mlsanityData to 14 Oct 2026

SplatPainter: Interactive Authoring of 3D Gaussians from 2D Edits via Test-Time Training

arXiv 2512.05354 ↗First posted 5 Dec 2025cs.CV cs.GRcited by 7

Not on the map. Why?

Abstract (arXiv)

The rise of 3D Gaussian Splatting has revolutionized photorealistic 3D asset creation, yet a critical gap remains for their interactive refinement and editing. Existing approaches based on diffusion or optimization are ill-suited for this task, as they are often prohibitively slow, destructive to the original asset's identity, or lack the precision for fine-grained control. To address this, we introduce SplatPainter, a state-aware feedforward model that enables continuous editing of 3D Gaussian assets from user-provided 2D view(s). Our method directly predicts updates to the attributes of a compact, feature-rich Gaussian representation and leverages Test-Time Training to create a state-aware, iterative workflow. The versatility of our approach allows a single architecture to perform diverse tasks, including high-fidelity local detail refinement, local paint-over, and consistent global recoloring, all at interactive speeds, paving the way for fluid and intuitive 3D content authoring.

Key facts. Cited by 7 papers in the map (data through 14 Oct 2026). Builds on 85 of its 106 references that are in the map. First posted on arXiv 5 Dec 2025.

Builds on

85 of 106 references.

References · 85

  1. 2017Attention Is All You Needcited by 60,361
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  3. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,177
  4. 20233D Gaussian Splatting for Real-Time Radiance Field Renderingcited by 6,226
  5. 2022DreamFusion: Text-to-3D using 2D Diffusioncited by 3,025
  6. 2023FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioningcited by 2,598
  7. 2020GLU Variants Improve Transformercited by 1,804
  8. 2022Objaverse: A Universe of Annotated 3D Objectscited by 1,540
  9. 2021Perceiver: General Perception with Iterative Attentioncited by 1,104
  10. 2015ORB-SLAM: a Versatile and Accurate Monocular SLAM Systemcited by 988
  11. 2023LRM: Large Reconstruction Model for Single Image to 3Dcited by 849
  12. 2024Structured 3D Latents for Scalable and Versatile 3D Generationcited by 770
38 more
  1. 2021DROID-SLAM: Deep Visual SLAM for Monocular, Stereo, and RGB-D Camerascited by 687
  2. 20163D-R2N2: A Unified Approach for Single and Multi-view 3D Object Reconstructioncited by 588
  3. 2024MVSplat: Efficient 3D Gaussian Splatting from Sparse Multi-View Imagescited by 566
  4. 2025Continuous 3D Perception Model with Persistent Statecited by 548
  5. 2021Linear Transformers Are Secretly Fast Weight Programmerscited by 502
  6. 2023Instruct-NeRF2NeRF: Editing 3D Scenes with Instructionscited by 484
  7. 2023PhysGaussian: Physics-Integrated 3D Gaussians for Generative Dynamicscited by 377
  8. 2022Decomposing NeRF for Editing via Feature Field Distillationcited by 371
  9. 2021CLIP-NeRF: Text-and-Image Driven Manipulation of Neural Radiance Fieldscited by 370
  10. 2023Feature 3DGS: Supercharging 3D Gaussian Splatting to Enable Distilled Feature Fieldscited by 364
  11. 2024SV3D: Novel Multi-view Synthesis and 3D Generation from a Single Image using Latent Video Diffusioncited by 364
  12. 2024Learning to (Learn at Test Time): RNNs with Expressive Hidden Statescited by 355
  13. 2024GS-LRM: Large Reconstruction Model for 3D Gaussian Splattingcited by 349
  14. 2023GaussianEditor: Swift and Controllable 3D Editing with Gaussian Splattingcited by 335
  15. 2021Text2Mesh: Text-Driven Neural Stylization for Meshescited by 334
  16. 2024Titans: Learning to Memorize at Test Timecited by 328
  17. 2024GRM: Large Gaussian Reconstruction Model for Efficient 3D Reconstruction and Generationcited by 301
  18. 20243D Reconstruction with Spatial Memorycited by 277
  19. 2021NeuralRecon: Real-Time Coherent 3D Reconstruction from Monocular Videocited by 241
  20. 2017Learning a Multi-View Stereo Machinecited by 194
  21. 2023NICER-SLAM: Neural Implicit Scene Encoding for RGB SLAMcited by 164
  22. 2023DreamEditor: Text-Driven 3D Scene Editing with Neural Fieldscited by 155
  23. 2025Test-Time Training Done Rightcited by 130
  24. 2025TTT3R: 3D Reconstruction as Test-Time Trainingcited by 129
  25. 2022NeRF-Art: Text-Driven Neural Radiance Fields Stylizationcited by 126
  26. 2023Vox-E: Text-guided Voxel Editing of 3D Objectscited by 115
  27. 2024GaussCtrl: Multi-View Consistent Text-Driven 3D Gaussian Splatting Editingcited by 115
  28. 20243D Gaussian as a New Era: A Surveycited by 104
  29. 2022NeRFusion: Fusing Radiance Fields for Large-Scale Scene Reconstructioncited by 100
  30. 2019Kaolin: A PyTorch Library for Accelerating 3D Deep Learning Researchcited by 98
  31. 2023TextDeformer: Geometry Manipulation using Text Guidancecited by 94
  32. 2024DGE: Direct Gaussian 3D Editing by Consistent Multi-view Editingcited by 94
  33. 2022TANGO: Text-driven Photorealistic and Robust 3D Stylization via Lighting Decompositioncited by 83
  34. 2024TIP-Editor: An Accurate 3D Editor Following Both Text-Prompts And Image-Promptscited by 77
  35. 2025Test-time regression: a unifying framework for designing sequence models with associative memorycited by 74
  36. 2025It's All Connected: A Journey Through Test-Time Memorization, Attentional Bias, Retention, and Online Optimizationcited by 64
  37. 2024Advances in 3D Generation: A Surveycited by 63
  38. 2023Instruct 3D-to-3D: Text Instruction Guided 3D-to-3D conversioncited by 48

50 most cited of 85 All 85 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 7.

7 papers.

  1. 2026VecSet-Edit: Unleashing Pre-trained LRM for Mesh Editing from Single Imagecited by 4
  2. 2026A Survey on Self-Improving Test-Time Intelligence: Feedback-Driven Adapting, Learning, and Scaling at Inferencecited by 3
  3. 2026UMI3D: Robust 3D Generation on Unconstrained Multi-Image Inputs via Simultaneous Focus Cross-Attention Routingcited by 2
  4. 2026SceneExpander: Text-Guided 3D Scene Expansion via Free-Form View Insertioncited by 0
  5. 2026Information-Regularized Constrained Inversion for Stable Avatar Editing from Sparse Supervisioncited by 0
  6. 2026LightBridge: Feed-Forward Generative Relighting for 3D Gaussian Splattingcited by 0
  7. 2026SCION: Scene Composition with Instanced Neural Primitivescited by 0