mlsanityData to 13 Oct 2026

Learned Initializations for Optimizing Coordinate-Based Neural Representations

arXiv 2012.02189 ↗First posted 3 Dec 2020cs.CVcited by 241

Not on the map. Why?

Abstract (arXiv)

Coordinate-based neural representations have shown significant promise as an alternative to discrete, array-based representations for complex low dimensional signals. However, optimizing a coordinate-based network from randomly initialized weights for each new signal is inefficient. We propose applying standard meta-learning algorithms to learn the initial weight parameters for these fully-connected networks based on the underlying class of signals being represented (e.g., images of faces or 3D models of chairs). Despite requiring only a minor change in implementation, using these learned initial weights enables faster convergence during optimization and can serve as a strong prior over the signal class being modeled, resulting in better generalization when only partial observations of a given signal are available. We explore these benefits across a variety of tasks, including representing 2D images, reconstructing CT scans, and recovering 3D shapes and scenes from 2D image observations.

Key facts. Cited by 241 papers in the map (data through 13 Oct 2026). First posted on arXiv 3 Dec 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 13 Oct)

Citing papers per year: 2021: 41; 2022: 49; 2023: 57; 2024: 44; 2025: 36; 2026: 14.

241 papers.

Sort:

By area:

  1. 2021Mip-NeRF: A Multiscale Representation for Anti-Aliasing Neural Radiance Fieldscited by 1,958
  2. 2021Plenoxels: Radiance Fields without Neural Networkscited by 1,610
  3. 2021Direct Voxel Grid Optimization: Super-fast Convergence for Radiance Fields Reconstructioncited by 962
  4. 2021PlenOctrees for Real-time Rendering of Neural Radiance Fieldscited by 905
  5. 2023K-Planes: Explicit Radiance Fields in Space, Time, and Appearancecited by 744
  6. 2021Depth-supervised NeRF: Fewer Views and Faster Training for Freecited by 733
  7. 2021KiloNeRF: Speeding up Neural Radiance Fields with Thousands of Tiny MLPscited by 672
  8. 2021RegNeRF: Regularizing Neural Radiance Fields for View Synthesis from Sparse Inputscited by 614
  9. 2021Neural Fields in Visual Computing and Beyondcited by 539
  10. 2021Zero-Shot Text-Guided Object Generation with Dream Fieldscited by 513
  11. 2021FastNeRF: High-Fidelity Neural Rendering at 200FPScited by 511
  12. 2021Putting NeRF on a Diet: Semantically Consistent Few-Shot View Synthesiscited by 464
38 more
  1. 2021Mega-NeRF: Scalable Construction of Large-Scale NeRFs for Virtual Fly-Throughscited by 396
  2. 2021Advances in Neural Renderingcited by 365
  3. 2021Mixture of Volumetric Primitives for Efficient Neural Renderingcited by 277
  4. 2023Generative Novel View Synthesis with 3D-Aware Diffusion Modelscited by 273
  5. 2022Panoptic Neural Fields: A Semantic Object-Aware Neural Scene Representationcited by 243
  6. 2021COIN: COmpression with Implicit Neural representationscited by 233
  7. 2021Editing Conditional Radiance Fieldscited by 222
  8. 2023EAGLES: Efficient Accelerated 3D Gaussians with Lightweight EncodingScited by 193
  9. 2021Neural Rays for Occlusion-aware Image-based Renderingcited by 192
  10. 2022From data to functa: Your data point is a function and you can treat it like onecited by 186
  11. 2022Fourier PlenOctrees for Dynamic Radiance Field Rendering in Real-timecited by 165
  12. 2021Implicit Neural Representations for Image Compressioncited by 162
  13. 2021NeRS: Neural Reflectance Surfaces for Sparse-view 3D Reconstruction in the Wildcited by 136
  14. 2021Local Texture Estimator for Implicit Representation Functioncited by 127
  15. 2021LENS: Localization enhanced by NeRF synthesiscited by 117
  16. 2021NeRF-VAE: A Geometry Aware 3D Scene Generative Modelcited by 111
  17. 2021A Structured Dictionary Perspective on Implicit Neural Representationscited by 100
  18. 2023Grid-guided Neural Radiance Fields for Large Urban Scenescited by 99
  19. 2021GANcraft: Unsupervised 3D Neural Rendering of Minecraft Worldscited by 98
  20. 20223DILG: Irregular Latent Grids for 3D Generative Modelingcited by 92
  21. 2023Implicit Neural Representation for Cooperative Low-light Image Enhancementcited by 92
  22. 2021Neural Trajectory Fields for Dynamic Novel View Synthesiscited by 86
  23. 2022COIN++: Neural Compression Across Modalitiescited by 82
  24. 2023Operator Learning with Neural Fields: Tackling PDEs on General Geometriescited by 81
  25. 2022Transformers as Meta-Learners for Implicit Neural Representationscited by 79
  26. 2021MetaAvatar: Learning Animatable Clothed Human Models from Few Depth Imagescited by 78
  27. 2021Learning to Stylize Novel Viewscited by 76
  28. 2022E-NeRV: Expedite Neural Video Representation with Disentangled Spatial-Temporal Contextcited by 76
  29. 2022Conditional-Flow NeRF: Accurate 3D Modelling with Reliable Uncertainty Quantificationcited by 72
  30. 2023HiNeRV: Video Compression with Hierarchical Encoding-based Neural Representationcited by 71
  31. 2023Learning Neural Volumetric Representations of Dynamic Humans in Minutescited by 70
  32. 2022Overview frequency principle/spectral bias in deep learningcited by 69
  33. 2021DIVeR: Real-time and Accurate Neural Radiance Fields with Deterministic Integration for Volume Renderingcited by 60
  34. 2021SAPE: Spatially-Adaptive Progressive Encoding for Neural Optimizationcited by 59
  35. 2022Control-NeRF: Editable Feature Volumes for Scene Rendering and Manipulationcited by 57
  36. 2022Masked Wavelet Representation for Compact Neural Radiance Fieldscited by 57
  37. 2023Cascaded Local Implicit Transformer for Arbitrary-Scale Super-Resolutioncited by 52
  38. 2022NIRVANA: Neural Implicit Representations of Videos with Adaptive Networks and Autoregressive Patch-wise Modelingcited by 51

50 most cited of 241 All 241 (JSON)