Parametric cad models created from sketches and descriptions
Rethinking Human Intent-to-CAD: Parametric CAD Model Generation via Cooperative Multi-Task Alignment and Spatial-Aware Reinforcement Learning
Computer Vision and Pattern RecognitionArtificial Intelligence
Summary
Designing 3D models using computers can be hard, especially when starting with rough ideas like sketches or text. The authors created a big dataset combining sketches, descriptions, and CAD codes to help computers learn these connections. They built a two-step method that first aligns different input types and then uses reinforcement learning to make sure the models are geometrically correct. Their approach works better than previous methods at turning human ideas into detailed CAD models.
What this means in practice
- •For industrial designers: Generate detailed parametric CAD models directly from early sketches and textual ideas to accelerate conceptual design phases.
- •For 3d modeling software developers: Integrate multi-modal intent interpretation into CAD platforms to improve user experience by converting sketches and text into editable models.
Authors
Qingwang Zhang, Jiahao Li, Xiangdong Zhou
Abstract
Parametric Computer-Aided-Design (CAD) modeling from human intent remains challenging, particularly during the conceptual design stage, where design goals are expressed through incomplete and unstructured modalities (e.g., hand-drawn sketches and textual descriptions). In this work, we rethink the human intent-to-CAD pipeline and propose a unified method that directly maps multi-level human intents to executable codes, without assuming the prior existence of target CAD models. To support our study, we construct HiCAD, the first large-scale dataset aligning hand-drawn sketches, textual descriptions, and parametric CAD codes. Based on this, we introduce HiCAD, a two-stage framework comprising Cooperative Multi-Task Alignment to bridge the representational gap between heterogeneous inputs, and Spatial-Aware Reinforcement Learning to enforce geometric and topological consistency. Extensive experiments demonstrate that our method significantly outperforms existing baselines across multiple tasks, validating its effectiveness and robustness in transforming heterogeneous human intents into high-fidelity parametric CAD models. Our project page: https://zqwlearning.github.io/HiCAD.