Research map / 3D vision / 3D generation
SuperDec: 3D Scene Decomposition with Superquadrics Primitives
Elisabetta Fedele, Boyang Sun, Leonidas Guibas, Marc Pollefeys, Francis Engelmann
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 |
|---|---|---|---|
| use SuperDec instead of SQ baseline | superquadric decomposition · ShapeNet in-category | L2 Chamfer: 0.279 -> 0.047 | Improved |
| use SuperDec instead of EMS optimization | superquadric decomposition · ShapeNet in-category | L2 Chamfer: 1.345 -> 0.047 | Improved |
| use SuperDec instead of CSA cuboids | decomposition · ShapeNet out-of-category | L2 Chamfer: 0.427 -> 0.061 | Improved |
| use SuperDec instead of CSA | decomposition · ScanNet++ objects | L2 Chamfer: 0.41 -> 0.11 | Improved |
| use SuperDec instead of EMS | decomposition · ScanNet++ objects | number of primitives: 4.25 -> 5.18 | Hurt |
| use SuperDec instead of CSA | decomposition · Replica objects | L2 Chamfer: 0.70 -> 0.19 | Improved |
| use SuperDec instead of cuboids for path planning | representation · 15 ScanNet++ scenes | success rate (%): 61.23 -> 91.71 | Improved |
| use SuperDec instead of dense point cloud for path planning | representation · 15 ScanNet++ scenes | memory (MB): 19.286 -> 0.042 | Improved |
| use SuperDec instead of occupancy grid for path planning | representation · 15 ScanNet++ scenes | success rate (%): 100.00 -> 91.71 | Hurt |
| add existence head alpha at training and inference (ablation) | SuperDec · ShapeNet in-category | L2 Chamfer: 0.049 -> 0.047 | Mixed |
| use predicted alpha at inference (ablation) | SuperDec · ShapeNet in-category | L2 Chamfer: 0.046 -> 0.047 | No clear effect |
| use farthest point sampling instead of random (ablation) | SuperDec · ScanNet++ objects | L2 Chamfer: 0.13 -> 0.11 | Improved |
Closest papers by meaning
- SuperFlex: Deformable Superquadrics for Point Cloud Decomposition (ECCV 2026 · similarity 0.91)
- QuadricFormer: Scene as Superquadrics for 3D Semantic Occupancy Prediction (NeurIPS 2025 · similarity 0.87)
- SuperPrimitive: Scene Reconstruction at a Primitive Level (CVPR 2024 · similarity 0.86)
- SCube: Instant Large-Scale Scene Reconstruction using VoxSplats (NeurIPS 2024 · similarity 0.85)
- SuperNormal: Neural Surface Reconstruction via Multi-View Normal Integration (CVPR 2024 · similarity 0.85)
- Total-Decom: Decomposed 3D Scene Reconstruction with Minimal Interaction (CVPR 2024 · similarity 0.85)
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.
