Sketch2Arti: Sketch-based Articulation Modeling of CAD Objects

2026-04-28

AI Summary

This work addresses the problem of automatically inferring movable parts and their motion parameters in CAD models from user-drawn 2D sketches. It proposes a category-agnostic, sketch-driven approach to articulated modeling, wherein users express motion intent through simple 2D strokes—such as arrows—drawn from a single viewpoint. The system integrates 2D sketch understanding, 3D geometric analysis, and kinematic constraint reasoning to automatically identify movable components and predict their motion parameters. The method supports iterative multi-joint modeling and can complete internal structures for shell-based models. To the best of our knowledge, this is the first end-to-end framework capable of generating articulated CAD models directly from sketches without requiring category labels, demonstrating strong generalization across diverse object types. Experiments show significant improvements over existing methods in modeling quality, control flexibility, and cross-category adaptability.

Abstract

Articulation modeling aims to infer movable parts and their motion parameters for a 3D object, enabling interactive animation, simulation, and shape editing. In this paper, we present Sketch2Arti, the first sketch-based articulation modeling system for CAD objects. Our key observation is that designers naturally communicate articulation intent through lightweight sketches (e.g., arrows and strokes) that indicate how parts should move, yet translating such sketches into articulated 3D models remains largely manual. Sketch2Arti bridges this gap by enabling users to specify articulation through simple 2D sketches drawn from a chosen viewpoint. Given a CAD model and user sketches, our approach automatically discovers the corresponding movable parts and predicts their motion parameters, allowing iterative modeling of multiple articulations on complex objects with fine-grained control. Importantly, Sketch2Arti is trained in a category-agnostic manner without requiring object category information, leading to strong generalization to diverse objects beyond existing articulation datasets. Moreover, for shell models lacking interior structures, Sketch2Arti supports controllable internal completion guided by user sketches, generating plausible internal components consistent with the existing geometry and predicted motion constraints. Comprehensive experiments and user evaluations demonstrate the effectiveness, controllability, and generalization of Sketch2Arti. The code, dataset, and the prototype system are at https://arlo-yang.github.io/Sketch2Arti.

Problem

• articulation modeling
• sketch-based modeling
• CAD objects
• motion parameters
• 3D object animation

Innovation

• sketch-based modeling
• articulation modeling
• CAD objects
• category-agnostic learning
• internal structure completion