mlsanityData to 14 Oct 2026

Nerfbusters: Removing Ghostly Artifacts from Casually Captured NeRFs

arXiv 2304.10532 ↗First posted 20 Apr 2023cs.CV cs.AI cs.GRcited by 76

Not on the map. Why?

Abstract (arXiv)

Casually captured Neural Radiance Fields (NeRFs) suffer from artifacts such as floaters or flawed geometry when rendered outside the camera trajectory. Existing evaluation protocols often do not capture these effects, since they usually only assess image quality at every 8th frame of the training capture. To push forward progress in novel-view synthesis, we propose a new dataset and evaluation procedure, where two camera trajectories are recorded of the scene: one used for training, and the other for evaluation. In this more challenging in-the-wild setting, we find that existing hand-crafted regularizers do not remove floaters nor improve scene geometry. Thus, we propose a 3D diffusion-based method that leverages local 3D priors and a novel density-based score distillation sampling loss to discourage artifacts during NeRF optimization. We show that this data-driven prior removes floaters and improves scene geometry for casual captures.

Key facts. Cited by 76 papers in the map (data through 14 Oct 2026). Builds on 40 of its 52 references that are in the map. First posted on arXiv 20 Apr 2023.

Builds on

40 of 52 references.

References · 40

  1. 2020Denoising Diffusion Probabilistic Modelscited by 21,076
  2. 2015U-Net: Convolutional Networks for Biomedical Image Segmentationcited by 19,713
  3. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,863
  4. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,058
  5. 2017Neural Discrete Representation Learningcited by 4,712
  6. 2019Generative Modeling by Estimating Gradients of the Data Distributioncited by 4,101
  7. 2021Improved Denoising Diffusion Probabilistic Modelscited by 3,761
  8. 2022Instant Neural Graphics Primitives with a Multiresolution Hash Encodingcited by 3,723
  9. 2022DreamFusion: Text-to-3D using 2D Diffusioncited by 3,025
  10. 2015ShapeNet: An Information-Rich 3D Model Repositorycited by 2,897
  11. 2019DeepSDF: Learning Continuous Signed Distance Functions for Shape Representationcited by 2,689
  12. 2021Mip-NeRF 360: Unbounded Anti-Aliased Neural Radiance Fieldscited by 2,032
28 more
  1. 2021Mip-NeRF: A Multiscale Representation for Anti-Aliasing Neural Radiance Fieldscited by 1,964
  2. 2021Plenoxels: Radiance Fields without Neural Networkscited by 1,613
  3. 2020pixelNeRF: Neural Radiance Fields from One or Few Imagescited by 1,584
  4. 2019Generating Diverse High-Fidelity Images with VQ-VAE-2cited by 1,387
  5. 2022TensoRF: Tensorial Radiance Fieldscited by 1,280
  6. 2020NeRF in the Wild: Neural Radiance Fields for Unconstrained Photo Collectionscited by 1,232
  7. 2021PlenOctrees for Real-time Rendering of Neural Radiance Fieldscited by 905
  8. 2023K-Planes: Explicit Radiance Fields in Space, Time, and Appearancecited by 746
  9. 2019Local Light Field Fusion: Practical View Synthesis with Prescriptive Sampling Guidelinescited by 672
  10. 2021Common Objects in 3D: Large-Scale Learning and Evaluation of Real-life 3D Category Reconstructioncited by 652
  11. 2021BARF: Bundle-Adjusting Neural Radiance Fieldscited by 642
  12. 2021RegNeRF: Regularizing Neural Radiance Fields for View Synthesis from Sparse Inputscited by 615
  13. 2023Nerfstudio: A Modular Framework for Neural Radiance Field Developmentcited by 600
  14. 2022Score Jacobian Chaining: Lifting Pretrained 2D Diffusion Models for 3D Generationcited by 597
  15. 2020Deep Local Shapes: Learning Local SDF Priors for Detailed 3D Reconstructioncited by 321
  16. 2023RealFusion: 360deg Reconstruction of Any Object from a Single Imagecited by 306
  17. 2020Infinite Nature: Perpetual View Generation of Natural Scenes from a Single Imagecited by 243
  18. 2020Image Matching across Wide Baselines: From Paper to Practicecited by 225
  19. 2022SparseFusion: Distilling View-conditioned Diffusion for 3D Reconstructioncited by 216
  20. 2022AutoSDF: Shape Priors for 3D Completion, Reconstruction and Generationcited by 208
  21. 2022DiffRF: Rendering-Guided 3D Radiance Field Diffusioncited by 204
  22. 2022Monocular Dynamic View Synthesis: A Reality Checkcited by 200
  23. 2023DiffusioNeRF: Regularizing Neural Radiance Fields with Denoising Diffusion Modelscited by 127
  24. 2018Deep Depth Completion of a Single RGB-D Imagecited by 110
  25. 2023RobustNeRF: Ignoring Distractors with Robust Lossescited by 101
  26. 2022InfiniteNature-Zero: Learning Perpetual View Generation of Natural Scenes from Single Imagescited by 82
  27. 2021Self-Supervised Depth Completion for Active Stereocited by 3
  28. 2022SparseFormer: Attention-based Depth Completion Networkcited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 16; 2024: 22; 2025: 18; 2026: 20.

76 papers.

Sort:

By area:

  1. 2023ReconFusion: 3D Reconstruction with Diffusion Priorscited by 318
  2. 2024A Hierarchical 3D Gaussian Representation for Real-Time Rendering of Very Large Datasetscited by 216
  3. 2025Difix3D+: Improving 3D Reconstructions with Single-Step Diffusion Modelscited by 161
  4. 20243DGS-Enhancer: Enhancing Unbounded 3D Gaussian Splatting with View-consistent 2D Diffusion Priorscited by 106
  5. 2023Generative Image Dynamicscited by 94
  6. 2023Bayes' Rays: Uncertainty Quantification for Neural Radiance Fieldscited by 91
  7. 2023NeRFiller: Completing Scenes via Generative 3D Inpaintingcited by 62
  8. 2024SpotlessSplats: Ignoring Distractors in 3D Gaussian Splattingcited by 61
  9. 2025GenFusion: Closing the Loop between Reconstruction and Generation via Videoscited by 41
  10. 2024GSEdit: Efficient Text-Guided Editing of 3D Objects via Gaussian Splattingcited by 37
  11. 2025MILo: Mesh-In-the-Loop Gaussian Splatting for Detailed and Efficient Surface Reconstructioncited by 37
  12. 2025Decompositional Neural Scene Reconstruction with Generative Diffusion Priorcited by 31
38 more
  1. 2025FlowR: Flowing from Sparse to Dense 3D Reconstructionscited by 28
  2. 2024Feat2GS: Probing Visual Foundation Models with Gaussian Splattingcited by 27
  3. 2024NerfBaselines: Consistent and Reproducible Evaluation of Novel View Synthesis Methodscited by 26
  4. 2024Novel View Extrapolation with Video Diffusion Priorscited by 25
  5. 2023SteinDreamer: Variance Reduction for Text-to-3D Score Distillation via Stein Identitycited by 23
  6. 2023Mirror-NeRF: Learning Neural Radiance Fields for Mirrors with Whitted-Style Ray Tracingcited by 22
  7. 2023ProteusNeRF: Fast Lightweight NeRF Editing using 3D-Aware Image Contextcited by 17
  8. 2023InNeRF360: Text-Guided 3D-Consistent Object Inpainting on 360-degree Neural Radiance Fieldscited by 16
  9. 2024SplatFormer: Point Transformer for Robust 3D Gaussian Splattingcited by 16
  10. 2024Robust 3D Gaussian Splatting for Novel View Synthesis in Presence of Distractorscited by 15
  11. 2024DeSplat: Decomposed Gaussian Splatting for Distractor-Free Renderingcited by 15
  12. 2023City-on-Web: Real-time Neural Rendering of Large-scale Scenes on the Webcited by 14
  13. 2023Inpaint3D: 3D Scene Content Generation using 2D Inpainting Diffusioncited by 13
  14. 2024CVT-xRF: Contrastive In-Voxel Transformer for 3D Consistent Radiance Fields from Sparse Inputscited by 13
  15. 2024T-3DGS: Removing Transient Objects for 3D Scene Reconstructioncited by 13
  16. 2024Neural Radiance Field-based Visual Rendering: A Comprehensive Reviewcited by 12
  17. 2026DiffusionHarmonizer: Bridging Neural Reconstruction and Photorealistic Simulation with Online Diffusion Enhancercited by 10
  18. 2024LoopGaussian: Creating 3D Cinemagraph with Multi-view Images via Eulerian Motion Fieldcited by 9
  19. 2026World Reconstruction From Inconsistent Viewscited by 8
  20. 2024NeRF-US: Removing Ultrasound Imaging Artifacts from Neural Radiance Fields in the Wildcited by 7
  21. 2024GaSpCT: Gaussian Splatting for Novel CT Projection View Synthesiscited by 6
  22. 2025PRIMU: Uncertainty Estimation for Novel Views in Gaussian Splatting from Primitive-Based Representations of Error and Coveragecited by 6
  23. 2025ExploreGS: Explorable 3D Scene Reconstruction with Virtual Camera Samplings and Diffusion Priorscited by 6
  24. 2026One-Shot Refiner: Boosting Feed-forward Novel View Synthesis via One-Step Diffusioncited by 6
  25. 2026ArtiFixer: Enhancing and Extending 3D Reconstruction with Auto-Regressive Diffusion Modelscited by 6
  26. 2023Reconstructive Latent-Space Neural Radiance Fields for Efficient 3D Scene Representationscited by 5
  27. 2024CT-NeRF: Incremental Optimizing Neural Radiance Field and Poses with Complex Trajectorycited by 5
  28. 2025Splat and Replace: 3D Reconstruction with Repetitive Elementscited by 5
  29. 2026Uncertainty-driven 3D Gaussian Splatting Active Mapping via Anisotropic Visibility Fieldcited by 5
  30. 2023Dynamic NeRFs for Soccer Scenescited by 4
  31. 2024SALVE: A 3D Reconstruction Benchmark of Wounds from Consumer-grade Videoscited by 4
  32. 2025How Many Tokens Do 3D Point Cloud Transformer Architectures Really Need?cited by 4
  33. 2026SyncFix: Fixing 3D Reconstructions via Multi-View Synchronizationcited by 4
  34. 2025Perspective-aware 3D Gaussian Inpainting with Multi-view Consistencycited by 3
  35. 2026NERFIFY: A Multi-Agent Framework for Turning NeRF Papers into Codecited by 3
  36. 2023Towards Visual Foundational Models of Physical Scenescited by 2
  37. 2024Oblique-MERF: Revisiting and Improving MERF for Oblique Photographycited by 2
  38. 2024Diversity-Driven View Subset Selection for Indoor Novel View Synthesiscited by 2

50 most cited of 76 All 76 (JSON)