Research map / 3D vision / 3D generation
Planner3D: LLM-enhanced Graph Prior Meets 3D Indoor Scene Explicit Regularization
Wei Yao, Martin Renqiang Min, George Vosselman A, Li Erran Li, Michael Ying Yang
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 LLM-based global scene graph prior (ablation) | Planner3D · SG-FRONT (all rooms) | KID: 7.94 -> 6.13 | Improved |
| add LLM-based global scene graph prior (ablation) | Planner3D · SG-FRONT (all rooms) | FID: 47.71 -> 47.34 | Improved |
| add LLM-based global scene graph prior (ablation) | Planner3D · SG-FRONT (all rooms) | symmetrical relationship consistency: 0.54 -> 0.66 | Improved |
| add LLM-based global scene graph prior (ablation) | Planner3D · SG-FRONT (all rooms) | mSG: 0.88 -> 0.90 | Improved |
| add IoU layout regularization (ablation) | Planner3D · SG-FRONT (all rooms) | FID: 48.15 -> 47.34 | Improved |
| add IoU layout regularization (ablation) | Planner3D · SG-FRONT (all rooms) | KID: 6.93 -> 6.13 | Improved |
| add IoU layout regularization (ablation) | Planner3D · SG-FRONT (all rooms) | mSG: 0.90 -> 0.90 | No clear effect |
| add adversarial layout discriminator (ablation) | Planner3D · SG-FRONT (all rooms) | FID: 47.34 -> 49.09 | Hurt |
| add adversarial layout discriminator (ablation) | Planner3D · SG-FRONT (all rooms) | KID: 6.13 -> 8.74 | Hurt |
| add adversarial layout discriminator (ablation) | Planner3D · SG-FRONT (all rooms) | mSG: 0.90 -> 0.91 | Mixed |
| remove layout discriminator (ablation) | CommonScenes · SG-FRONT (all rooms) | FID: 49.89 -> 48.20 | Improved |
| add unified graph encoder (ablation) | CommonScenes · SG-FRONT (all rooms) | FID: 49.89 -> 47.69 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- CommonScenes — CommonScenes: Generating Commonsense 3D Indoor Scenes with Scene Graph Diffusion (NeurIPS 2023) · better on 9 of 10 shared results
- Graph-to-3D — Graph-to-3D: End-to-End Generation and Manipulation of 3D Scenes Using Scene Graphs (ICCV 2021) · better on 9 of 10 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- InstructScene: Instruction-Driven 3D Indoor Scene Synthesis with Semantic Graph Prior (ICLR 2024 · similarity 0.89)
- Hierarchically-Structured Open-Vocabulary Indoor Scene Synthesis with Pre-trained Large Language Model (AAAI 2025 · similarity 0.89)
- FreeScene: Mixed Graph Diffusion for 3D Scene Synthesis from Free Prompts (CVPR 2025 · similarity 0.88)
- DiffuScene: Denoising Diffusion Models for Generative Indoor Scene Synthesis (CVPR 2024 · similarity 0.88)
- LT3SD: Latent Trees for 3D Scene Diffusion (CVPR 2025 · similarity 0.88)
- MMGDreamer: Mixed-Modality Graph for Geometry-Controllable 3D Indoor Scene Generation (AAAI 2025 · similarity 0.88)
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.
