Highlights
Introduce Articulation-XL, a large-scale dataset containing over 33k 3D models with high-quality articulation annotations carefully curated from Objaverse-XL.
Formulate skeleton generation as a sequence modeling problem, leveraging an auto-regressive transformer to naturally handle varying numbers of bones or joints within skeletons and their inherent
dependencies across different 3D models.
Predict skinning weights using a functional diffusion process
that incorporates volumetric geodesic distance priors between vertices and joints.
Abstract
With the explosive growth of 3D content creation, there is an increasing demand for automatically converting static 3D models into articulation-ready versions that support realistic animation. Traditional approaches rely heavily on manual annotation, which is both time-consuming and labor-intensive. Moreover, the lack of large-scale benchmarks has hindered the development of learning-based solutions.
In this work, we present MagicArticulate, an effective framework that automatically transforms static 3D models into articulation-ready assets. Our key contributions are threefold. First, we introduce Articulation-XL, a large-scale benchmark containing over 33k 3D models with high-quality articulation annotations, carefully curated from Objaverse-XL. Second, we propose a novel skeleton generation method that formulates the task as a sequence modeling problem, leveraging an auto-regressive transformer to naturally handle varying numbers of bones or joints within skeletons and their inherent dependencies across different 3D models. Third, we predict skinning weights using a functional diffusion process that incorporates volumetric geodesic distance priors between vertices and joints. Extensive experiments demonstrate that MagicArticulate significantly outperforms existing methods across diverse object categories, achieving high-quality articulation that enables realistic animation.
Dataset: Articulation-XL
To enable large-scale learning of 3D model articulation, we introduce Articulation-XL, a curated dataset of over 33k 3D models with high-quality articulation annotations from Objaverse-XL. Articulation-XL statistics are shown above. We have expanded the dataset to over 48K models in Articulation-XL2.0. For further details, please check here.
Method overview
Overview of our method for auto-regressive skeleton generation. Given an input mesh, we begin by sampling point clouds from its surface. These sampled points are then encoded into fixed-length shape tokens, which are appended to the start of skeleton tokens to achieve auto-regressive skeleton generation conditioned on input shapes.
Video
Sequence ordering
We present demos of spatial sequence ordering versus hierarchical sequence ordering.
Skeleton generation results
We present demos of auto-regressive skeleton generation.
Comparison of skinning weight prediction
We compare our skinning weight prediction results with RigNet and Geodesic Voxel Binding (GVB).
Animation results
BibTeX
@article{song2025magicarticulate,
title={MagicArticulate: Make Your 3D Models Articulation-Ready},
author={Chaoyue Song and Jianfeng Zhang and Xiu Li and Fan Yang and Yiwen Chen and Zhongcong Xu and Jun Hao Liew and Xiaoyang Guo and Fayao Liu and Jiashi Feng and Guosheng Lin},
journal={arXiv preprint arXiv:2502.12135},
year={2025},
}