Research map / Point clouds / Shape and surface completion
MonteBoxFinder: Detecting and Filtering Primitives to Fit a Noisy Point Cloud
Michaël Ramamonjisoa, Sinisa Stekovic, Vincent Lepetit
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 MonteBoxFinder search instead of hill-climbing | cuboid primitive selection · ScanNet (1613 scenes) | loss: 0.383 -> 0.201 | Improved |
| use MonteBoxFinder search instead of MCTS | cuboid primitive selection · ScanNet (1613 scenes) | loss: 0.247 -> 0.201 | Improved |
| use MonteBoxFinder search instead of MCTS-Binary | cuboid primitive selection · ScanNet (1613 scenes) | loss: 0.292 -> 0.201 | Improved |
| use MonteBoxFinder instead of MCTS | cuboid primitive selection · ScanNet (1613 scenes) | AUC: 0.427 -> 0.322 | Improved |
| use MonteBoxFinder instead of MCTS-Binary | cuboid primitive selection · ScanNet (1613 scenes) | AUC: 0.370 -> 0.322 | Improved |
| use MonteBoxFinder instead of MCTS | cuboid primitive selection · ScanNet (1613 scenes) | precision: 0.966 -> 0.982 | Improved |
| use MonteBoxFinder instead of MCTS | cuboid primitive selection · ScanNet (1613 scenes) | avg. number of cuboids: 28 -> 37 | Improved |
| use MonteBoxFinder instead of hill-climbing | cuboid primitive selection · ScanNet (1613 scenes) | avg. number of cuboids: 12 -> 37 | Improved |
| use MCTS-Binary instead of original MCTS | MCTS baseline · ScanNet (1613 scenes) | loss: 0.247 -> 0.292 | Hurt |
| use MCTS-Binary instead of original MCTS | MCTS baseline · ScanNet (1613 scenes) | AUC: 0.427 -> 0.370 | Improved |
Closest papers by meaning
- Robust Shape Fitting for 3D Scene Abstraction (TPAMI 2024 · similarity 0.87)
- Cuboids Revisited: Learning Robust 3D Shape Fitting to Single RGB Images (CVPR 2021 · similarity 0.87)
- 3DMambaIPF: A State Space Model for Iterative Point Cloud Filtering via Differentiable Rendering (AAAI 2025 · similarity 0.86)
- CASAGPT: Cuboid Arrangement and Scene Assembly for Interior Design (CVPR 2025 · similarity 0.85)
- On Geometry-Enhanced Parameter-Efficient Fine-Tuning for 3D Scene Segmentation (NeurIPS 2025 · similarity 0.85)
- SAMBLE: Shape-Specific Point Cloud Sampling for an Optimal Trade-Off Between Local Detail and Global Uniformity (CVPR 2025 · 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.
