KineDiff3D: Kinematic-Aware Diffusion for Category-Level Articulated Object Shape Reconstruction and Generation
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866915563420254208 |
|---|---|
| author | Xu, WenBo Liu, Liu Zhang, Li Zhang, Ran Wu, Hao Guo, Dan Wang, Meng |
| author_facet | Xu, WenBo Liu, Liu Zhang, Li Zhang, Ran Wu, Hao Guo, Dan Wang, Meng |
| contents | Articulated objects, such as laptops and drawers, exhibit significant challenges for 3D reconstruction and pose estimation due to their multi-part geometries and variable joint configurations, which introduce structural diversity across different states. To address these challenges, we propose KineDiff3D: Kinematic-Aware Diffusion for Category-Level Articulated Object Shape Reconstruction and Generation, a unified framework for reconstructing diverse articulated instances and pose estimation from single view input. Specifically, we first encode complete geometry (SDFs), joint angles, and part segmentation into a structured latent space via a novel Kinematic-Aware VAE (KA-VAE). In addition, we employ two conditional diffusion models: one for regressing global pose (SE(3)) and joint parameters, and another for generating the kinematic-aware latent code from partial observations. Finally, we produce an iterative optimization module that bidirectionally refines reconstruction accuracy and kinematic parameters via Chamfer-distance minimization while preserving articulation constraints. Experimental results on synthetic, semi-synthetic, and real-world datasets demonstrate the effectiveness of our approach in accurately reconstructing articulated objects and estimating their kinematic properties. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_17137 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | KineDiff3D: Kinematic-Aware Diffusion for Category-Level Articulated Object Shape Reconstruction and Generation Xu, WenBo Liu, Liu Zhang, Li Zhang, Ran Wu, Hao Guo, Dan Wang, Meng Computer Vision and Pattern Recognition Articulated objects, such as laptops and drawers, exhibit significant challenges for 3D reconstruction and pose estimation due to their multi-part geometries and variable joint configurations, which introduce structural diversity across different states. To address these challenges, we propose KineDiff3D: Kinematic-Aware Diffusion for Category-Level Articulated Object Shape Reconstruction and Generation, a unified framework for reconstructing diverse articulated instances and pose estimation from single view input. Specifically, we first encode complete geometry (SDFs), joint angles, and part segmentation into a structured latent space via a novel Kinematic-Aware VAE (KA-VAE). In addition, we employ two conditional diffusion models: one for regressing global pose (SE(3)) and joint parameters, and another for generating the kinematic-aware latent code from partial observations. Finally, we produce an iterative optimization module that bidirectionally refines reconstruction accuracy and kinematic parameters via Chamfer-distance minimization while preserving articulation constraints. Experimental results on synthetic, semi-synthetic, and real-world datasets demonstrate the effectiveness of our approach in accurately reconstructing articulated objects and estimating their kinematic properties. |
| title | KineDiff3D: Kinematic-Aware Diffusion for Category-Level Articulated Object Shape Reconstruction and Generation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2510.17137 |