# SqueezeNeRF: Further factorized FastNeRF for memory-efficient inference

arXiv: https://arxiv.org/abs/2204.02585 · HTML: https://mlsanity.com/paper/2204.02585/

## Key facts

- Cited by 16 papers in the map (data through 14 Oct 2026).
- Builds on 16 of its 25 references that are in the map.
- First posted on arXiv 6 Apr 2022.

## Builds on (most cited first, 16 of 16)

- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [Plenoxels: Radiance Fields without Neural Networks](https://mlsanity.com/paper/2112.05131/) (2021)
- [pixelNeRF: Neural Radiance Fields from One or Few Images](https://mlsanity.com/paper/2012.02190/) (2020)
- [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)
- [NeRF++: Analyzing and Improving Neural Radiance Fields](https://mlsanity.com/paper/2010.07492/) (2020)
- [Depth-supervised NeRF: Fewer Views and Faster Training for Free](https://mlsanity.com/paper/2107.02791/) (2021)
- [KiloNeRF: Speeding up Neural Radiance Fields with Thousands of Tiny MLPs](https://mlsanity.com/paper/2103.13744/) (2021)
- [Local Light Field Fusion: Practical View Synthesis with Prescriptive Sampling Guidelines](https://mlsanity.com/paper/1905.00889/) (2019)
- [FastNeRF: High-Fidelity Neural Rendering at 200FPS](https://mlsanity.com/paper/2103.10380/) (2021)
- [Baking Neural Radiance Fields for Real-Time View Synthesis](https://mlsanity.com/paper/2103.14645/) (2021)
- [DONeRF: Towards Real-Time Rendering of Compact Neural Radiance Fields using Depth Oracle Networks](https://mlsanity.com/paper/2103.03231/) (2021)
- [NeX: Real-time View Synthesis with Neural Basis Expansion](https://mlsanity.com/paper/2103.05606/) (2021)
- [AutoInt: Automatic Integration for Fast Neural Volume Rendering](https://mlsanity.com/paper/2012.01714/) (2020)
- [DIVeR: Real-time and Accurate Neural Radiance Fields with Deterministic Integration for Volume Rendering](https://mlsanity.com/paper/2111.10427/) (2021)
- [CoordX: Accelerating Implicit Neural Representation with a Split MLP Architecture](https://mlsanity.com/paper/2201.12425/) (2022)

## Built on by (most cited first, 16 of 16)

- [Compressed 3D Gaussian Splatting for Accelerated Novel View Synthesis](https://mlsanity.com/paper/2401.02436/) (2023)
- [GS-Hider: Hiding Messages into 3D Gaussian Splatting](https://mlsanity.com/paper/2405.15118/) (2024)
- [Mirror-3DGS: Incorporating Mirror Reflections into 3D Gaussian Splatting](https://mlsanity.com/paper/2404.01168/) (2024)
- [SteerNeRF: Accelerating NeRF Rendering via Smooth Viewpoint Trajectory](https://mlsanity.com/paper/2212.08476/) (2022)
- [Regularize implicit neural representation by itself](https://mlsanity.com/paper/2303.15484/) (2023)
- [CF-NeRF: Camera Parameter Free Neural Radiance Fields with Incremental Learning](https://mlsanity.com/paper/2312.08760/) (2023)
- [Neural radiance fields in the industrial and robotics domain: applications, research opportunities and use cases](https://mlsanity.com/paper/2308.07118/) (2023)
- [Content-Aware Radiance Fields: Aligning Model Complexity with Scene Intricacy Through Learned Bitwidth Quantization](https://mlsanity.com/paper/2410.19483/) (2024)
- [Anatomizing Deep Learning Inference in Web Browsers](https://mlsanity.com/paper/2402.05981/) (2024)
- [HiCoM: Hierarchical Coherent Motion for Streamable Dynamic Scene with 3D Gaussian Splatting](https://mlsanity.com/paper/2411.07541/) (2024)
- [Reconstructive Latent-Space Neural Radiance Fields for Efficient 3D Scene Representations](https://mlsanity.com/paper/2310.17880/) (2023)
- [Dynamic NeRFs for Soccer Scenes](https://mlsanity.com/paper/2309.06802/) (2023)
- [NGP-RT: Fusing Multi-Level Hash Features with Lightweight Attention for Real-Time Novel View Synthesis](https://mlsanity.com/paper/2407.10482/) (2024)
- [Learning Neural Radiance Fields from Multi-View Geometry](https://mlsanity.com/paper/2210.13041/) (2022)
- [Intensity Field Decomposition for Tissue-Guided Neural Tomography](https://mlsanity.com/paper/2411.00900/) (2024)
- [Survey on Controlable Image Synthesis with Deep Learning](https://mlsanity.com/paper/2307.10275/) (2023)

## Abstract (arXiv)

Neural Radiance Fields (NeRF) has emerged as the state-of-the-art method for novel view generation of complex scenes, but is very slow during inference. Recently, there have been multiple works on speeding up NeRF inference, but the state of the art methods for real-time NeRF inference rely on caching the neural network output, which occupies several giga-bytes of disk space that limits their real-world applicability. As caching the neural network of original NeRF network is not feasible, Garbin et al. proposed "FastNeRF" which factorizes the problem into 2 sub-networks - one which depends only on the 3D coordinate of a sample point and one which depends only on the 2D camera viewing direction. Although this factorization enables them to reduce the cache size and perform inference at over 200 frames per second, the memory overhead is still substantial. In this work, we propose SqueezeNeRF, which is more than 60 times memory-efficient than the sparse cache of FastNeRF and is still able to render at more than 190 frames per second on a high spec GPU during inference.