Infinite Mobility: Scalable High-Fidelity Synthesis of Articulated Objects via Procedural Generation

TL;DR

This work tackles the scarcity of high-quality, scalable articulated-object data for embodied AI by introducing Infinite Mobility, a procedural pipeline that builds URDF-like articulation trees and procedurally generates geometry, materials, and joints, with mesh retrieval and refinement to ensure plausibility. It demonstrates superior physical fidelity and mesh quality relative to human-annotated datasets and state-of-the-art generative approaches, while providing synthetic data that can train large generative models and support embodied AI tasks in simulators. The authors also address practical issues of physical plausibility, such as ground collisions and joint stability, through targeted structural adjustments. Overall, the approach enables scalable production of diverse, high-fidelity articulated objects and provides a foundation for advancing sim-to-real and embodied AI research, with code released for community use.

Abstract

Large-scale articulated objects with high quality are desperately needed for multiple tasks related to embodied AI. Most existing methods for creating articulated objects are either data-driven or simulation based, which are limited by the scale and quality of the training data or the fidelity and heavy labour of the simulation. In this paper, we propose Infinite Mobility, a novel method for synthesizing high-fidelity articulated objects through procedural generation. User study and quantitative evaluation demonstrate that our method can produce results that excel current state-of-the-art methods and are comparable to human-annotated datasets in both physics property and mesh quality. Furthermore, we show that our synthetic data can be used as training data for generative models, enabling next-step scaling up. Code is available at https://github.com/Intern-Nexus/Infinite-Mobility

Figures (12)

• Figure 1: We design probalistic programs to generate $22$ common articulated objects. We demonstrate the motion sequence of the generated articulated objects. Ours generated articulated objects bear accurate geometry, realistic textures, and reasonable joints.
• Figure 2: The whole pipeline can be devided into four parts: articulation tree structure generation, geometry generation, material generation and joint generation.
• Figure 3: Structure of the URDF file. Each link is a part of the object, which is represented as a textured mesh in our case. Each joint connects two links and describes the articulation structure between them.
• Figure 4: We implement $6$ kinds of joints in our articulated objects. First three are simple joints, and the last three are compound joints.
• Figure 5: Examples of our generated cabinets, chairs, lamps and windows. For each object, we display the textured mesh with the corresponding articulation tree above. The generated objects are diverse in both shapes and articulation structures.