Research map / 3D vision / 3D generation
IM2CAD
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 |
|---|---|---|---|
| use IM2CAD room-layout estimation | room layout estimation · Hedau [15] test | pixel misclassification error (%): 10.15 vs. 9.73 | Mixed |
| use IM2CAD room-layout estimation | room layout estimation · LSUN | pixel misclassification error (%): 10.04 vs. 9.31 | Mixed |
| use IM2CAD room geometry estimation | 3D room layout estimation · SUN RGB-D | voxel IoU: 49.4 -> 62.6 | Helped |
| use IM2CAD room geometry estimation | 3D room layout estimation · 3DGP | voxel IoU: 47.3 -> 63.2 | Helped |
| add joint scene optimization (ablation) | IM2CAD · SUN RGB-D | voxel IoU: 46.1 -> 49.0 | Helped |
| add joint scene optimization (ablation) | IM2CAD · SUN RGB-D | mAP: 74.7 -> 75.6 | Helped |
| add joint scene optimization (ablation) | IM2CAD · 3DGP | voxel IoU: 53.5 -> 53.8 | Helped |
| add joint scene optimization (ablation) | IM2CAD · 3DGP | mAP: 86.6 -> 86.6 | No clear effect |
| use FCN geometric features without 3D box estimation (ablation) | room layout estimation · Hedau [15] benchmark | pixel misclassification error (%): 28.9% -> 12.4% | Helped |
| combine the method with Hedau [15] room hypotheses (ablation) | IM2CAD room-layout estimation · Hedau [15] test | pixel misclassification error (%): 10.15% -> 11.02% | Hurt |
| combine the method with Lee [22] room hypotheses (ablation) | IM2CAD room-layout estimation · Hedau [15] test | pixel misclassification error (%): 10.15% -> 11.13% | Hurt |
Closest papers by meaning
- Scan2CAD: Learning CAD Model Alignment in RGB-D Scans (CVPR 2019 · similarity 0.87)
- HomeDiffusion: Zero-Shot Object Customization with Multi-View Representation Learning for Indoor Scenes (AAAI 2025 · similarity 0.85)
- SceneCraft: Layout-Guided 3D Scene Generation (NeurIPS 2024 · similarity 0.84)
- S-INF: Towards Realistic Indoor Scene Synthesis via Scene Implicit Neural Field (AAAI 2025 · similarity 0.84)
- InternScenes: A Large-scale Simulatable Indoor Scene Dataset with Realistic Layouts (NeurIPS 2025 · similarity 0.83)
- InstructScene: Instruction-Driven 3D Indoor Scene Synthesis with Semantic Graph Prior (ICLR 2024 · similarity 0.83)
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.