mlsanityData to 7 Oct 2026

FrugalNeRF: Fast Convergence for Extreme Few-shot Novel View Synthesis without Learned Priors

arXiv 2410.16271 ↗First posted 21 Oct 2024cs.CVcited by 18

Not on the map. Why?

Abstract (arXiv)

Neural Radiance Fields (NeRF) face significant challenges in extreme few-shot scenarios, primarily due to overfitting and long training times. Existing methods, such as FreeNeRF and SparseNeRF, use frequency regularization or pre-trained priors but struggle with complex scheduling and bias. We introduce FrugalNeRF, a novel few-shot NeRF framework that leverages weight-sharing voxels across multiple scales to efficiently represent scene details. Our key contribution is a cross-scale geometric adaptation scheme that selects pseudo ground truth depth based on reprojection errors across scales. This guides training without relying on externally learned priors, enabling full utilization of the training data. It can also integrate pre-trained priors, enhancing quality without slowing convergence. Experiments on LLFF, DTU, and RealEstate-10K show that FrugalNeRF outperforms other few-shot NeRF methods while significantly reducing training time, making it a practical solution for efficient and accurate 3D scene reconstruction.

Key facts. Cited by 18 papers in the map (data through 7 Oct 2026). Builds on 89 of its 100 references that are in the map. First posted on arXiv 21 Oct 2024.

Builds on

89 of 100 references.

References · 89

  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. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,893
  4. 2022Instant Neural Graphics Primitives with a Multiresolution Hash Encodingcited by 3,698
  5. 2017ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenescited by 3,088
  6. 2020Implicit Neural Representations with Periodic Activation Functionscited by 2,214
  7. 2020Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domainscited by 2,170
  8. 2021Mip-NeRF 360: Unbounded Anti-Aliased Neural Radiance Fieldscited by 2,013
  9. 2021Vision Transformers for Dense Predictioncited by 1,789
  10. 2021Plenoxels: Radiance Fields without Neural Networkscited by 1,607
  11. 2020pixelNeRF: Neural Radiance Fields from One or Few Imagescited by 1,574
  12. 2019Towards Robust Monocular Depth Estimation: Mixing Datasets for Zero-shot Cross-dataset Transfercited by 1,484
38 more
  1. 2020D-NeRF: Neural Radiance Fields for Dynamic Scenescited by 1,459
  2. 2022TensoRF: Tensorial Radiance Fieldscited by 1,275
  3. 2020NeRF in the Wild: Neural Radiance Fields for Unconstrained Photo Collectionscited by 1,230
  4. 2021Efficient Geometry-aware 3D Generative Adversarial Networkscited by 1,204
  5. 2021Direct Voxel Grid Optimization: Super-fast Convergence for Radiance Fields Reconstructioncited by 961
  6. 2019Scene Representation Networks: Continuous 3D-Structure-Aware Neural Scene Representationscited by 864
  7. 2020NeRF++: Analyzing and Improving Neural Radiance Fieldscited by 857
  8. 2021MVSNeRF: Fast Generalizable Radiance Field Reconstruction from Multi-View Stereocited by 735
  9. 2021Depth-supervised NeRF: Fewer Views and Faster Training for Freecited by 732
  10. 2020pi-GAN: Periodic Implicit Generative Adversarial Networks for 3D-Aware Image Synthesiscited by 708
  11. 2019Local Light Field Fusion: Practical View Synthesis with Prescriptive Sampling Guidelinescited by 672
  12. 2021UNISURF: Unifying Neural Implicit Surfaces and Radiance Fields for Multi-View Reconstructioncited by 643
  13. 2021RegNeRF: Regularizing Neural Radiance Fields for View Synthesis from Sparse Inputscited by 614
  14. 2021Zero-Shot Text-Guided Object Generation with Dream Fieldscited by 513
  15. 2021Putting NeRF on a Diet: Semantically Consistent Few-Shot View Synthesiscited by 464
  16. 2018DeepVoxels: Learning Persistent 3D Feature Embeddingscited by 401
  17. 2019SynSin: End-to-end View Synthesis from a Single Imagecited by 364
  18. 2021NeRF in the Dark: High Dynamic Range View Synthesis from Noisy Raw Imagescited by 336
  19. 2023FreeNeRF: Improving Few-shot Neural Rendering with Free Frequency Regularizationcited by 310
  20. 2021Dense Depth Priors for Neural Radiance Fields from Sparse Input Viewscited by 301
  21. 2023FSGS: Real-Time Few-shot View Synthesis using Gaussian Splattingcited by 278
  22. 2020Single-View View Synthesis with Multiplane Imagescited by 263
  23. 2024DNGaussian: Optimizing Sparse-View 3D Gaussian Radiance Fields with Global-Local Depth Normalizationcited by 256
  24. 2023SparseNeRF: Distilling Depth Ranking for Few-shot Novel View Synthesiscited by 247
  25. 2022NoPe-NeRF: Optimising Neural Radiance Field with No Pose Priorcited by 235
  26. 2020Consistent Video Depth Estimationcited by 220
  27. 2022NeRF-Editing: Geometry Editing of Neural Radiance Fieldscited by 216
  28. 2022SparseFusion: Distilling View-conditioned Diffusion for 3D Reconstructioncited by 216
  29. 2022Geo-Neus: Geometry-Consistent Neural Implicit Surfaces Learning for Multi-view Reconstructioncited by 211
  30. 2021GeoNeRF: Generalizing NeRF with Geometry Priorscited by 195
  31. 2023Robust Dynamic Radiance Fieldscited by 193
  32. 2021Stereo Radiance Fields (SRF): Learning View Synthesis for Sparse Views of Novel Scenescited by 190
  33. 2020Robust Consistent Video Depth Estimationcited by 188
  34. 2021InfoNeRF: Ray Entropy Minimization for Few-Shot Neural Volume Renderingcited by 180
  35. 2016Single-Image Depth Perception in the Wildcited by 179
  36. 2022SinNeRF: Training Neural Radiance Fields on Complex Scenes from a Single Imagecited by 165
  37. 2022NeRDi: Single-View NeRF Synthesis with Language-Guided Diffusion as General Image Priorscited by 156
  38. 2022NeuralLift-360: Lifting An In-the-wild 2D Photo to A 3D Object with 360deg Viewscited by 148

50 most cited of 89 All 89 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 15; 2026: 3.

18 papers.

Sort:

By area:

  1. 2025AuraFusion360: Augmented Unseen Region Alignment for Reference-based 360deg Unbounded Scene Inpaintingcited by 21
  2. 2025Skyfall-GS: Synthesizing Immersive 3D Urban Scenes from Satellite Imagerycited by 16
  3. 2025StealthAttack: Robust 3D Gaussian Splatting Poisoning via Density-Guided Illusionscited by 6
  4. 2025GaMO: Geometry-aware Multi-view Diffusion Outpainting for Sparse-View 3D Reconstructioncited by 4
  5. 2025Sparse-View 3D Reconstruction: Recent Advances and Open Challengescited by 3
  6. 2025UFV-Splatter: Pose-Free Feed-Forward 3D Gaussian Splatting Adapted to Unfavorable Viewscited by 3
  7. 2025Voxify3D: Pixel Art Meets Volumetric Renderingcited by 3
  8. 2025Splannequin: Freezing Monocular Mannequin-Challenge Footage with Dual-Detection Splattingcited by 2
  9. 2025TriDF: Triplane-Accelerated Density Fields for Few-Shot Remote Sensing Novel View Synthesiscited by 1
  10. 2025CamC2V: Context-aware Controllable Video Generationcited by 1
  11. 2025SVR-GS: Spatially Variant Regularization for Probabilistic Masks in 3D Gaussian Splattingcited by 1
  12. 2025DivCon-NeRF: Diverse and Consistent Ray Augmentation for Few-Shot NeRFcited by 0
6 more
  1. 2025Every Camera Effect, Every Time, All at Once: 4D Gaussian Ray Tracing for Physics-based Camera Effect Data Generationcited by 0
  2. 2025HybridSplat: Fast Reflection-baked Gaussian Tracing using Hybrid Splattingcited by 0
  3. 2026LGDWT-GS: Local and Global Discrete Wavelet-Regularized 3D Gaussian Splatting for Sparse-View Scene Reconstructioncited by 0
  4. 2026HeroGS: Hierarchical Guidance for Robust 3D Gaussian Splatting under Sparse Viewscited by 0
  5. 2025SAC-NeRF: Adaptive Ray Sampling for Neural Radiance Fields via Soft Actor-Critic Reinforcement Learningcited by 0
  6. 20263DTV: A Feedforward Interpolation Network for Real-Time View Synthesiscited by 0