user instructions and the standard input line drawing data, it constructs initial generated materials, performs multi-view joint generation processing to generate a joint multi-view 2D detailed image, and performs local updates to obtain the updated generated materials; Engineering constraint verification module: Used to perform engineering constraint verification and multi-view consistency verification on the updated generated materials, generate correction instructions, perform iterative engineering constraint verification, and obtain compliant 2D detailed map, final verification report and correction log; 3D Model Module: Based on the compliant 2D detailed image, a parametric 3D model is obtained through multi-view generation, geometric reconstruction, and parametric fitting. Then, 3D constraint verification and accuracy improvement are performed to obtain a high-precision parametric 3D model. The parametric fitting process includes extracting feature lines and cross-sectional curves from the mesh, fitting and generating parametric surfaces, and establishing editable parameters, including cross-sectional parameters, curvature control lines, and key feature points. The constraint traceability mapping module is used to acquire environmental photos, use an environmental fusion rendering algorithm to obtain multi-view renderings, bind constraint version numbers, and obtain the final deliverable after establishing a mapping relationship. The final deliverable includes the high-precision parametric 3D model, the multi-view renderings, and constraint traceability mapping records. The constraint traceability mapping records include the correspondence between constraint IDs and 3D parameters.

9. A traceable mapping 2D-3D through-design device, characterized in that, The traceable mapping 2D-3D through-design processor; a memory storing computer-readable instructions that, when executed by the processor, implement the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium contains program code that can be invoked by a processor to execute the method as described in any one of claims 1 to 7.

Images