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

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

Xinyu Lian1,2
Zichao Yu3
Ruiming Liang7,8
Yitong Wang5
Li Ray Luo1
Kaixu Chen1,4
Yuanzhen Zhou1
Qihong Tang6
Xudong Xu1
Zhaoyang Lyu1
Bo Dai9
Jiangmiao Pang1
1Shanghai AI Laboratory
2South China University of Technology
3University of Science and Technology of China
4Tongji University
5Fudan University
6Harbin Institute of Technology, Shenzhen
7Institute of Automation, Chinese Academy of Sciences
8School of Artificial Intelligence, University of Chinese Academy of Sciences
9The University of Hong Kong
description Paper description BibTeX description Code

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.


The whole pipeline can be devided into four parts: articulation tree structure generation, geometry generation, material generation and joint generation.

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.

Paper

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

Xinyu Lian, Zichao Yu, Ruiming Liang, Yitong Wang, Li Ray Luo, Kaixu Chen, Qihong Tang, Xudong XU, Zhaoyang Lyu, Bo Dai, Jiangmiao Pang

description PDF
description arXiv version
insert_comment BibTeX

Videos of Generated Objects

Cabinet

Pan

Pot

Display

Dishwasher

Microwave

Fridge

Oven

Bar Chair

Office Chair

Window

Lamp

Tap

Toilet

Jar

Door

Rack

Cocktail Table

Dining Table

Plate Rack

Bottle

Vase

Robotic Arm Manipulating

Robotic Arm Operating Cabinet

Robotic Arm Operating Microwave

Citation


@misc{lian2025infinitemobilityscalablehighfidelity,
      title={Infinite Mobility: Scalable High-Fidelity Synthesis of Articulated Objects via Procedural Generation}, 
      author={Xinyu Lian and Zichao Yu and Ruiming Liang and Yitong Wang and Li Ray Luo and Kaixu Chen and Yuanzhen Zhou and Qihong Tang and Xudong Xu and Zhaoyang Lyu and Bo Dai and Jiangmiao Pang},
      year={2025},
      eprint={2503.13424},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2503.13424}, 
}