Research map / Point clouds / Segmentation and detection
3D Semantic Parsing of Large-Scale Indoor Spaces
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 |
|---|---|---|---|
| Parse building point clouds into final spaces using void-based dividers, merging, and element-based updates | Ours · Building space parsing, mean | ARI: 0.2 -> 0.77; 0.31 -> 0.77 | Helped |
| Add the merging and updating steps to the over-segmented space parsing (ablation) | Ours · Building space parsing, mean | ARI: 0.55 -> 0.77 | Helped |
| Use the full hierarchical point-cloud parsing method instead of Vote3D | Semantic element detection · Semantic element detection | mAP: 39.21 -> 49.93 | Helped |
| Use the full hierarchical point-cloud parsing method instead of mBOW | Semantic element detection · Semantic element detection | mAP: 36.11 -> 49.93 | Helped |
| Add local geometry features (ablation) | Ours · Semantic element detection | mAP: 44.19 -> 49.93 | Helped |
| Add global geometry features (ablation) | Ours · Semantic element detection | mAP: 41.87 -> 49.93 | Helped |
| Add color features (ablation) | Ours · Semantic element detection | mAP: 45.41 -> 49.93 | Helped |
| Transfer 3D point-cloud parsing results to RGB-D images rather than use the RGB-D segmentation baseline | Ours · RGB-D segmentation | mIOU: 20.9 -> 38.5 | Helped |
Closest papers by meaning
- SpatialLM: Training Large Language Models for Structured Indoor Modeling (NeurIPS 2025 · similarity 0.86)
- Geometric Exploitation for Indoor Panoramic Semantic Segmentation (NeurIPS 2024 · similarity 0.86)
- From Indoor to Open World: Revealing the Spatial Reasoning Gap in MLLMs (CVPR 2026 · similarity 0.85)
- Rooms from Motion: Un-posed Indoor 3D Object Detection as Localization and Mapping (NeurIPS 2025 · similarity 0.85)
- HouseLayout3D: A Benchmark and Training-free Baseline for 3D Layout Estimation in the Wild (NeurIPS 2025 · similarity 0.85)
- Open-Vocabulary Functional 3D Scene Graphs for Real-World Indoor Spaces (CVPR 2025 · 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.