Research map / 3D vision / Face and shape
Superquadrics Revisited: Learning 3D Shape Parsing Beyond Cuboids
Despoina Paschalidou, Ali Osman Ulusoy, Andreas Geiger
Check 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 cuboid primitives with superquadrics | Tulsiani et al. cuboid parsing · ShapeNet Chairs | Chamfer distance: 0.0121 -> 0.0006 | Improved |
| Replace cuboid primitives with superquadrics | Tulsiani et al. cuboid parsing · ShapeNet Aeroplanes | Chamfer distance: 0.0153 -> 0.0003 | Improved |
| Replace cuboid primitives with superquadrics | Tulsiani et al. cuboid parsing · ShapeNet Animals | Chamfer distance: 0.0110 -> 0.0003 | Improved |
| Replace cuboid primitives with superquadrics | Tulsiani et al. cuboid parsing · ShapeNet Chairs | Volumetric IoU: 0.1288 -> 0.1408 | Improved |
| Replace cuboid primitives with superquadrics | Tulsiani et al. cuboid parsing · ShapeNet Aeroplanes | Volumetric IoU: 0.0650 -> 0.1808 | Improved |
| Replace cuboid primitives with superquadrics | Tulsiani et al. cuboid parsing · ShapeNet Animals | Volumetric IoU: 0.3339 -> 0.7506 | Improved |
| Use standard Chamfer distance instead of truncated bidirectional loss (ablation) | Superquadric parsing · 2D toy fitting experiment | Local-minimum avoidance | Improved |
| Use the analytical expected existence-variable loss instead of sampling-based gradient estimation | Superquadric parsing · Shape parsing training | Gradient variance | Improved |
| Use the analytical expected existence-variable loss instead of sampling-based gradient estimation | Superquadric parsing · Shape parsing training | Iteration time | Improved |
| Replace cuboids with superquadrics (ablation) | Proposed parsing model · ShapeNet Animals reconstruction with varying primitive limits | Reconstruction loss | Improved |
Closest papers by meaning
- Self-supervised Learning of Hybrid Part-aware 3D Representations of 2D Gaussians and Superquadrics (ICCV 2025 · similarity 0.89)
- SuperDec: 3D Scene Decomposition with Superquadrics Primitives (ICCV 2025 · similarity 0.88)
- QuadricFormer: Scene as Superquadrics for 3D Semantic Occupancy Prediction (NeurIPS 2025 · similarity 0.86)
- Collaborative Learning for 3D Hand-Object Reconstruction and Compositional Action Recognition from Egocentric RGB Videos Using Superquadrics (AAAI 2025 · similarity 0.85)
- Learning Partonomic 3D Reconstruction from Image Collections (CVPR 2025 · similarity 0.84)
- 3DCoMPaT200: Language Grounded Large-Scale 3D Vision Dataset for Compositional Recognition (NeurIPS 2024 · similarity 0.84)
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.
