Research map / 3D vision / 3D generation
Holistic 3D Scene Parsing and Reconstruction from a Single RGB Image
Siyuan Huang, Siyuan Qi, Yixin Zhu, Yinxue Xiao, Yuanlu Xu, Song-Chun Zhu
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 |
|---|---|---|---|
| add joint inference of objects and layout over initialization (ablation) | Ours (init.) · SUN RGB-D (5050 images) | 3D layout IoU: 46.7 -> 54.9 | Improved |
| add joint inference (ablation) | Ours (init.) · SUN RGB-D (5050 images) | geometric precision Pg: 25.9 -> 37.7 | Improved |
| add joint inference (ablation) | Ours (init.) · SUN RGB-D (749 images) | free-space IoU: 47.1 -> 52.1 | Improved |
| add joint inference (ablation) | Ours (init.) · SUN RGB-D 3D object detection | mAP: 7.35 -> 12.07 | Improved |
| add support relation energy term (C1) (ablation) | Ours · SUN RGB-D offices | free-space IoU: 42.3 -> 44.7 | Improved |
| add physical constraint term (C2) (ablation) | Ours · SUN RGB-D offices | free-space IoU: 41.3 -> 44.7 | Improved |
| add physical constraint term (C2) (ablation) | Ours · SUN RGB-D offices | Pg: 23.5 -> 34.4 | Improved |
| add latent human imagination term (C3) (ablation) | Ours · SUN RGB-D offices | free-space IoU: 43.8 -> 44.7 | Improved |
| add all three terms C1, C2, C3 (ablation) | Ours · SUN RGB-D offices | free-space IoU: 38.4 -> 44.7 | Improved |
| optimize both layout and objects jointly instead of layout only (S4) (ablation) | Ours · SUN RGB-D offices | free-space IoU: 39.4 -> 44.7 | Improved |
| optimize both layout and objects jointly instead of objects only (S5) (ablation) | Ours · SUN RGB-D offices | free-space IoU: 36.3 -> 44.7 | Improved |
| use Ours (joint.) instead of IM2CAD | Ours · SUN RGB-D (749 images) | free-space IoU: 49.0 -> 52.1 | Improved |
Closest papers by meaning
- Single-View 3D Scene Reconstruction and Parsing by Attribute Grammar (TPAMI 2017 · similarity 0.91)
- Human-Centric Indoor Scene Synthesis Using Stochastic Grammar (CVPR 2018 · similarity 0.91)
- Holistic++ Scene Understanding: Single-View 3D Holistic Scene Parsing and Human Pose Estimation With Human-Object Interaction and Physical Commonsense (ICCV 2019 · similarity 0.90)
- Open-Vocabulary Functional 3D Scene Graphs for Real-World Indoor Spaces (CVPR 2025 · similarity 0.87)
- InstructScene: Instruction-Driven 3D Indoor Scene Synthesis with Semantic Graph Prior (ICLR 2024 · similarity 0.87)
- SceneFun3D: Fine-Grained Functionality and Affordance Understanding in 3D Scenes (CVPR 2024 · similarity 0.86)
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.
