mlsanityData to 7 Oct 2026

Neural Implicit Dictionary via Mixture-of-Expert Training

arXiv 2207.03691 ↗First posted 8 Jul 2022cs.CVcited by 9

On the map: Mixture-of-experts → Vision mixture-of-experts

Abstract (arXiv)

Representing visual signals by coordinate-based deep fully-connected networks has been shown advantageous in fitting complex details and solving inverse problems than discrete grid-based representation. However, acquiring such a continuous Implicit Neural Representation (INR) requires tedious per-scene training on tons of signal measurements, which limits its practicality. In this paper, we present a generic INR framework that achieves both data and training efficiency by learning a Neural Implicit Dictionary (NID) from a data collection and representing INR as a functional combination of basis sampled from the dictionary. Our NID assembles a group of coordinate-based subnetworks which are tuned to span the desired function space. After training, one can instantly and robustly acquire an unseen scene representation by solving the coding coefficients. To parallelly optimize a large group of networks, we borrow the idea from Mixture-of-Expert (MoE) to design and train our network with a sparse gating mechanism. Our experiments show that, NID can improve reconstruction of 2D images or 3D scenes by 2 orders of magnitude faster with up to 98% less input data. We further demonstrate various applications of NID in image inpainting and occlusion removal, which are considered to be challenging with vanilla INR. Our codes are available in https://github.com/VITA-Group/Neural-Implicit-Dict.

Key facts. Cited by 9 papers in the map (data through 7 Oct 2026). Placed in Vision mixture-of-experts, one of 12 directions in Mixture-of-experts. Builds on 55 of its 81 references that are in the map. First posted on arXiv 8 Jul 2022.

Builds on

55 of 81 references.

References · 55

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,029
  3. 2016PointNet: Deep Learning on Point Sets for 3D Classification and Segmentationcited by 5,223
  4. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
  5. 2014Deep Learning Face Attributes in the Wildcited by 3,679
  6. 2017Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layercited by 3,413
  7. 2021Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsitycited by 3,014
  8. 2015ShapeNet: An Information-Rich 3D Model Repositorycited by 2,880
  9. 2020Fourier Neural Operator for Parametric Partial Differential Equationscited by 2,806
  10. 2019DeepSDF: Learning Continuous Signed Distance Functions for Shape Representationcited by 2,682
  11. 2020Implicit Neural Representations with Periodic Activation Functionscited by 2,214
  12. 2020Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domainscited by 2,170
38 more
  1. 2018Occupancy Networks: Learning 3D Reconstruction in Function Spacecited by 2,037
  2. 2020GShard: Scaling Giant Models with Conditional Computation and Automatic Shardingcited by 1,601
  3. 2020pixelNeRF: Neural Radiance Fields from One or Few Imagescited by 1,574
  4. 2017Deep Setscited by 1,394
  5. 2020NeRF in the Wild: Neural Radiance Fields for Unconstrained Photo Collectionscited by 1,230
  6. 2016HyperNetworkscited by 1,003
  7. 2019PIFu: Pixel-Aligned Implicit Function for High-Resolution Clothed Human Digitizationcited by 818
  8. 2020Convolutional Occupancy Networkscited by 760
  9. 2021MVSNeRF: Fast Generalizable Radiance Field Reconstruction from Multi-View Stereocited by 735
  10. 2020Implicit Geometric Regularization for Learning Shapescited by 707
  11. 2021IBRNet: Learning Multi-View Image-Based Renderingcited by 698
  12. 2021KiloNeRF: Speeding up Neural Radiance Fields with Thousands of Tiny MLPscited by 672
  13. 2018Free-Form Image Inpainting with Gated Convolutioncited by 594
  14. 20163D-R2N2: A Unified Approach for Single and Multi-view 3D Object Reconstructioncited by 587
  15. 2020Learning Continuous Image Representation with Local Implicit Image Functioncited by 550
  16. 2017Solving high-dimensional partial differential equations using deep learningcited by 549
  17. 2018Image Inpainting for Irregular Holes Using Partial Convolutionscited by 505
  18. 2021Mega-NeRF: Scalable Construction of Large-Scale NeRFs for Virtual Fly-Throughscited by 396
  19. 2020Implicit Functions in Feature Space for 3D Shape Reconstruction and Completioncited by 343
  20. 2021Light Field Networks: Neural Scene Representations with Single-Evaluation Renderingcited by 330
  21. 2020Deep Local Shapes: Learning Local SDF Priors for Detailed 3D Reconstructioncited by 321
  22. 2021BASE Layers: Simplifying Training of Large, Sparse Modelscited by 315
  23. 2021NeRV: Neural Representations for Videoscited by 255
  24. 2022Unified Scaling Laws for Routed Language Modelscited by 249
  25. 2020Learned Initializations for Optimizing Coordinate-Based Neural Representationscited by 240
  26. 2021Hash Layers For Large Sparse Modelscited by 236
  27. 2020MetaSDF: Meta-learning Signed Distance Functionscited by 214
  28. 2021CodeNeRF: Disentangled Neural Radiance Fields for Object Categoriescited by 183
  29. 2015Conditional Computation in Neural Networks for faster modelscited by 166
  30. 2017Optimizing the Latent Space of Generative Networkscited by 157
  31. 2019Convolutional Conditional Neural Processescited by 139
  32. 2021FastMoE: A Fast Mixture-of-Expert Training Systemcited by 110
  33. 2021ShaRF: Shape-conditioned Radiance Fields from a Single Viewcited by 92
  34. 2021T"oRF: Time-of-Flight Radiance Fields for Dynamic Scene View Synthesiscited by 88
  35. 2015On the Error of Random Fourier Featurescited by 76
  36. 2014PCANet: A Simple Deep Learning Baseline for Image Classification?cited by 70
  37. 2014From Denoising to Compressed Sensingcited by 67
  38. 2022Unified Implicit Neural Stylizationcited by 67

50 most cited of 55 All 55 (JSON)

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 1; 2023: 3; 2024: 1; 2025: 3; 2026: 1.

9 papers.

  1. 2022Signal Processing for Implicit Neural Representationscited by 46
  2. 2023Enhancing NeRF akin to Enhancing LLMs: Generalizable NeRF Transformer with Mixture-of-View-Expertscited by 27
  3. 2023Partition Speeds Up Learning Implicit Neural Representations Based on Exponential-Increase Hypothesiscited by 13
  4. 2024Nonparametric Teaching of Implicit Neural Representationscited by 13
  5. 2023Sparsity-Guided Holistic Explanation for LLMs with Interpretable Inference-Time Interventioncited by 12
  6. 2025SNeRV: Spectra-preserving Neural Representation for Videocited by 9
  7. 2025Mixture-of-Shape-Experts (MoSE): End-to-End Shape Dictionary Framework to Prompt SAM for Generalizable Medical Segmentationcited by 4
  8. 2025NTK-Guided Implicit Neural Teachingcited by 2
  9. 2026MoE-dqINR: A Unified Mixture-of-Experts Implicit Neural Representation Framework for Scan-Specific Dynamic and Quantitative MRI Reconstructioncited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Neural Experts: Mixture of Experts for Implicit Neural Representations
  2. 2023MoEC: Mixture of Experts Implicit Neural Compression
  3. 2023Towards Scalable Neural Representation for Diverse Videos
  4. 2026From Scalars to Time Series: Rethinking Implicit Neural Representations for Time-Varying Volumetric Data
  5. 2025Efficient Data Driven Mixture-of-Expert Extraction from Trained Networks
  6. 2026Weight-Space Mixture-of-Experts for Implicit Neural Representation Classification
  7. 2025Few-shot Implicit Function Generation via Equivariance
  8. 2023O2-Recon: Completing 3D Reconstruction of Occluded Objects in the Scene with a Pre-trained 2D Diffusion Model