Research map / 3D vision / Gaussian splatting
ArtPro: Self-Supervised Articulated Object Reconstruction with Adaptive Integration of Mobility Proposals
Xuelu Li, Zhaonan Wang, Xiaogang Wang, Lei Wu, Manyi Li, Changhe Tu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add over-segmentation initialization (OverSeg) on top of MoI without knowing GT part number (ablation) | ArtPro baseline · Own multi-part dataset | CD-m: 6.97 -> 5.01 | Improved |
| add motion initialization (MoI) to DBSCAN baseline (ablation) | ArtPro baseline · Own multi-part dataset | CD-m: 9.27 -> 6.97 | Improved |
| add adaptive proposal integration (Merge) (ablation) | ArtPro without Prune · Own multi-part dataset | CD-m: 5.01 -> 4.57 | Improved |
| add collision-aware motion pruning (ablation) | ArtPro Merge variant · Own multi-part dataset | CD-m: 4.57 -> 3.63 | Improved |
| add collision-aware motion pruning (ablation) | ArtPro Merge variant · Own multi-part dataset | CD-w: 0.67 -> 0.65 | Improved |
| add progressive merging after over-segmentation (ablation) | ArtPro · Own multi-part dataset | CD-m: 4.26 -> 3.63 | Improved |
| add principal-axis motion initialization (MoI) (ablation) | ArtPro · Own multi-part dataset | CD-m: 5.62 -> 3.63 | Improved |
| add MoI and pruning together (ablation) | ArtPro · Own multi-part dataset | CD-m: 5.81 -> 3.63 | Improved |
| add over-segmentation and merging instead of DBSCAN with GT part number (ablation) | ArtPro baseline · Own multi-part dataset | CD-w: 0.93 -> 0.65 | Improved |
| use ArtPro instead of ArtGS on complex multi-part objects | ArtGS · Own multi-part dataset | Part motion error: 9.50 -> 0.04 | Improved |
| use ArtPro instead of ArtGS | ArtGS · ArtGS-Multi dataset | CD-m: 2.14 -> 0.89 | Improved |
| use ArtPro instead of ArtGS for visual quality | ArtGS · Own multi-part dataset | PSNR: 36.10 -> 48.02 | Improved |
Closest papers by meaning
- ArticulatedGS: Self-supervised Digital Twin Modeling of Articulated Objects using 3D Gaussian Splatting (CVPR 2025 · similarity 0.92)
- Building Interactable Replicas of Complex Articulated Objects via Gaussian Splatting (ICLR 2025 · similarity 0.91)
- SINGAPO: Single Image Controlled Generation of Articulated Parts in Objects (ICLR 2025 · similarity 0.87)
- Template-free Articulated Gaussian Splatting for Real-time Reposable Dynamic View Synthesis (NeurIPS 2024 · similarity 0.87)
- TAGA: Self-supervised Learning for Template-free Animatable Gaussian Articulated Model (CVPR 2025 · similarity 0.87)
- Articulate your NeRF: Unsupervised articulated object modeling via conditional view synthesis (NeurIPS 2024 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
