mlsanityData to 7 Oct 2026

Direct Voxel Grid Optimization: Super-fast Convergence for Radiance Fields Reconstruction

arXiv 2111.11215 ↗First posted 22 Nov 2021cs.CVcited by 961

Not on the map. Why?

Cited by 34 in Diffusion models, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

We present a super-fast convergence approach to reconstructing the per-scene radiance field from a set of images that capture the scene with known poses. This task, which is often applied to novel view synthesis, is recently revolutionized by Neural Radiance Field (NeRF) for its state-of-the-art quality and flexibility. However, NeRF and its variants require a lengthy training time ranging from hours to days for a single scene. In contrast, our approach achieves NeRF-comparable quality and converges rapidly from scratch in less than 15 minutes with a single GPU. We adopt a representation consisting of a density voxel grid for scene geometry and a feature voxel grid with a shallow network for complex view-dependent appearance. Modeling with explicit and discretized volume representations is not new, but we propose two simple yet non-trivial techniques that contribute to fast convergence speed and high-quality output. First, we introduce the post-activation interpolation on voxel density, which is capable of producing sharp surfaces in lower grid resolution. Second, direct voxel density optimization is prone to suboptimal geometry solutions, so we robustify the optimization process by imposing several priors. Finally, evaluation on five inward-facing benchmarks shows that our method matches, if not surpasses, NeRF's quality, yet it only takes about 15 minutes to train from scratch for a new scene.

Key facts. Cited by 961 papers in the map (data through 7 Oct 2026). Builds on 52 of its 74 references that are in the map. First posted on arXiv 22 Nov 2021.

Builds on

52 of 74 references.

References · 52

  1. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  2. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,029
  3. 2020Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domainscited by 2,170
  4. 2021Mip-NeRF: A Multiscale Representation for Anti-Aliasing Neural Radiance Fieldscited by 1,957
  5. 2020pixelNeRF: Neural Radiance Fields from One or Few Imagescited by 1,574
  6. 2020D-NeRF: Neural Radiance Fields for Dynamic Scenescited by 1,459
  7. 2020NeRF in the Wild: Neural Radiance Fields for Unconstrained Photo Collectionscited by 1,230
  8. 2020Neural Sparse Voxel Fieldscited by 959
  9. 2021PlenOctrees for Real-time Rendering of Neural Radiance Fieldscited by 904
  10. 2019Scene Representation Networks: Continuous 3D-Structure-Aware Neural Scene Representationscited by 864
  11. 2020NeRF++: Analyzing and Improving Neural Radiance Fieldscited by 857
  12. 2020Neural Scene Flow Fields for Space-Time View Synthesis of Dynamic Scenescited by 750
38 more
  1. 2021MVSNeRF: Fast Generalizable Radiance Field Reconstruction from Multi-View Stereocited by 735
  2. 2021Depth-supervised NeRF: Fewer Views and Faster Training for Freecited by 732
  3. 2020pi-GAN: Periodic Implicit Generative Adversarial Networks for 3D-Aware Image Synthesiscited by 708
  4. 2021IBRNet: Learning Multi-View Image-Based Renderingcited by 698
  5. 2020GRAF: Generative Radiance Fields for 3D-Aware Image Synthesiscited by 688
  6. 2021KiloNeRF: Speeding up Neural Radiance Fields with Thousands of Tiny MLPscited by 672
  7. 2021UNISURF: Unifying Neural Implicit Surfaces and Radiance Fields for Multi-View Reconstructioncited by 643
  8. 2021BARF: Bundle-Adjusting Neural Radiance Fieldscited by 641
  9. 2021FastNeRF: High-Fidelity Neural Rendering at 200FPScited by 510
  10. 2020NeRV: Neural Reflectance and Visibility Fields for Relighting and View Synthesiscited by 493
  11. 2019BlendedMVS: A Large-scale Dataset for Generalized Multi-view Stereo Networkscited by 487
  12. 2020Non-Rigid Neural Radiance Fields: Reconstruction and Novel View Synthesis of a Dynamic Scene From Monocular Videocited by 470
  13. 2020Dynamic Neural Radiance Fields for Monocular 4D Facial Avatar Reconstructioncited by 465
  14. 2021Baking Neural Radiance Fields for Real-Time View Synthesiscited by 465
  15. 2020NeRD: Neural Reflectance Decomposition from Image Collectionscited by 446
  16. 2021Dynamic View Synthesis from Dynamic Monocular Videocited by 420
  17. 2020Space-time Neural Irradiance Fields for Free-Viewpoint Videocited by 414
  18. 2018DeepVoxels: Learning Persistent 3D Feature Embeddingscited by 401
  19. 2020INeRF: Inverting Neural Radiance Fields for Pose Estimationcited by 381
  20. 2020Single-View View Synthesis with Multiplane Imagescited by 263
  21. 2019Deferred Neural Rendering: Image Synthesis using Neural Texturescited by 253
  22. 2019DeepView: View Synthesis with Learned Gradient Descentcited by 253
  23. 2020Learned Initializations for Optimizing Coordinate-Based Neural Representationscited by 240
  24. 2021NeX: Real-time View Synthesis with Neural Basis Expansioncited by 215
  25. 2018Stereo Magnification: Learning View Synthesis using Multiplane Imagescited by 206
  26. 20203D Photography using Context-aware Layered Depth Inpaintingcited by 206
  27. 2020AutoInt: Automatic Integration for Fast Neural Volume Renderingcited by 204
  28. 2021Self-Calibrating Neural Radiance Fieldscited by 204
  29. 2020Neural Reflectance Fields for Appearance Acquisitioncited by 196
  30. 2021Neural Articulated Radiance Fieldcited by 196
  31. 2021Neural Rays for Occlusion-aware Image-based Renderingcited by 192
  32. 2019Pushing the Boundaries of View Extrapolation with Multiplane Imagescited by 191
  33. 2020DeRF: Decomposed Radiance Fieldscited by 163
  34. 2021NeRF--: Neural Radiance Fields Without Known Camera Parameterscited by 155
  35. 2021GNeRF: GAN-based Neural Radiance Field without Posed Cameracited by 153
  36. 2021NeRFactor: Neural Factorization of Shape and Reflectance Under an Unknown Illuminationcited by 135
  37. 2021NeRF-VAE: A Geometry Aware 3D Scene Generative Modelcited by 111
  38. 2018Layer-structured 3D Scene Inference via View Synthesiscited by 82

50 most cited of 52 All 52 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 3; 2022: 100; 2023: 309; 2024: 341; 2025: 148; 2026: 60.

961 papers.

Sort:

By area:

  1. 20233D Gaussian Splatting for Real-Time Radiance Field Renderingcited by 6,141
  2. 2022Instant Neural Graphics Primitives with a Multiresolution Hash Encodingcited by 3,698
  3. 2023Zero-1-to-3: Zero-shot One Image to 3D Objectcited by 1,483
  4. 2022TensoRF: Tensorial Radiance Fieldscited by 1,275
  5. 20234D Gaussian Splatting for Real-Time Dynamic Scene Renderingcited by 1,144
  6. 20242D Gaussian Splatting for Geometrically Accurate Radiance Fieldscited by 1,061
  7. 2023LRM: Large Reconstruction Model for Single Image to 3Dcited by 834
  8. 2023Deformable 3D Gaussians for High-Fidelity Monocular Dynamic Scene Reconstructioncited by 753
  9. 2023ScanNet++: A High-Fidelity Dataset of 3D Indoor Scenescited by 747
  10. 2023K-Planes: Explicit Radiance Fields in Space, Time, and Appearancecited by 742
  11. 2022Block-NeRF: Scalable Large Scene Neural View Synthesiscited by 720
  12. 2023Zip-NeRF: Anti-Aliased Grid-Based Neural Radiance Fieldscited by 667
38 more
  1. 2023Mip-Splatting: Alias-free 3D Gaussian Splattingcited by 661
  2. 2023SuGaR: Surface-Aligned Gaussian Splatting for Efficient 3D Mesh Reconstruction and High-Quality Mesh Renderingcited by 611
  3. 2022Score Jacobian Chaining: Lifting Pretrained 2D Diffusion Models for 3D Generationcited by 597
  4. 2023Nerfstudio: A Modular Framework for Neural Radiance Field Developmentcited by 596
  5. 2023HexPlane: A Fast Representation for Dynamic Scenescited by 582
  6. 2023Real-time Photorealistic Dynamic Scene Representation and Rendering with 4D Gaussian Splattingcited by 470
  7. 2023Neuralangelo: High-Fidelity Neural Surface Reconstructioncited by 464
  8. 2023LightGaussian: Unbounded 3D Gaussian Compression with 15x Reduction and 200+ FPScited by 420
  9. 2023Gaussian Splatting SLAMcited by 410
  10. 2023LangSplat: 3D Language Gaussian Splattingcited by 410
  11. 2023PhysGaussian: Physics-Integrated 3D Gaussians for Generative Dynamicscited by 368
  12. 2023Compact 3D Gaussian Representation for Radiance Fieldcited by 365
  13. 2023Splatter Image: Ultra-Fast Single-View 3D Reconstructioncited by 359
  14. 2022Fast Dynamic Radiance Fields with Time-Aware Neural Voxelscited by 344
  15. 2024GS-LRM: Large Reconstruction Model for 3D Gaussian Splattingcited by 344
  16. 2023Spacetime Gaussian Feature Splatting for Real-Time Dynamic View Synthesiscited by 341
  17. 2023OmniObject3D: Large-Vocabulary 3D Object Dataset for Realistic Perception, Reconstruction and Generationcited by 324
  18. 2022NeuS2: Fast Learning of Neural Implicit Surfaces for Multi-view Reconstructioncited by 311
  19. 2022MobileNeRF: Exploiting the Polygon Rasterization Pipeline for Efficient Neural Field Rendering on Mobile Architecturescited by 308
  20. 2023Text-to-3D using Gaussian Splattingcited by 299
  21. 2023UniSim: A Neural Closed-Loop Sensor Simulatorcited by 294
  22. 2023FSGS: Real-Time Few-shot View Synthesis using Gaussian Splattingcited by 278
  23. 2023GaussianAvatars: Photorealistic Head Avatars with Rigged 3D Gaussianscited by 256
  24. 2024DNGaussian: Optimizing Sparse-View 3D Gaussian Radiance Fields with Global-Local Depth Normalizationcited by 256
  25. 2023Compressed 3D Gaussian Splatting for Accelerated Novel View Synthesiscited by 232
  26. 2023LucidDreamer: Domain-free Generation of 3D Gaussian Splatting Scenescited by 228
  27. 2024Gaussian Opacity Fields: Efficient Adaptive Surface Reconstruction in Unbounded Scenescited by 226
  28. 2024Octree-GS: Towards Consistent Real-time Rendering with LOD-Structured 3D Gaussianscited by 225
  29. 2024A Survey on 3D Gaussian Splattingcited by 221
  30. 2023BakedSDF: Meshing Neural SDFs for Real-Time View Synthesiscited by 220
  31. 2023GS-IR: 3D Gaussian Splatting for Inverse Renderingcited by 220
  32. 2022Tensor4D : Efficient Neural 4D Decomposition for High-fidelity Dynamic Reconstruction and Renderingcited by 217
  33. 2023MERF: Memory-Efficient Radiance Fields for Real-time View Synthesis in Unbounded Scenescited by 215
  34. 2024A Hierarchical 3D Gaussian Representation for Real-Time Rendering of Very Large Datasetscited by 214
  35. 2022Geo-Neus: Geometry-Consistent Neural Implicit Surfaces Learning for Multi-view Reconstructioncited by 211
  36. 2023Gaussian-Flow: 4D Reconstruction with Dynamic 3D Gaussian Particlecited by 210
  37. 2024VastGaussian: Vast 3D Gaussians for Large Scene Reconstructioncited by 209
  38. 2024Mini-Splatting: Representing Scenes with a Constrained Number of Gaussianscited by 208

50 most cited of 961 All 961 (JSON)