mlsanityData to 7 Oct 2026

PlanCraft: Sketch, Refine, and Furnish for Architect-Inspired Progressive 3D Residential Scene Generation

arXiv 2607.23491 ↗First posted 26 Jul 2026cs.CV cs.CLcited by 1

Not on the map. Why?

Abstract (arXiv)

Two structural insights have been overlooked in automated residential floor plan generation. First, design is inherently progressive. Architects begin with rough strokes and refine them over time, whereas existing methods typically require their conditioning representation to be fully specified before generation, a fundamental mismatch with how design actually works. Second, the 2D floor plan is not an optional intermediate but an irreplaceable spatial contract. Once room boundaries, doors, and windows are fixed, furnishing reduces from open-ended spatial reasoning to bounded constraint satisfaction. Bypassing this contract, as existing 3D systems do by delegating layout to language models, yields overlapping rooms and implausible proportions; directly calling general-purpose language models likewise produces geometrically invalid layouts. Guided by these insights, we present PlanCraft. SketchPlan supplies the missing training signal by replaying the architect's drawing process on 80K real floor plans, producing partial sketches at every completeness level. PlanCraft-Diff progressively sharpens an incomplete sketch into a geometrically precise, vectorizable floor plan through a coarse-to-fine strategy. With the spatial contract established, PlanCraft-Agent then furnishes the scene within well-defined room boundaries. Experiments show that PlanCraft achieves a 61.1% lower FID than the best existing 2D method and surpasses existing 3D systems by 15 points in expert-rated spatial rationality, with a sketch at only 25% completion already outperforming all fully specified baselines.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 19 of its 34 references that are in the map. First posted on arXiv 26 Jul 2026.

Builds on

19 of 34 references.

References · 19

  1. 2017Attention Is All You Needcited by 59,932
  2. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  3. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  4. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  5. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,604
  6. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,049
  7. 2016SGDR: Stochastic Gradient Descent with Warm Restartscited by 4,993
  8. 2022Objaverse: A Universe of Annotated 3D Objectscited by 1,522
  9. 2019Generalized Intersection over Union: A Metric and A Loss for Bounding Box Regressioncited by 1,361
  10. 2017AI2-THOR: An Interactive 3D Environment for Visual AIcited by 850
  11. 2025Mean Flows for One-step Generative Modelingcited by 524
  12. 2022ProcTHOR: Large-Scale Embodied AI Using Procedural Generationcited by 369
7 more
  1. 2023Holodeck: Language Guided Generation of 3D Embodied AI Environmentscited by 257
  2. 2024InstructScene: Instruction-Driven 3D Indoor Scene Synthesis with Semantic Graph Priorcited by 110
  3. 2024I-Design: Personalized LLM Interior Designercited by 87
  4. 2022HouseDiffusion: Vector Floorplan Generation via a Diffusion Model with Discrete and Continuous Denoisingcited by 72
  5. 2023Tell2Design: A Dataset for Language-Guided Floor Plan Generationcited by 19
  6. 2025WorldGen: From Text to Traversable and Interactive 3D Worldscited by 18
  7. 2026MANSION: Multi-floor lANguage-to-3D Scene generatIOn for loNg-horizon taskscited by 8

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 1.

1 paper.

Similar, not cited

Close in wording, no citation either way

  1. 2026Directly from Alpha to Omega: Controllable End-to-End Vector Floor Plan Generation ↗
  2. 2025FloorPlan-DeepSeek (FPDS): A multimodal approach to floorplan generation using vector-based next room prediction
  3. 2026TLC-Plan: A Two-Level Codebook Based Network for End-to-End Vector Floorplan Generation
  4. 2026Agentic Designer: Progressive Multi-Agent Collaboration for Structure-Aware Interior Layout Generation
  5. 2026FuncRoom-Agent: Sequential Feed-Forward 3D Functional Indoor Scene Generation
  6. 2026Generative Floor Plan Design with LLMs via Reinforcement Learning with Verifiable Rewards
  7. 2026Tokenization Allows Multimodal Large Language Models to Understand, Generate and Edit Architectural Floor Plans
  8. 2026HomeWorld: A Unified Floorplan-to-Furnished Framework for Generating Controllable, Densely Interactive Whole-Home Scenes