HSM: Hierarchical Scene Motifs for Multi-Scale Indoor Scene Generation
Not on the map. Why?
Abstract (arXiv)
Despite advances in indoor 3D scene layout generation, synthesizing scenes with dense object arrangements remains challenging. Existing methods focus on large furniture while neglecting smaller objects, resulting in unrealistically empty scenes. Those that place small objects typically do not honor arrangement specifications, resulting in largely random placement not following the text description. We present Hierarchical Scene Motifs (HSM): a hierarchical framework for indoor scene generation with dense object arrangements across spatial scales. Indoor scenes are inherently hierarchical, with surfaces supporting objects at different scales, from large furniture on floors to smaller objects on tables and shelves. HSM embraces this hierarchy and exploits recurring cross-scale spatial patterns to generate complex and realistic scenes in a unified manner. Our experiments show that HSM outperforms existing methods by generating scenes that better conform to user input across room types and spatial configurations. Project website is available at https://3dlg-hcvc.github.io/hsm .
Key facts. Cited by 24 papers in the map (data through 14 Oct 2026). Builds on 52 of its 136 references that are in the map. First posted on arXiv 21 Mar 2025.
Builds on
52 of 136 references.
References · 52
- 2023GPT-4 Technical Reportcited by 18,396
- 2023BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Modelscited by 6,137
- 2022Objaverse: A Universe of Annotated 3D Objectscited by 1,540
- 2024Evaluating Text-to-Visual Generation with Image-to-Text Generationcited by 474
- 20203D-FRONT: 3D Furnished Rooms with layOuts and semaNTicscited by 380
- 2022ProcTHOR: Large-Scale Embodied AI Using Procedural Generationcited by 380
- 2023LayoutGPT: Compositional Visual Planning and Generation with Large Language Modelscited by 347
- 20203D-FUTURE: 3D Furniture shape with TextUREcited by 311
- 2024Long-CLIP: Unlocking the Long-Text Capability of CLIPcited by 300
- 2023Holodeck: Language Guided Generation of 3D Embodied AI Environmentscited by 268
- 2021ATISS: Autoregressive Transformers for Indoor Scene Synthesiscited by 200
- 2023Habitat Synthetic Scenes Dataset (HSSD-200): An Analysis of 3D Scene Scale and Realism Tradeoffs for ObjectGoal Navigationcited by 167
38 more
- 2023DiffuScene: Denoising Diffusion Models for Generative Indoor Scene Synthesiscited by 166
- 2020SceneFormer: Indoor Scene Generation with Transformerscited by 158
- 2024Infinigen Indoors: Photorealistic Indoor Scenes using Procedural Generationcited by 137
- 2024SceneCraft: An LLM Agent for Synthesizing 3D Scene as Blender Codecited by 134
- 2024PhyScene: Physically Interactable 3D Scene Synthesis for Embodied AIcited by 125
- 2024InstructScene: Instruction-Driven 3D Indoor Scene Synthesis with Semantic Graph Priorcited by 117
- 2018Fast and Flexible Indoor Scene Synthesis via Deep Convolutional Generative Modelscited by 106
- 2024LayoutVLM: Differentiable Optimization of 3D Layout via Vision-Language Modelscited by 96
- 2024I-Design: Personalized LLM Interior Designercited by 87
- 2021Graph-to-3D: End-to-End Generation and Manipulation of 3D Scenes Using Scene Graphscited by 84
- 2023AnyHome: Open-Vocabulary Generation of Structured and Textured 3D Homescited by 82
- 2018Human-centric Indoor Scene Synthesis Using Stochastic Grammarcited by 74
- 2024EchoScene: Indoor Scene Generation via Information Echo over Scene Graph Diffusioncited by 67
- 2018Deep Generative Modeling for Scene Synthesis via Hybrid Representationscited by 63
- 2023CommonScenes: Generating Commonsense 3D Indoor Scenes with Scene Graph Diffusioncited by 54
- 2024Open-Universe Indoor Scene Generation using LLM Program Synthesis and Uncurated Object Databasescited by 45
- 2023SceneHGN: Hierarchical Graph Networks for 3D Indoor Scene Generation with Fine-Grained Geometrycited by 39
- 2018GRAINS: Generative Recursive Autoencoders for INdoor Scenescited by 38
- 2024Architect: Generating Vivid and Interactive 3D Scenes with Hierarchical 2D Inpaintingcited by 38
- 2024Mixed Diffusion for 3D Indoor Scene Synthesiscited by 32
- 2025SceneEval: Evaluating Semantic Coherence in Text-Conditioned 3D Indoor Scene Synthesiscited by 26
- 2025FirePlace: Geometric Refinements of LLM Common Sense Reasoning for 3D Object Placementcited by 25
- 2025FlairGPT: Repurposing LLMs for Interior Designscited by 24
- 2020General 3D Room Layout from a Single View by Render-and-Comparecited by 21
- 2024Chat2Layout: Interactive 3D Furniture Layout with a Multimodal LLMcited by 21
- 2025WorldCraft: Photo-Realistic 3D World Creation and Customization via LLM Agentscited by 21
- 2025MMGDreamer: Mixed-Modality Graph for Geometry-Controllable 3D Indoor Scene Generationcited by 19
- 2024SceneMotifCoder: Example-driven Visual Program Learning for Generating 3D Object Arrangementscited by 17
- 20233D Copy-Paste: Physically Plausible Object Insertion for Monocular 3D Detectioncited by 15
- 2022COFS: Controllable Furniture layout Synthesiscited by 14
- 2024SceneGenAgent: Precise Industrial Scene Generation with Coding Agentcited by 13
- 2024Forest2Seq: Revitalizing Order Prior for Sequential Indoor Scene Synthesiscited by 12
- 2020SceneGen: Generative Contextual Scene Augmentation using Scene Graph Priorscited by 11
- 2024Diorama: Unleashing Zero-shot Single-view 3D Indoor Scene Modelingcited by 11
- 2025CHOrD: Generation of Collision-Free, House-Scale, and Organized Digital Twins for 3D Indoor Scenes with Controllable Floor Plans and Optimal Layoutscited by 10
- 2024Planner3D: LLM-enhanced graph prior meets 3D indoor scene explicit regularizationcited by 9
- 2024External Knowledge Enhanced 3D Scene Generation from Sketchcited by 9
- 2017Automatic Generation of Constrained Furniture Layoutscited by 8
50 most cited of 52 All 52 (JSON)
Built on by
Citing papers per year: 2025: 5; 2026: 19.
24 papers.
Sort:
By area:
- 2026SceneCode: Executable World Programs for Editable Indoor Scenes with Articulated Objectscited by 11
- 2025Steerable Scene Generation with Post Training and Inference-Time Searchcited by 9
- 2026WorldClaw: Agentic 3D Open-World Generation at Scalecited by 8
- 2025Video Perception Models for 3D Scene Synthesiscited by 7
- 2026PhyScensis: Physics-Augmented LLM Agents for Complex Physical Scene Arrangementcited by 7
- 2025DisCo-Layout: Disentangling and Coordinating Semantic and Physical Refinement in a Multi-Agent Framework for 3D Indoor Layout Synthesiscited by 6
- 2026AnchoredDream: Zero-Shot 360deg Indoor Scene Generation from a Single View via Geometric Groundingcited by 5
- 2026MUSE: Agentic 3D Scene Authoring via Memory-Grounded Incremental Requirement Satisfactioncited by 3
- 2026SceneNAT: Masked Generative Modeling for Language-Guided Indoor Scene Synthesiscited by 2
- 2026HDSL: A Hierarchical Domain-Specific Language for Structured 3D Indoor Scene Generation and Localized Editing with LLM Agentscited by 2
- 2026Beyond Placement and Articulation: Usage-Driven Code Scenes for Embodied Interactioncited by 2
- 2025GeoSceneGraph: Geometric Scene Graph Diffusion Model for Text-guided 3D Indoor Scene Synthesiscited by 1
12 more
- 2026Pair2Scene: Learning Local Object Relations for Procedural Scene Generationcited by 1
- 2026Co-generation of Layout and Shape from Text via Autoregressive 3D Diffusioncited by 1
- 2026Function2Scene: 3D Indoor Scene Layout from Functional Specificationscited by 1
- 2026Global Graph-Validated Optimization for VLM-based 3D Indoor Scene Generationcited by 1
- 2025RoomPilot: Controllable Indoor Scene Synthesis via Multimodal Semantic Parsingcited by 0
- 2026SDesc3D: Towards Layout-Aware 3D Indoor Scene Generation from Short Descriptionscited by 0
- 2026ThinkBLOX: 3D Indoor Scene Generation with Progressive Reasoningcited by 0
- 2026Learning an Interior Layout Policy in a Domain Specific Language Action Spacecited by 0
- 2026SceneMosaic: Efficient and Diverse Simulation-Ready Scene Generation via Hybrid Agentic Layout Evolutioncited by 0
- 2026WorldWeave: Growing Persistent Geometric Worlds for Video Generationcited by 0
- 2026Frozen Scenes, Shifting Winners: Configuration Fragility in Text-to-3D Evaluationcited by 0
- 2026EmbodiedSmith: Scaling Embodied Data through Recursive Self-Improvement Flywheel in Simulationcited by 0
- 2026EmbodiedSmith: Scaling Embodied Data through Recursive Self-Improvement Flywheel in Simulation6 Oct 2026
- 2026Frozen Scenes, Shifting Winners: Configuration Fragility in Text-to-3D Evaluation30 Sep 2026
- 2026WorldWeave: Growing Persistent Geometric Worlds for Video Generation28 Sep 2026
- 2026SceneMosaic: Efficient and Diverse Simulation-Ready Scene Generation via Hybrid Agentic Layout Evolution4 Sep 2026
- 2026Beyond Placement and Articulation: Usage-Driven Code Scenes for Embodied Interaction19 Aug 2026
- 2026WorldClaw: Agentic 3D Open-World Generation at Scale5 Aug 2026
- 2026Global Graph-Validated Optimization for VLM-based 3D Indoor Scene Generation4 Aug 2026
- 2026Learning an Interior Layout Policy in a Domain Specific Language Action Space31 Jul 2026
- 2026ThinkBLOX: 3D Indoor Scene Generation with Progressive Reasoning15 Jul 2026
- 2026MUSE: Agentic 3D Scene Authoring via Memory-Grounded Incremental Requirement Satisfaction12 Jun 2026
- 2026HDSL: A Hierarchical Domain-Specific Language for Structured 3D Indoor Scene Generation and Localized Editing with LLM Agents8 Jun 2026
- 2026Function2Scene: 3D Indoor Scene Layout from Functional Specifications29 May 2026
12 more
- 2026SceneCode: Executable World Programs for Editable Indoor Scenes with Articulated Objects19 May 2026
- 2026Co-generation of Layout and Shape from Text via Autoregressive 3D Diffusion17 Apr 2026
- 2026Pair2Scene: Learning Local Object Relations for Procedural Scene Generation13 Apr 2026
- 2026SDesc3D: Towards Layout-Aware 3D Indoor Scene Generation from Short Descriptions2 Apr 2026
- 2026PhyScensis: Physics-Augmented LLM Agents for Complex Physical Scene Arrangement16 Feb 2026
- 2026AnchoredDream: Zero-Shot 360deg Indoor Scene Generation from a Single View via Geometric Grounding23 Jan 2026
- 2026SceneNAT: Masked Generative Modeling for Language-Guided Indoor Scene Synthesis12 Jan 2026
- 2025RoomPilot: Controllable Indoor Scene Synthesis via Multimodal Semantic Parsing12 Dec 2025
- 2025GeoSceneGraph: Geometric Scene Graph Diffusion Model for Text-guided 3D Indoor Scene Synthesis18 Nov 2025
- 2025DisCo-Layout: Disentangling and Coordinating Semantic and Physical Refinement in a Multi-Agent Framework for 3D Indoor Layout Synthesis2 Oct 2025
- 2025Video Perception Models for 3D Scene Synthesis25 Jun 2025
- 2025Steerable Scene Generation with Post Training and Inference-Time Search7 May 2025
- 2026WorldClaw: Agentic 3D Open-World Generation at Scalecited by 8
- 2026PhyScensis: Physics-Augmented LLM Agents for Complex Physical Scene Arrangementcited by 7
- 2025DisCo-Layout: Disentangling and Coordinating Semantic and Physical Refinement in a Multi-Agent Framework for 3D Indoor Layout Synthesiscited by 6
- 2026MUSE: Agentic 3D Scene Authoring via Memory-Grounded Incremental Requirement Satisfactioncited by 3
- 2026HDSL: A Hierarchical Domain-Specific Language for Structured 3D Indoor Scene Generation and Localized Editing with LLM Agentscited by 2
- 2026SceneMosaic: Efficient and Diverse Simulation-Ready Scene Generation via Hybrid Agentic Layout Evolutioncited by 0
- 2026SceneMosaic: Efficient and Diverse Simulation-Ready Scene Generation via Hybrid Agentic Layout Evolution4 Sep 2026
- 2026WorldClaw: Agentic 3D Open-World Generation at Scale5 Aug 2026
- 2026MUSE: Agentic 3D Scene Authoring via Memory-Grounded Incremental Requirement Satisfaction12 Jun 2026
- 2026HDSL: A Hierarchical Domain-Specific Language for Structured 3D Indoor Scene Generation and Localized Editing with LLM Agents8 Jun 2026
- 2026PhyScensis: Physics-Augmented LLM Agents for Complex Physical Scene Arrangement16 Feb 2026
- 2025DisCo-Layout: Disentangling and Coordinating Semantic and Physical Refinement in a Multi-Agent Framework for 3D Indoor Layout Synthesis2 Oct 2025
- 2026SceneCode: Executable World Programs for Editable Indoor Scenes with Articulated Objectscited by 11
- 2025Video Perception Models for 3D Scene Synthesiscited by 7
- 2026AnchoredDream: Zero-Shot 360deg Indoor Scene Generation from a Single View via Geometric Groundingcited by 5
- 2026SceneNAT: Masked Generative Modeling for Language-Guided Indoor Scene Synthesiscited by 2
- 2026Beyond Placement and Articulation: Usage-Driven Code Scenes for Embodied Interactioncited by 2
- 2026Pair2Scene: Learning Local Object Relations for Procedural Scene Generationcited by 1
- 2026Co-generation of Layout and Shape from Text via Autoregressive 3D Diffusioncited by 1
- 2026Function2Scene: 3D Indoor Scene Layout from Functional Specificationscited by 1
- 2026Global Graph-Validated Optimization for VLM-based 3D Indoor Scene Generationcited by 1
- 2025RoomPilot: Controllable Indoor Scene Synthesis via Multimodal Semantic Parsingcited by 0
- 2026SDesc3D: Towards Layout-Aware 3D Indoor Scene Generation from Short Descriptionscited by 0
- 2026Learning an Interior Layout Policy in a Domain Specific Language Action Spacecited by 0
12 most cited of 13 All 13 (JSON)
- 2026Frozen Scenes, Shifting Winners: Configuration Fragility in Text-to-3D Evaluation30 Sep 2026
- 2026Beyond Placement and Articulation: Usage-Driven Code Scenes for Embodied Interaction19 Aug 2026
- 2026Global Graph-Validated Optimization for VLM-based 3D Indoor Scene Generation4 Aug 2026
- 2026Learning an Interior Layout Policy in a Domain Specific Language Action Space31 Jul 2026
- 2026Function2Scene: 3D Indoor Scene Layout from Functional Specifications29 May 2026
- 2026SceneCode: Executable World Programs for Editable Indoor Scenes with Articulated Objects19 May 2026
- 2026Co-generation of Layout and Shape from Text via Autoregressive 3D Diffusion17 Apr 2026
- 2026Pair2Scene: Learning Local Object Relations for Procedural Scene Generation13 Apr 2026
- 2026SDesc3D: Towards Layout-Aware 3D Indoor Scene Generation from Short Descriptions2 Apr 2026
- 2026AnchoredDream: Zero-Shot 360deg Indoor Scene Generation from a Single View via Geometric Grounding23 Jan 2026
- 2026SceneNAT: Masked Generative Modeling for Language-Guided Indoor Scene Synthesis12 Jan 2026
- 2025RoomPilot: Controllable Indoor Scene Synthesis via Multimodal Semantic Parsing12 Dec 2025
12 newest of 13 All 13 (JSON)