Articulated objects are common in the real world, yet modeling their structure and motion remains a challenging task for 3D reconstruction methods. In this work, we introduce PartGS, a novel framework for modeling articulated digital twins of multi-part objects with high-fidelity geometry and physically consistent articulation. PartGS leverages a part-aware 3D Gaussian representation that encodes articulated components with learnable attributes, enabling structured, disentangled transformations that preserve high-fidelity geometry. To ensure physically consistent motion, we propose a motion-aware canonical representation guided by physics-based constraints, including contact enforcement, velocity consistency, and vector-field alignment. Furthermore, we introduce a field of repel points to prevent part collisions and maintain stable articulation paths, significantly improving motion coherence over baselines. Extensive evaluations on both synthetic and real-world datasets show that PartGS consistently outperforms state-of-the-art methods by up to 10 in Chamfer Distance for movable parts.
View on arXiv@article{yu2025_2506.17212,
title={ Part$^{2}$GS: Part-aware Modeling of Articulated Objects using 3D Gaussian Splatting },
author={ Tianjiao Yu and Vedant Shah and Muntasir Wahed and Ying Shen and Kiet A. Nguyen and Ismini Lourentzou },
journal={arXiv preprint arXiv:2506.17212},
year={ 2025 }
}