# R2L: Distilling Neural Radiance Field to Neural Light Field for Efficient Novel View Synthesis

arXiv: https://arxiv.org/abs/2203.17261 · HTML: https://mlsanity.com/paper/2203.17261/

## Key facts

- Cited by 69 papers in the map (data through 14 Oct 2026).
- Builds on 43 of its 62 references that are in the map.
- First posted on arXiv 31 Mar 2022.

## Builds on (most cited first, 30 of 43)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift](https://mlsanity.com/paper/1502.03167/) (2015)
- [DeepSDF: Learning Continuous Signed Distance Functions for Shape Representation](https://mlsanity.com/paper/1901.05103/) (2019)
- [Occupancy Networks: Learning 3D Reconstruction in Function Space](https://mlsanity.com/paper/1812.03828/) (2018)
- [Mip-NeRF: A Multiscale Representation for Anti-Aliasing Neural Radiance Fields](https://mlsanity.com/paper/2103.13415/) (2021)
- [FitNets: Hints for Thin Deep Nets](https://mlsanity.com/paper/1412.6550/) (2014)
- [pixelNeRF: Neural Radiance Fields from One or Few Images](https://mlsanity.com/paper/2012.02190/) (2020)
- [TinyBERT: Distilling BERT for Natural Language Understanding](https://mlsanity.com/paper/1909.10351/) (2019)
- [Paying More Attention to Attention: Improving the Performance of Convolutional Neural Networks via Attention Transfer](https://mlsanity.com/paper/1612.03928/) (2016)
- [Learning Implicit Fields for Generative Shape Modeling](https://mlsanity.com/paper/1812.02822/) (2018)
- [Neural Sparse Voxel Fields](https://mlsanity.com/paper/2007.11571/) (2020)
- [PlenOctrees for Real-time Rendering of Neural Radiance Fields](https://mlsanity.com/paper/2103.14024/) (2021)
- [Relational Knowledge Distillation](https://mlsanity.com/paper/1904.05068/) (2019)
- [Neural Scene Flow Fields for Space-Time View Synthesis of Dynamic Scenes](https://mlsanity.com/paper/2011.13084/) (2020)
- [Contrastive Representation Distillation](https://mlsanity.com/paper/1910.10699/) (2019)
- [KiloNeRF: Speeding up Neural Radiance Fields with Thousands of Tiny MLPs](https://mlsanity.com/paper/2103.13744/) (2021)
- [Training Region-based Object Detectors with Online Hard Example Mining](https://mlsanity.com/paper/1604.03540/) (2016)
- [Similarity-Preserving Knowledge Distillation](https://mlsanity.com/paper/1907.09682/) (2019)
- [FastNeRF: High-Fidelity Neural Rendering at 200FPS](https://mlsanity.com/paper/2103.10380/) (2021)
- [Do Deep Nets Really Need to be Deep?](https://mlsanity.com/paper/1312.6184/) (2013)
- [Baking Neural Radiance Fields for Real-Time View Synthesis](https://mlsanity.com/paper/2103.14645/) (2021)
- [Neural Geometric Level of Detail: Real-time Rendering with Implicit 3D Shapes](https://mlsanity.com/paper/2101.10994/) (2021)
- [Light Field Networks: Neural Scene Representations with Single-Evaluation Rendering](https://mlsanity.com/paper/2106.02634/) (2021)
- [Knowledge Distillation and Student-Teacher Learning for Visual Intelligence: A Review and New Outlooks](https://mlsanity.com/paper/2004.05937/) (2020)
- [Correlation Congruence for Knowledge Distillation](https://mlsanity.com/paper/1904.01802/) (2019)

## Built on by (most cited first, 30 of 69)

- [LightGaussian: Unbounded 3D Gaussian Compression with 15x Reduction and 200+ FPS](https://mlsanity.com/paper/2311.17245/) (2023)
- [Spacetime Gaussian Feature Splatting for Real-Time Dynamic View Synthesis](https://mlsanity.com/paper/2312.16812/) (2023)
- [MERF: Memory-Efficient Radiance Fields for Real-time View Synthesis in Unbounded Scenes](https://mlsanity.com/paper/2302.12249/) (2023)
- [VastGaussian: Vast 3D Gaussians for Large Scene Reconstruction](https://mlsanity.com/paper/2402.17427/) (2024)
- [3DGS-Avatar: Animatable Avatars via Deformable 3D Gaussian Splatting](https://mlsanity.com/paper/2312.09228/) (2023)
- [HyperReel: High-Fidelity 6-DoF Video with Ray-Conditioned Sampling](https://mlsanity.com/paper/2301.02238/) (2023)
- [NeRF: Neural Radiance Field in 3D Vision: A Comprehensive Review (Updated Post-Gaussian Splatting)](https://mlsanity.com/paper/2210.00379/) (2022)
- [Fully Explicit Dynamic Gaussian Splatting](https://mlsanity.com/paper/2410.15629/) (2024)
- [SPAD : Spatially Aware Multiview Diffusers](https://mlsanity.com/paper/2402.05235/) (2024)
- [Masked Wavelet Representation for Compact Neural Radiance Fields](https://mlsanity.com/paper/2212.09069/) (2022)
- [Real-Time Neural Light Field on Mobile Devices](https://mlsanity.com/paper/2212.08057/) (2022)
- [Masked Space-Time Hash Encoding for Efficient Dynamic Scene Reconstruction](https://mlsanity.com/paper/2310.17527/) (2023)
- [ExBluRF: Efficient Radiance Fields for Extreme Motion Blurred Images](https://mlsanity.com/paper/2309.08957/) (2023)
- [Vidu4D: Single Generated Video to High-Fidelity 4D Reconstruction with Dynamic Gaussian Surfels](https://mlsanity.com/paper/2405.16822/) (2024)
- [StegaNeRF: Embedding Invisible Information within Neural Radiance Fields](https://mlsanity.com/paper/2212.01602/) (2022)
- [ZPressor: Bottleneck-Aware Compression for Scalable Feed-Forward 3DGS](https://mlsanity.com/paper/2505.23734/) (2025)
- [Learning Neural Duplex Radiance Fields for Real-Time View Synthesis](https://mlsanity.com/paper/2304.10537/) (2023)
- [Enhancing NeRF akin to Enhancing LLMs: Generalizable NeRF Transformer with Mixture-of-View-Experts](https://mlsanity.com/paper/2308.11793/) (2023)
- [Multi-view Inverse Rendering for Large-scale Real-world Indoor Scenes](https://mlsanity.com/paper/2211.10206/) (2022)
- [Interactive Segment Anything NeRF with Feature Imitation](https://mlsanity.com/paper/2305.16233/) (2023)
- [Dynamic Mesh-Aware Radiance Fields](https://mlsanity.com/paper/2309.04581/) (2023)
- [DyLiN: Making Light Field Networks Dynamic](https://mlsanity.com/paper/2303.14243/) (2023)
- [One is All: Bridging the Gap Between Neural Radiance Fields Architectures with Progressive Volume Distillation](https://mlsanity.com/paper/2211.15977/) (2022)
- [Protecting NeRFs' Copyright via Plug-And-Play Watermarking Base Model](https://mlsanity.com/paper/2407.07735/) (2024)
- [Re-ReND: Real-time Rendering of NeRFs across Devices](https://mlsanity.com/paper/2303.08717/) (2023)
- [SteerNeRF: Accelerating NeRF Rendering via Smooth Viewpoint Trajectory](https://mlsanity.com/paper/2212.08476/) (2022)
- [1000+ FPS 4D Gaussian Splatting for Dynamic Scene Rendering](https://mlsanity.com/paper/2503.16422/) (2025)
- [RayDF: Neural Ray-surface Distance Fields with Multi-view Consistency](https://mlsanity.com/paper/2310.19629/) (2023)
- [DeblurGS: Gaussian Splatting for Camera Motion Blur](https://mlsanity.com/paper/2404.11358/) (2024)
- [StyleRF-VolVis: Style Transfer of Neural Radiance Fields for Expressive Volume Visualization](https://mlsanity.com/paper/2408.00150/) (2024)

## Abstract (arXiv)

Recent research explosion on Neural Radiance Field (NeRF) shows the encouraging potential to represent complex scenes with neural networks. One major drawback of NeRF is its prohibitive inference time: Rendering a single pixel requires querying the NeRF network hundreds of times. To resolve it, existing efforts mainly attempt to reduce the number of required sampled points. However, the problem of iterative sampling still exists. On the other hand, Neural Light Field (NeLF) presents a more straightforward representation over NeRF in novel view synthesis -- the rendering of a pixel amounts to one single forward pass without ray-marching. In this work, we present a deep residual MLP network (88 layers) to effectively learn the light field. We show the key to successfully learning such a deep NeLF network is to have sufficient data, for which we transfer the knowledge from a pre-trained NeRF model via data distillation. Extensive experiments on both synthetic and real-world scenes show the merits of our method over other counterpart algorithms. On the synthetic scenes, we achieve 26-35x FLOPs reduction (per camera ray) and 28-31x runtime speedup, meanwhile delivering significantly better (1.4-2.8 dB average PSNR improvement) rendering quality than NeRF without any customized parallelism requirement.