Research map / 3D vision / NeRF and radiance fields
Zip-NeRF: Anti-Aliased Grid-Based Neural Radiance Fields
Jonathan T. Barron, Ben Mildenhall, Dor Verbin, Pratul P. Srinivasan, Peter Hedman
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add multisampling and downweighting versus naive sampling (ablation) | Zip-NeRF · Multiscale 360 (8x) | PSNR: 26.53 -> 32.52 | Improved |
| 6x supersampling instead of multisampling (ablation) | Zip-NeRF · Multiscale 360 (8x) | PSNR: 26.53 -> 25.97 | No clear effect |
| add multisampling (ablation) | Zip-NeRF · Multiscale 360 (8x) | PSNR: 32.12 -> 32.52 | Improved |
| add downweighting (ablation) | Zip-NeRF · Multiscale 360 (8x) | PSNR: 31.67 -> 32.52 | Improved |
| append scale omega as feature (ablation) | Zip-NeRF · Multiscale 360 (1x) | PSNR: 28.23 -> 28.25 | Improved |
| hexagonal multisampling instead of random or unscented (ablation) | Zip-NeRF · Multiscale 360 (8x) | PSNR: 32.03 / 32.46 -> 32.52 | No clear effect |
| add anti-aliased interlevel loss (ablation) | Zip-NeRF · Multiscale 360 (1x) | PSNR: 28.12 -> 28.25 | No clear effect |
| reduce number of NeRF samples with anti-aliased loss versus mip-NeRF 360 loss | Zip-NeRF · Multiscale 360 | PSNR | Improved |
| add normalized weight decay on grid features (ablation) | Zip-NeRF · Multiscale 360 (1x) | PSNR: 27.34 -> 28.25 | Improved |
| normalized weight decay instead of un-normalized weight decay (ablation) | Zip-NeRF · Multiscale 360 (1x) | PSNR: 27.99 -> 28.25 | Improved |
| use large view-dependent MLP instead of iNGP small MLP (ablation) | Zip-NeRF · Multiscale 360 (1x) | PSNR: 27.41 -> 28.25 | Improved |
| combine mip-NeRF 360 with iNGP without anti-aliasing (naive combination) at coarse scale | mip-NeRF 360 + iNGP · Multiscale 360 (8x) | PSNR: 28.42 (iNGP) -> 25.58 | Hurt |
Later papers that compared with this method
33 later accepted papers report ZipNeRF as a baseline in their results tables; 19 of them report a better number on most of the results they share with it.
- Neural Harmonic Textures for High-Quality Primitive Based Neural Reconstruction (ECCV 2026 · better on 2 of 2 shared results)
- Augmented Radiance Field: A General Framework for Enhanced Gaussian Splatting (ICLR 2026 · better on 1 of 1 shared result)
- PPISP: Physically-Plausible Compensation and Control of Photometric Variations in Radiance Field Reconstruction (CVPR 2026 · better on 1 of 1 shared result)
- Spherical Voronoi: Directional Appearance as a Differentiable Partition of the Sphere (CVPR 2026 · better on 1 of 1 shared result)
- Deformable Triangle Splatting: Flexible Primitives for Real-Time Radiance Field Rendering (ECCV 2026 · better on 1 of 1 shared result)
- Fourier Splatting: Generalized Fourier encoded primitives for scalable radiance fields (ECCV 2026 · better on 0 of 1 shared result)
- HoGS: Unified Near and Far Object Reconstruction via Homogeneous Gaussian Splatting (CVPR 2025 · better on 1 of 5 shared results)
- EnvGS: Modeling View-Dependent Appearance with Environment Gaussian (CVPR 2025 · better on 0 of 3 shared results)
25 more not listed.
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- mip-NeRF 360 — Mip-NeRF 360: Unbounded Anti-Aliased Neural Radiance Fields (CVPR 2022) · better on 5 of 5 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- PyNeRF: Pyramidal Neural Radiance Fields (NeurIPS 2023 · similarity 0.94)
- Mip-Grid: Anti-aliased Grid Representations for Neural Radiance Fields (NeurIPS 2023 · similarity 0.94)
- Mip-NeRF: A Multiscale Representation for Anti-Aliasing Neural Radiance Fields (ICCV 2021 · similarity 0.93)
- Rad-NeRF: Ray-decoupled Training of Neural Radiance Field (NeurIPS 2024 · similarity 0.92)
- Neural NeRF Compression (ICML 2024 · similarity 0.90)
- CF-NeRF: Camera Parameter Free Neural Radiance Fields with Incremental Learning (AAAI 2024 · similarity 0.90)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
