Research map / 3D vision / 3D generation
Ditto: Building Digital Twins of Articulated Objects From Interaction
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| use dense per-point joint predictions instead of global joint regression | Ditto · Synthetic Dataset | Revolute angle error: 52.0 -> 0.72 | Helped |
| use dense per-point joint predictions instead of global joint regression | Ditto · Shape2Motion | Revolute angle error: 79.8 -> 1.36 | Helped |
| add attention-based fusion across the two observations (ablation) | Ditto · Shape2Motion | Mobile Chamfer Distance: 0.42 vs 3.09 | Helped |
| keep geometry and articulation feature representations separate (ablation) | Ditto · Shape2Motion | Mobile Chamfer Distance: 0.42 vs 10.7 | Helped |
| use separate decoders for geometry and articulation (ablation) | Ditto · Shape2Motion | Mobile Chamfer Distance: 0.42 vs 3.30 | Helped |
| use separate decoders for geometry and articulation (ablation) | Ditto · Shape2Motion | Revolute angle error: 1.93 vs 1.36 | Helped |
| use 3D grids for geometry and 2D planes for articulation rather than 3D grids for both (ablation) | Ditto · Synthetic Dataset | Whole Chamfer Distance: 0.27 vs 0.25 | No clear effect |
| use 3D grids for geometry and 2D planes for articulation rather than 3D grids for both (ablation) | Ditto · Synthetic Dataset | Mobile Chamfer Distance: 0.12 vs 0.16 | No clear effect |
| use 3D grids for both geometry and articulation (ablation) | Ditto · Synthetic Dataset | memory usage and training time: around 20% more | Hurt |
| fix A-SDF's articulation code to its ground-truth value (ablation) | A-SDF · Synthetic Dataset | Whole Chamfer Distance: 2.48 -> 0.66 | Helped |
| use explicit part meshes and joint parameters rather than an implicit articulation code | Ditto · Synthetic Dataset articulated motion synthesis | Chamfer Distance: 3.57 -> 0.37 | Helped |
| replace attention fusion with concatenated structured features (ablation) | Ditto · Shape2Motion | Joint type accuracy: 100 -> 96 | Hurt |
Closest papers by meaning
- Articulate-Anything: Automatic Modeling of Articulated Objects via a Vision-Language Foundation Model (ICLR 2025 · similarity 0.89)
- IAAO: Interactive Affordance Learning for Articulated Objects in 3D Environments (CVPR 2025 · similarity 0.88)
- DIPO: Dual-State Images Controlled Articulated Object Generation Powered by Diverse Data (NeurIPS 2025 · similarity 0.87)
- ArticulatedGS: Self-supervised Digital Twin Modeling of Articulated Objects using 3D Gaussian Splatting (CVPR 2025 · similarity 0.87)
- SINGAPO: Single Image Controlled Generation of Articulated Parts in Objects (ICLR 2025 · similarity 0.87)
- Building Interactable Replicas of Complex Articulated Objects via Gaussian Splatting (ICLR 2025 · 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.