ArtPro: Self-Supervised Articulated Object Reconstruction with Adaptive Integration of Mobility Proposals
This work provides a more robust method for reconstructing complex articulated objects, which is important for robotics and simulation applications where high-fidelity digital twins are needed.
This paper addresses the challenge of reconstructing articulated objects into high-fidelity digital twins, which is crucial for applications like robotic manipulation. The authors propose ArtPro, a self-supervised framework that uses adaptive integration of mobility proposals to overcome the sensitivity of existing methods to initial part segmentation, achieving robust reconstruction of complex multi-part objects.
Reconstructing articulated objects into high-fidelity digital twins is crucial for applications such as robotic manipulation and interactive simulation. Recent self-supervised methods using differentiable rendering frameworks like 3D Gaussian Splatting remain highly sensitive to the initial part segmentation. Their reliance on heuristic clustering or pre-trained models often causes optimization to converge to local minima, especially for complex multi-part objects. To address these limitations, we propose ArtPro, a novel self-supervised framework that introduces adaptive integration of mobility proposals. Our approach begins with an over-segmentation initialization guided by geometry features and motion priors, generating part proposals with plausible motion hypotheses. During optimization, we dynamically merge these proposals by analyzing motion consistency among spatial neighbors, while a collision-aware motion pruning mechanism prevents erroneous kinematic estimation. Extensive experiments on both synthetic and real-world objects demonstrate that ArtPro achieves robust reconstruction of complex multi-part objects, significantly outperforming existing methods in accuracy and stability.