Research map / 3D vision / Gaussian splatting
SplArt: Articulation Estimation and Part-Level Reconstruction with 3D Gaussian Splatting
Shengjie Lin, Jiading Fang, Muhammad Zubair Irshad, Vitor Campagnolo Guizilini, Rares Andrei Ambrus, Greg Shakhnarovich, Matthew R. Walter
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 |
|---|---|---|---|
| replace NeRF-based PARIS with 3DGS multi-stage SplArt | SplArt · PARIS-PMS | success rate: 42.0% -> 99.0% | Improved |
| SplArt versus DTA (which uses ground-truth depth) on mesh Chamfer distance | SplArt · PARIS-PMS (overall) | CD_w (x10^-3): 1.83 (DTA) vs 3.61 | Hurt |
| add Stage 3(a) geometric consistency for articulation estimation (ablation) | SplArt · SplArt-PMS | success rate: 6.4% -> 85.5% | Improved |
| add Stage 2(b) photometric mobility estimation (ablation) | SplArt · SplArt-PMS | success rate: 74.7% -> 85.5% | Improved |
| add Stage 2(a) coarse geometric mobility estimation (ablation) | SplArt · SplArt-PMS | success rate: success 83.7% -> 85.5%; err_a 13.85 -> 23.02 | Mixed |
| add Stage 3(b) photometric articulation refinement (ablation) | SplArt · SplArt-PMS | err_a (x10^-2 deg): 73.62 -> 23.02 | Improved |
| use cross-mobile consistency for toaster (flat slider) instead of mobile-only (ablation) | Stage 3(a) · 103549-Toaster | articulation estimation success | Hurt |
| combine mobile-only and cross-mobile geometric consistency with randomized tries (ablation) | Stage 3(a) · SplArt-PMS | robustness | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- PARIS — PARIS: Part-level Reconstruction and Motion Analysis for Articulated Objects (ICCV 2023) · better on 9 of 9 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- ArticulatedGS: Self-supervised Digital Twin Modeling of Articulated Objects using 3D Gaussian Splatting (CVPR 2025 · similarity 0.91)
- Building Interactable Replicas of Complex Articulated Objects via Gaussian Splatting (ICLR 2025 · similarity 0.90)
- SplatFormer: Point Transformer for Robust 3D Gaussian Splatting (ICLR 2025 · similarity 0.88)
- REArtGS: Reconstructing and Generating Articulated Objects via 3D Gaussian Splatting with Geometric and Motion Constraints (NeurIPS 2025 · similarity 0.88)
- Template-free Articulated Gaussian Splatting for Real-time Reposable Dynamic View Synthesis (NeurIPS 2024 · similarity 0.88)
- PF3plat: Pose-Free Feed-Forward 3D Gaussian Splatting for Novel View Synthesis (ICML 2025 · similarity 0.88)
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.
