Research map / Detection and segmentation / 3D detection and driving
MultiScan: Scalable RGBD scanning for 3D environments with articulated objects
Read the paperCheck 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 SSTNet instead of PointGroup | object instance segmentation · MultiScan object seg | AP: 26.2 -> 32.6 | Helped |
| use HAIS instead of PointGroup | object instance segmentation · MultiScan object seg | AP: 26.2 -> 30.1 | Helped |
| use predicted instead of ground-truth object segmentation | SSTNet part segmentation · MultiScan part seg | AP: 29.8 -> 9.5 | Hurt |
| use predicted instead of ground-truth objects | SSTNet static part · MultiScan part seg | AP: 53.4 -> 8.5 | Hurt |
| closed instead of opened state | SSTNet part segmentation · MultiScan part seg | AP: 28.2 -> 19.1 | Hurt |
| closed instead of opened state | PointGroup door parts · MultiScan part seg | AP: 59.4 -> 18.6 | Hurt |
| closed instead of opened state | SSTNet window parts · MultiScan part seg | AP: 0.0 -> 0.5 | Helped |
| use HAIS instead of PointGroup | segmentation consistency across scans · MultiScan consistency | consistency: 44.43 -> 44.57 | No clear effect |
| use Shape2Motion instead of OPDPN | mobility estimation with GT seg · MultiScan mobility | movable part IoU: 54.80 -> 70.09 | Helped |
| use Shape2Motion instead of OPDPN | mobility estimation with GT seg · MultiScan mobility | Motion+Axis+Origin F1: 1.06 -> 7.15 | Helped |
| add data augmentation to Shape2Motion | mobility with GT seg · MultiScan mobility | Motion+Axis+Origin F1: 7.15 -> 5.81 | Hurt |
| use SSTNet predicted segmentation instead of GT | Shape2Motion · MultiScan mobility | movable part IoU: 70.09 -> 70.50 | No clear effect |
Closest papers by meaning
- SCRREAM : SCan, Register, REnder And Map: A Framework for Annotating Accurate and Dense 3D Indoor Scenes with a Benchmark (NeurIPS 2024 · similarity 0.86)
- SceneFun3D: Fine-Grained Functionality and Affordance Understanding in 3D Scenes (CVPR 2024 · similarity 0.86)
- PartGen: Part-level 3D Generation and Reconstruction with Multi-view Diffusion Models (CVPR 2025 · similarity 0.85)
- Stable Part Diffusion 4D: Multi-View RGB and Kinematic Parts Video Generation (NeurIPS 2025 · similarity 0.85)
- InternScenes: A Large-scale Simulatable Indoor Scene Dataset with Realistic Layouts (NeurIPS 2025 · similarity 0.85)
- PartNeXt: A Next-Generation Dataset for Fine-Grained and Hierarchical 3D Part Understanding (NeurIPS 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.