parametric feature modeling

This work addresses the geometrically driven parametric CAD editing challenge: simultaneously and structurally preserving updates to a CAD model’s underlying parametric construction sequence during shape editing—while ensuring structural consistency, semantic validity, and high shape fidelity—particularly under scarce triplet supervision. We propose a “Plan-Generate-Verify” closed-loop framework: (1) zero-shot editing via a parameter-to-shape (P2S) latent diffusion model coupled with masked parameter prediction (MPP); (2) cross-attention mechanisms to localize edit regions and inject semantic context; and (3) automatic verification using shape latent-space distance metrics. Our method requires no manually annotated triplets and significantly outperforms GPT-4o and specialized CAD baselines. It achieves comprehensive success across structural preservation, semantic plausibility, and shape fidelity, enabling high-fidelity iterative editing and enhanced reverse engineering capabilities.

Existing CAD generation datasets lack the capability to evaluate design intent preservation—specifically, constraint consistency—after parametric edits. This work proposes HistCAD, a CAD-software-agnostic intermediate language that explicitly encodes sketch primitives, geometric and dimensional constraints, feature operations, and 3D boundary references within sequential representations. Leveraging this representation, we construct a dataset of 170,000 executable sequences and establish a new benchmark focused on editability. We introduce three metrics—Edit Reachability (ER), constraint-preserving command sequence ratio (cPCSR), and operation edit similarity (OES)—to distinguish between edit accessibility and constraint preservation. Experiments demonstrate that explicit constraint modeling is crucial for maintaining design intent. HistCAD enables both text-supervised generation and end-to-end CAD synthesis with large language models, validated on industrial-scale complex models, thereby advancing CAD generation from static shape imitation toward reusable, parameterized sequence synthesis.

To address the lack of flexibility and controllability in 3D human hair modeling and editing, this paper proposes Perm—a learnable parametric representation for multi-style hairstyles. Our method tackles three core challenges: (1) We introduce a novel strand representation in the frequency domain, coupled with frequency-domain PCA decomposition, to disentangle global hair structure from local curl patterns; (2) We design a geometric decomposition mechanism that separates guide texture (structural backbone) from residual texture (fine-scale details), enabling hierarchical modeling of geometry; (3) We formulate hair styling as a layered generative process. Evaluated on single-view 3D hair reconstruction, interactive editing, and hairstyle-conditional image generation, Perm achieves state-of-the-art performance across all tasks. It demonstrates strong generalization capability and seamless cross-task deployment, establishing a unified, controllable framework for diverse hair-related applications.

Existing CAD deep learning methods neglect geometric constraint modeling, leading to poor discrimination between shapes with similar appearances but distinct constraints. To address this, we propose CstNet—a two-stage constraint-aware network for parametric point cloud analysis—that explicitly encodes CAD constraints as learnable ternary vectors and supports both B-Rep and point cloud inputs. Our contributions are threefold: (1) the first vectorized representation of CAD constraints and an end-to-end constraint learning paradigm; (2) Param20K—the first large-scale multimodal dataset of parametric point clouds, comprising 20,000 samples across 75 classes; and (3) state-of-the-art performance on Param20K, achieving a 3.52% absolute improvement in classification accuracy and a 26.17% gain in rotational robustness over prior methods.

Existing CAD systems suffer from a fundamental disconnect between feature-based parametric modeling and B-rep–based direct modeling, hindering cross-paradigm collaborative editing of geometry, topology, and parametric constraints. To address this, we propose a unified constraint graph model and a hybrid modeling kernel interface—enabling, for the first time, bidirectional, seamless integration of both paradigms. Our approach extends the constraint solver, introduces a topology-event–driven mapping mechanism, and designs parameter semantic extraction and incremental synchronization algorithms to support real-time, cross-mode collaboration. Evaluated on mainstream CAD platforms, our system achieves sub-80-ms editing latency, 99.2% constraint fidelity, and efficient handling of complex assemblies. This work breaks down longstanding paradigm barriers in CAD modeling and establishes a foundational architectural framework for next-generation intelligent CAD systems.

Existing CAD generation methods prioritize visual similarity over geometric precision and suffer from quantization-induced errors in parametric representations, rendering them inadequate for industrial applications demanding exact dimensional accuracy. To address this, this work proposes a Plan-Then-Construct paradigm: first generating a structured design plan with explicit continuous dimensional parameters, then employing a pointer mechanism during the construction phase to directly reference these parameters, thereby enabling high-fidelity modeling. This approach achieves the first end-to-end prediction of continuous parameters, eliminating quantization errors, and ensures dimensional consistency by decoupling parameter inference from geometric construction. We introduce the first large-scale CAD dataset annotated with design plans and propose a three-level geometric evaluation metric—vertices, edges, and faces. Experiments demonstrate that our method significantly outperforms existing approaches across all geometric levels, making it suitable for precision-sensitive engineering tasks.

Existing B-Rep generation methods rely on non-native representations such as point clouds or meshes, which discard the semantic information of parametric surfaces and thereby limit geometric accuracy and downstream usability. This work proposes ParaCAD, an autoregressive framework that, for the first time, enables native parametric B-Rep generation conditioned on input point clouds. By employing a surface-oriented tokenization strategy, ParaCAD explicitly encodes surface types and continuous parameters, generates parametric surfaces within constrained UV domains, and constructs valid B-Reps through global intersection computation. The method fully preserves CAD geometric semantics and substantially improves geometric accuracy, robustness, watertightness, and fidelity to the input point cloud, outperforming all existing point-cloud-based baselines across the board.

This work addresses the limitation of existing CAD model evaluation methods, which predominantly emphasize visual fidelity while neglecting engineering functionality. To bridge this gap, the authors propose CADEngBench, a dual-track benchmark that systematically incorporates engineering behavior validation—including finite element analysis (FEA) alignment, design-for-manufacturing (DFM) checks, and kinematic joint dynamics—into the assessment framework, covering both parametric parts and assemblies. The benchmark employs techniques such as B-Rep validity verification, parameter perturbation tests, functional editing tasks, and linear static simulations using CalculiX to comprehensively evaluate the engineering-grade capabilities of generated and edited models. Experimental results reveal that while current multimodal models outperform in CAD editing over generation, they still struggle with complex edits, FEA consistency, and accurately reconstructing real-world assembly mating relationships.

Existing CAD generation methods struggle to simultaneously preserve modeling history, topological reference stability, and feature-level editability in cross-platform scenarios. This work proposes CADIR—an agent-oriented, executable intermediate representation that explicitly constructs a procedural graph encompassing operation sequences, parameter dependencies, constraints, and topological selections based on the OpenCASCADE (OCCT) geometric kernel. To enable faithful cross-platform model reconstruction, CADIR introduces a geometric signature matching mechanism. It is the first approach to support explicit procedural graph representations that allow editing across heterogeneous CAD backends. By integrating text- or image-driven procedural graph retrieval, CADIR demonstrates high-fidelity, editable reuse of complete models and substructures across FreeCAD, SolidWorks, and Fusion 360, enabling seamless subsequent modifications.