mlsanityData to 14 Oct 2026

SceneMaker: Open-set 3D Scene Generation with Decoupled De-occlusion and Pose Estimation Model

arXiv 2512.10957 ↗First posted 11 Dec 2025cs.CV cs.AIcited by 17

Not on the map. Why?

Abstract (arXiv)

We propose a decoupled 3D scene generation framework called SceneMaker in this work. Due to the lack of sufficient open-set de-occlusion and pose estimation priors, existing methods struggle to simultaneously produce high-quality geometry and accurate poses under severe occlusion and open-set settings. To address these issues, we first decouple the de-occlusion model from 3D object generation, and enhance it by leveraging image datasets and collected de-occlusion datasets for much more diverse open-set occlusion patterns. Then, we propose a unified pose estimation model that integrates global and local mechanisms for both self-attention and cross-attention to improve accuracy. Besides, we construct an open-set 3D scene dataset to further extend the generalization of the pose estimation model. Comprehensive experiments demonstrate the superiority of our decoupled framework on both indoor and open-set scenes. Our codes and datasets is released at https://idea-research.github.io/SceneMaker/.

Key facts. Cited by 17 papers in the map (data through 14 Oct 2026). Builds on 55 of its 59 references that are in the map. First posted on arXiv 11 Dec 2025.

Builds on

55 of 59 references.

References · 55

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  3. 2023GPT-4 Technical Reportcited by 18,396
  4. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,974
  5. 2022Scalable Diffusion Models with Transformerscited by 6,054
  6. 2022Flow Matching for Generative Modelingcited by 5,702
  7. 2021RoFormer: Enhanced Transformer with Rotary Position Embeddingcited by 4,703
  8. 2022LAION-5B: An open large-scale dataset for training next generation image-text modelscited by 4,152
  9. 2023Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detectioncited by 3,332
  10. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,117
  11. 2023DUSt3R: Geometric 3D Vision Made Easycited by 1,567
  12. 2022Objaverse: A Universe of Annotated 3D Objectscited by 1,540
38 more
  1. 2025VGGT: Visual Geometry Grounded Transformercited by 1,539
  2. 2025FLUX.1 Kontext: Flow Matching for In-Context Image Generation and Editing in Latent Spacecited by 1,006
  3. 2024Grounded SAM: Assembling Open-World Models for Diverse Visual Taskscited by 929
  4. 2024Structured 3D Latents for Scalable and Versatile 3D Generationcited by 770
  5. 20213D Shape Generation and Completion through Point-Voxel Diffusioncited by 516
  6. 2019Normalized Object Coordinate Space for Category-Level 6D Object Pose and Size Estimationcited by 495
  7. 2024CLAY: A Controllable Large-scale Generative Model for Creating High-quality 3D Assetscited by 455
  8. 2023FoundationPose: Unified 6D Pose Estimation and Tracking of Novel Objectscited by 429
  9. 20203D-FRONT: 3D Furnished Rooms with layOuts and semaNTicscited by 380
  10. 2025Hunyuan3D 2.0: Scaling Diffusion Models for High Resolution Textured 3D Assets Generationcited by 356
  11. 2021Neural RGB-D Surface Reconstructioncited by 313
  12. 2024MoGe: Unlocking Accurate Monocular Geometry Estimation for Open-Domain Images with Optimal Training Supervisioncited by 299
  13. 2025TripoSG: High-Fidelity 3D Shape Synthesis using Large-Scale Rectified Flow Modelscited by 213
  14. 2024Direct3D: Scalable Image-to-3D Generation via 3D Latent Diffusion Transformercited by 210
  15. 2024BrushNet: A Plug-and-Play Image Inpainting Model with Decomposed Dual-Branch Diffusioncited by 202
  16. 2022MegaPose: 6D Pose Estimation of Novel Objects via Render & Comparecited by 201
  17. 2020Total3DUnderstanding: Joint Layout, Object Pose and Mesh Reconstruction for Indoor Scenes from a Single Imagecited by 172
  18. 2023DiffuScene: Denoising Diffusion Models for Generative Indoor Scene Synthesiscited by 166
  19. 2020Shape Prior Deformation for Categorical 6D Object Pose and Size Estimationcited by 161
  20. 2022Omni3D: A Large Benchmark and Model for 3D Object Detection in the Wildcited by 160
  21. 2024MIDI: Multi-Instance Diffusion for Single Image to 3D Scene Generationcited by 117
  22. 2025CAST: Component-Aligned 3D Scene Reconstruction from an RGB Imagecited by 106
  23. 2024Dora: Sampling and Benchmarking for 3D Shape Variational Auto-Encoderscited by 98
  24. 2024Automated Creation of Digital Cousins for Robust Policy Learningcited by 91
  25. 2025Step1X-3D: Towards High-Fidelity and Controllable Generation of Textured 3D Assetscited by 87
  26. 2025PartCrafter: Structured 3D Mesh Generation via Compositional Latent Diffusion Transformerscited by 84
  27. 2022OSOP: A Multi-Stage One Shot Object Pose Estimation Frameworkcited by 81
  28. 2021Panoptic 3D Scene Reconstruction From a Single RGB Imagecited by 70
  29. 2024Gen3DSR: Generalizable 3D Scene Reconstruction via Divide and Conquer from a Single Viewcited by 61
  30. 2025Amodal3R: Amodal 3D Reconstruction from Occluded 2D Imagescited by 54
  31. 2025SceneGen: Single-Image 3D Scene Generation in One Feedforward Passcited by 54
  32. 2024Omni6DPose: A Benchmark and Model for Universal 6D Object Pose Estimation and Trackingcited by 50
  33. 2023GenPose: Generative Category-level Object Pose Estimation via Diffusion Modelscited by 48
  34. 2023DiffCAD: Weakly-Supervised Probabilistic CAD Model Retrieval and Alignment from an RGB Imagecited by 48
  35. 2024REPARO: Compositional 3D Assets Generation with Differentiable 3D Layout Alignmentcited by 45
  36. 2022Towards High-Fidelity Single-view Holistic Reconstruction of Indoor Scenescited by 44
  37. 2022SSP-Pose: Symmetry-Aware Shape Prior Deformation for Direct Category-Level Object Pose Estimationcited by 41
  38. 2023Single-view 3D Scene Reconstruction with High-fidelity Shape and Texturecited by 37

50 most cited of 55 All 55 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 17.

17 papers.

Sort:

By area:

  1. 2026WorldClaw: Agentic 3D Open-World Generation at Scalecited by 8
  2. 2026One Video, One World: Turning Monocular Video into Physical 4D Scenescited by 6
  3. 2026φ-Scene: Physically Grounded Image-to-3D Scene Reconstructioncited by 3
  4. 2026WorldAct: Activating Monolithic 3D Worlds into Interactive-Ready Object-Centric Scenescited by 2
  5. 2026SceneConductor: 3D Scene Generation from a Single Image with Multi-Agent Orchestrationcited by 2
  6. 2026WorldSculpt: Generating Compositional Worlds from Grounded Videoscited by 2
  7. 2026GASE: Gaussian Splatting-Based Automated System for Reconstructing Embodied-Simulation Environmentscited by 1
  8. 2026Scenix: Sparse-View 3D Scene Reconstruction via Executable Scene Programscited by 1
  9. 2026StateFlow: Building, Evolving, and Accessing 3D World States for Previsualizationcited by 1
  10. 2026Can Single-View Mesh Reconstruction Generalize to Robot Camera Rotation?cited by 0
  11. 2026Scene-SAM3D: Multi-View Scene Asset Generation Without Fine-Tuningcited by 0
  12. 2026SceneReGen: Generative Reconstruction of 3D Scenes from a Single Imagecited by 0
5 more
  1. 2026Mira-Scene: Pixel-Aligned Layouts for Generative 3D Scene Reconstructioncited by 0
  2. 2026Fysiverse-3D-Vision Technical Report: Generating Executable 3D Worlds from Images through Unified Spatial Reasoningcited by 0
  3. 2026CoDimRecon: Agentic Reconstruction of Sim-Ready 3D Scenes with Deformable Curves, Surfaces, and Volumescited by 0
  4. 2026DistScene: Object-to-Scene Distillation for 3D Scene Generationcited by 0
  5. 2026ScenePlacer: Reformulating Scene-Level Object Pose Estimation as Visibility Field Learningcited by 0