Research map / 3D vision / 3D generation
Orchestrating Spatial Semantics via a Zone-Graph Paradigm for Intricate Indoor Scene Generation
Meisheng Zhang, Shizhao Sun, Yang Zhao, Ziyuan Liu, Zhijun Gao, Jiang Bian
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 Zone-Graph derivation reasoning (ablation) | ZoneMaestro SFT · SCALE | asset count: 38.45 -> 30.96 | Improved |
| add Zone-Graph derivation reasoning (ablation) | ZoneMaestro SFT · SCALE | OOB: 0.25 -> 0.21 | Improved |
| add Zone-Graph derivation reasoning (ablation) | ZoneMaestro SFT · SCALE | collision volume: 0.18 -> 0.13 | Improved |
| add Z-GRPO after Zone-Graph SFT (ablation) | ZoneMaestro · SCALE | GPT-4o-mini Avg: 6.84 -> 6.85 | Improved |
| add Z-GRPO after Zone-Graph SFT (ablation) | ZoneMaestro · SCALE | OOB: 0.21 -> 0.11 | Improved |
| use two alternating cycles instead of Z-GRPO only (ablation) | ZoneMaestro · SCALE | Realism: 4.14 -> 4.95 | Improved |
| use two alternating cycles instead of one (ablation) | ZoneMaestro · SCALE | GPT-4o-mini Avg: 6.75 -> 6.96 | Improved |
| use two alternating cycles instead of one (ablation) | ZoneMaestro · SCALE | OOB: 0.14 -> 0.09 | Improved |
| use two alternating cycles instead of Z-GRPO only (ablation) | ZoneMaestro · SCALE | Structure score: 6.28 -> 5.19 | Hurt |
| use two alternating cycles instead of Z-GRPO only (ablation) | ZoneMaestro · SCALE | asset count: 21.35 -> 23.35 | 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.
- DiffuScene — DiffuScene: Denoising Diffusion Models for Generative Indoor Scene Synthesis (CVPR 2024) · better on 4 of 4 shared results
- Holodeck — Holodeck: Language Guided Generation of 3D Embodied AI Environments (CVPR 2024) · better on 4 of 4 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
- Direct Numerical Layout Generation for 3D Indoor Scene Synthesis via Spatial Reasoning (NeurIPS 2025 · similarity 0.90)
- InstructScene: Instruction-Driven 3D Indoor Scene Synthesis with Semantic Graph Prior (ICLR 2024 · similarity 0.88)
- SceneWeaver: All-in-One 3D Scene Synthesis with an Extensible and Self-Reflective Agent (NeurIPS 2025 · similarity 0.88)
- OptiScene: LLM-driven Indoor Scene Layout Generation via Scaled Human-aligned Data Synthesis and Multi-Stage Preference Optimization (NeurIPS 2025 · similarity 0.87)
- Thinking in Structures: Evaluating Spatial Intelligence in Constraint-Governed Spaces (ICML 2026 · similarity 0.87)
- LayoutVLM: Differentiable Optimization of 3D Layout via Vision-Language Models (CVPR 2025 · similarity 0.87)
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.
