UniArt: Unified 3D Representation for Generating 3D Articulated Objects with Open-Set Articulation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908677634523136 |
|---|---|
| author | Jin, Bu Li, Weize Gu, Songen Zheng, Yupeng Zheng, Yuhang Zhou, Zhengyi Yao, Yao |
| author_facet | Jin, Bu Li, Weize Gu, Songen Zheng, Yupeng Zheng, Yuhang Zhou, Zhengyi Yao, Yao |
| contents | Articulated 3D objects play a vital role in realistic simulation and embodied robotics, yet manually constructing such assets remains costly and difficult to scale. In this paper, we present UniArt, a diffusion-based framework that directly synthesizes fully articulated 3D objects from a single image in an end-to-end manner. Unlike prior multi-stage techniques, UniArt establishes a unified latent representation that jointly encodes geometry, texture, part segmentation, and kinematic parameters. We introduce a reversible joint-to-voxel embedding, which spatially aligns articulation features with volumetric geometry, enabling the model to learn coherent motion behaviors alongside structural formation. Furthermore, we formulate articulation type prediction as an open-set problem, removing the need for fixed joint semantics and allowing generalization to novel joint categories and unseen object types. Experiments on the PartNet-Mobility benchmark demonstrate that UniArt achieves state-of-the-art mesh quality and articulation accuracy. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_21887 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | UniArt: Unified 3D Representation for Generating 3D Articulated Objects with Open-Set Articulation Jin, Bu Li, Weize Gu, Songen Zheng, Yupeng Zheng, Yuhang Zhou, Zhengyi Yao, Yao Computer Vision and Pattern Recognition Articulated 3D objects play a vital role in realistic simulation and embodied robotics, yet manually constructing such assets remains costly and difficult to scale. In this paper, we present UniArt, a diffusion-based framework that directly synthesizes fully articulated 3D objects from a single image in an end-to-end manner. Unlike prior multi-stage techniques, UniArt establishes a unified latent representation that jointly encodes geometry, texture, part segmentation, and kinematic parameters. We introduce a reversible joint-to-voxel embedding, which spatially aligns articulation features with volumetric geometry, enabling the model to learn coherent motion behaviors alongside structural formation. Furthermore, we formulate articulation type prediction as an open-set problem, removing the need for fixed joint semantics and allowing generalization to novel joint categories and unseen object types. Experiments on the PartNet-Mobility benchmark demonstrate that UniArt achieves state-of-the-art mesh quality and articulation accuracy. |
| title | UniArt: Unified 3D Representation for Generating 3D Articulated Objects with Open-Set Articulation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2511.21887 |