mlsanityData to 7 Oct 2026

SPARF: Neural Radiance Fields from Sparse and Noisy Poses

arXiv 2211.11738 ↗First posted 21 Nov 2022cs.CVcited by 195

Not on the map. Why?

Abstract (arXiv)

Neural Radiance Field (NeRF) has recently emerged as a powerful representation to synthesize photorealistic novel views. While showing impressive performance, it relies on the availability of dense input views with highly accurate camera poses, thus limiting its application in real-world scenarios. In this work, we introduce Sparse Pose Adjusting Radiance Field (SPARF), to address the challenge of novel-view synthesis given only few wide-baseline input images (as low as 3) with noisy camera poses. Our approach exploits multi-view geometry constraints in order to jointly learn the NeRF and refine the camera poses. By relying on pixel matches extracted between the input views, our multi-view correspondence objective enforces the optimized scene and camera poses to converge to a global and geometrically accurate solution. Our depth consistency loss further encourages the reconstructed scene to be consistent from any viewpoint. Our approach sets a new state of the art in the sparse-view regime on multiple challenging datasets.

Key facts. Cited by 195 papers in the map (data through 7 Oct 2026). Builds on 50 of its 74 references that are in the map. First posted on arXiv 21 Nov 2022.

Builds on

50 of 74 references.

References · 50

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,029
  3. 2022Instant Neural Graphics Primitives with a Multiresolution Hash Encodingcited by 3,698
  4. 2021Mip-NeRF 360: Unbounded Anti-Aliased Neural Radiance Fieldscited by 2,013
  5. 2021Mip-NeRF: A Multiscale Representation for Anti-Aliasing Neural Radiance Fieldscited by 1,957
  6. 2020pixelNeRF: Neural Radiance Fields from One or Few Imagescited by 1,574
  7. 2019SuperGlue: Learning Feature Matching with Graph Neural Networkscited by 1,437
  8. 2017SuperPoint: Self-Supervised Interest Point Detection and Descriptioncited by 1,313
  9. 2018On the Continuity of Rotation Representations in Neural Networkscited by 1,186
  10. 2019The Replica Dataset: A Digital Replica of Indoor Spacescited by 885
  11. 2021MVSNeRF: Fast Generalizable Radiance Field Reconstruction from Multi-View Stereocited by 735
  12. 2021Depth-supervised NeRF: Fewer Views and Faster Training for Freecited by 732
38 more
  1. 2021IBRNet: Learning Multi-View Image-Based Renderingcited by 698
  2. 2021BARF: Bundle-Adjusting Neural Radiance Fieldscited by 641
  3. 2021RegNeRF: Regularizing Neural Radiance Fields for View Synthesis from Sparse Inputscited by 614
  4. 2021NICE-SLAM: Neural Implicit Scalable Encoding for SLAMcited by 586
  5. 2021iMAP: Implicit Mapping and Positioning in Real-Timecited by 476
  6. 2022MonoSDF: Exploring Monocular Geometric Cues for Neural Implicit Surface Reconstructioncited by 467
  7. 2021Putting NeRF on a Diet: Semantically Consistent Few-Shot View Synthesiscited by 464
  8. 2020INeRF: Inverting Neural Radiance Fields for Pose Estimationcited by 381
  9. 2021Neural RGB-D Surface Reconstructioncited by 312
  10. 2021Dense Depth Priors for Neural Radiance Fields from Sparse Input Viewscited by 301
  11. 2021Self-Calibrating Neural Radiance Fieldscited by 204
  12. 2021Neural Rays for Occlusion-aware Image-based Renderingcited by 192
  13. 2021NerfingMVS: Guided Optimization of Neural Radiance Fields for Indoor Multi-view Stereocited by 191
  14. 2021Stereo Radiance Fields (SRF): Learning View Synthesis for Sparse Views of Novel Scenescited by 190
  15. 2020GRF: Learning a General Radiance Field for 3D Representation and Renderingcited by 183
  16. 2021InfoNeRF: Ray Entropy Minimization for Few-Shot Neural Volume Renderingcited by 180
  17. 2022SparseNeuS: Fast Generalizable Neural Surface Reconstruction from Sparse Viewscited by 175
  18. 2021NeRF--: Neural Radiance Fields Without Known Camera Parameterscited by 155
  19. 2021GNeRF: GAN-based Neural Radiance Field without Posed Cameracited by 153
  20. 2019GLU-Net: Global-Local Universal Network for Dense Flow and Correspondencescited by 139
  21. 2021NeRS: Neural Reflectance Surfaces for Sparse-view 3D Reconstruction in the Wildcited by 136
  22. 2021Improving neural implicit surfaces geometry with patch warpingcited by 136
  23. 2021Learning Accurate Dense Correspondences and When to Trust Themcited by 113
  24. 2022RelPose: Predicting Probabilistic Relative Rotation for Single Objects in the Wildcited by 102
  25. 2022SAMURAI: Shape And Material from Unconstrained Real-world Arbitrary Image collectionscited by 89
  26. 2022GARF: Gaussian Activated Radiance Fields for High Fidelity Reconstruction and Pose Estimationcited by 86
  27. 2021PDC-Net+: Enhanced Probabilistic Dense Correspondence Networkcited by 80
  28. 2022PS-NeRF: Neural Inverse Rendering for Multi-view Photometric Stereocited by 69
  29. 2021Wide-Baseline Relative Camera Pose Estimation with Directional Learningcited by 49
  30. 2022SiNeRF: Sinusoidal Neural Radiance Fields for Joint Pose Estimation and Scene Reconstructioncited by 49
  31. 2017Relative Camera Pose Estimation Using Convolutional Neural Networkscited by 47
  32. 2022NeROIC: Neural Rendering of Objects from Online Image Collectionscited by 36
  33. 2018RPNet: an End-to-End Network for Relative Camera Pose Estimationcited by 31
  34. 2019GLAMpoints: Greedily Learned Accurate Match pointscited by 30
  35. 2019Backpropagation-Friendly Eigendecompositioncited by 27
  36. 2021Fusing the Old with the New: Learning Relative Camera Pose with Geometry-Guided Uncertaintycited by 9
  37. 2021Learning Neural Transmittance for Efficient Rendering of Reflectance Fieldscited by 3
  38. 2021End-to-end learning of keypoint detection and matching for relative pose estimationcited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 39; 2024: 75; 2025: 62; 2026: 18.

195 papers.

Sort:

By area:

  1. 2024MVSplat: Efficient 3D Gaussian Splatting from Sparse Multi-View Imagescited by 553
  2. 2023FSGS: Real-Time Few-shot View Synthesis using Gaussian Splattingcited by 278
  3. 2025FLARE: Feed-forward Geometry, Appearance and Camera Estimation from Uncalibrated Sparse Viewscited by 176
  4. 2024Splatt3R: Zero-shot Gaussian Splatting from Uncalibrated Image Pairscited by 171
  5. 2025Difix3D+: Improving 3D Reconstructions with Single-Step Diffusion Modelscited by 160
  6. 2024Cameras as Rays: Pose Estimation via Ray Diffusioncited by 129
  7. 2024MVSplat360: Feed-Forward 360 Scene Synthesis from Sparse Viewscited by 111
  8. 2025VMem: Consistent Interactive Video Scene Generation with Surfel-Indexed View Memorycited by 102
  9. 2025RayZer: A Self-supervised Large View Synthesis Modelcited by 74
  10. 2024TranSplat: Generalizable 3D Gaussian Splatting from Sparse Multi-View Images with Transformerscited by 71
  11. 2023MuRF: Multi-Baseline Radiance Fieldscited by 70
  12. 2024InstantSplat: Sparse-view Gaussian Splatting in Secondscited by 67
38 more
  1. 2023LEAP: Liberate Sparse-view 3D Modeling from Camera Posescited by 65
  2. 2024Advances in 3D Generation: A Surveycited by 62
  3. 2023HeadGaS: Real-Time Animatable Head Avatars via 3D Gaussian Splattingcited by 60
  4. 2024PF3plat: Pose-Free Feed-Forward 3D Gaussian Splattingcited by 60
  5. 2023CamP: Camera Preconditioning for Neural Radiance Fieldscited by 55
  6. 2024LaRa: Efficient Large-Baseline Radiance Fieldscited by 54
  7. 2023Explicit Correspondence Matching for Generalizable Neural Radiance Fieldscited by 46
  8. 2024Binocular-Guided 3D Gaussian Splatting with View Consistency for Sparse View Synthesiscited by 41
  9. 2023Deceptive-NeRF/3DGS: Diffusion-Generated Pseudo-Observations for High-Quality Sparse-View Reconstructioncited by 39
  10. 2025No Pose at All: Self-Supervised Pose-Free 3D Gaussian Splatting from Sparse Viewscited by 38
  11. 2023Preface: A Data-driven Volumetric Prior for Few-shot Ultra High-resolution Face Synthesiscited by 37
  12. 2023Unifying Correspondence, Pose and NeRF for Pose-Free Novel View Synthesis from Stereo Pairscited by 37
  13. 2025GSFixer: Improving 3D Gaussian Splatting with Reference-Guided Video Diffusion Priorscited by 36
  14. 2024SpaRP: Fast 3D Object Reconstruction and Pose Estimation from Sparse Viewscited by 35
  15. 2025Instant Gaussian Stream: Fast and Generalizable Streaming of Dynamic Scene Reconstruction via Gaussian Splattingcited by 35
  16. 2025Decompositional Neural Scene Reconstruction with Generative Diffusion Priorcited by 30
  17. 2023UniSDF: Unifying Neural Representations for High-Fidelity 3D Reconstruction of Complex Scenes with Reflectionscited by 29
  18. 2025DropoutGS: Dropping Out Gaussians for Better Sparse-view Renderingcited by 29
  19. 20244D Gaussian Splatting in the Wild with Uncertainty-Aware Regularizationcited by 28
  20. 2025FlowR: Flowing from Sparse to Dense 3D Reconstructionscited by 28
  21. 2024Feat2GS: Probing Visual Foundation Models with Gaussian Splattingcited by 26
  22. 2025Gaussian Graph Network: Learning Efficient and Generalizable Gaussian Representations from Multi-view Imagescited by 25
  23. 2023ZeroRF: Fast Sparse View 360deg Reconstruction with Zero Pretrainingcited by 24
  24. 2023UC-NeRF: Neural Radiance Field for Under-Calibrated Multi-view Cameras in Autonomous Drivingcited by 23
  25. 2024Splatter-360: Generalizable 360circ Gaussian Splatting for Wide-baseline Panoramic Imagescited by 23
  26. 2024PanSplat: 4K Panorama Synthesis with Feed-Forward Gaussian Splattingcited by 23
  27. 2023ARTIC3D: Learning Robust Articulated 3D Shapes from Noisy Web Image Collectionscited by 22
  28. 2024UFORecon: Generalizable Sparse-View Surface Reconstruction from Arbitrary and UnFavOrable Setscited by 22
  29. 2024Neural Fields in Robotics: A Surveycited by 21
  30. 2023Deep Learning for Camera Calibration and Beyond: A Surveycited by 20
  31. 2023PoRF: Pose Residual Field for Accurate Neural Surface Reconstructioncited by 19
  32. 2025AE-NeRF: Augmenting Event-Based Neural Radiance Fields for Non-ideal Conditions and Larger Scenecited by 19
  33. 2024SHINOBI: Shape and Illumination using Neural Object Decomposition via BRDF Optimization In-the-wildcited by 18
  34. 2024FewViewGS: Gaussian Splatting with Few View Matching and Multi-stage Trainingcited by 18
  35. 2024Structure Consistent Gaussian Splatting with Matching Prior for Few-shot Novel View Synthesiscited by 18
  36. 2023ConsistentNeRF: Enhancing Neural Radiance Fields with 3D Consistency for Sparse View Synthesiscited by 17
  37. 2024NeRFs in Robotics: A Surveycited by 16
  38. 2023CorresNeRF: Image Correspondence Priors for Neural Radiance Fieldscited by 15

50 most cited of 195 All 195 (JSON)