Research map / 3D vision / NeRF and radiance fields
Dynamic and editable NeRF
Extends radiance fields to dynamic scenes, editing, large scenes and faster real-time rendering.
88 accepted papers at ICML, NeurIPS, ICLR, CVPR and AAAI (2021–2026) · an approach in NeRF and radiance fields, keyword cluster “radiance fields · fields nerf · nerfs”
State History Benchmarks Tried Baselines Checklist Preprints Who Export
Where it stands
Still moving The best reported result is still improving on 1 of 3 tracked benchmarks.
60 papers in 2023–24 and 13 in 2025–26 (at ICML, NeurIPS, ICLR, CVPR and AAAI); its share of the topic shrank, from 22.6% to 14.4%.
Open questions the data points to
- NeRF synthetic (PSNR): best went from 32.24 (up to 2024) to 34.53 in 2025 — Normal-NeRF: Ambiguity-Robust Normal Estimation for Highly Reflective Scenes (AAAI 2025 · 0 citations)
- Papers report Mip-NeRF on Blender (LPIPS) anywhere from 0.043 to 0.0675: comparisons against it depend on whose number you take.
- Papers report NeRF on DTU, 3-view (PSNR) anywhere from 6.68 to 9.9: comparisons against it depend on whose number you take.
No longer reported by papers here: LLFF (PSNR, last in 2024); Tanks and Temples (PSNR, last in 2024).
How this is decided
A benchmark is tracked when at least four papers of this approach report their own method on the same dataset and metric in at least two different years. Its record is the best number reported up to 2024 compared with the best reported in 2025–26: still improving means at least 1% better, flat means at most 0.2% better with at least two papers from 2025–26 trying. Numbers on a different scale from the rest (a fraction among percentages) and far outliers are left out. Results come from different papers with different setups, so they are signals, not head-to-head comparisons. The share of papers is counted at ICML, NeurIPS, ICLR, CVPR and AAAI.
How it got here
Most cited paper of each year
- 2021 · D-NeRF: Neural Radiance Fields for Dynamic Scenes (CVPR 2021 · 1,356 citations)
- 2022 · Mip-NeRF 360: Unbounded Anti-Aliased Neural Radiance Fields (CVPR 2022 · 1,901 citations)
- 2023 · SPIn-NeRF: Multiview Segmentation and Perceptual Inpainting With Neural Radiance Fields (CVPR 2023 · 139 citations)
- 2024 · NeRF On-the-go: Exploiting Uncertainty for Distractor-free NeRFs in the Wild (CVPR 2024 · 48 citations)
- 2025 · AE-NeRF: Augmenting Event-Based Neural Radiance Fields for Non-ideal Conditions and Larger Scenes (AAAI 2025 · 11 citations)
- 2026 · To View Transform or Not to View Transform: NeRF-based Pre-training Perspective (ICLR 2026 · 0 citations)
Earliest papers here
- D-NeRF: Neural Radiance Fields for Dynamic Scenes (CVPR 2021 · 1,356 citations)
- Mip-NeRF 360: Unbounded Anti-Aliased Neural Radiance Fields (CVPR 2022 · 1,901 citations)
- Depth-Supervised NeRF: Fewer Views and Faster Training for Free (CVPR 2022 · 825 citations)
Benchmark records
The best number each year that beat every earlier one, among papers of this approach reporting their own method. Numbers are extracted by AI from each paper's text; check the paper before citing.
LLFF · PSNR (higher is better)
8 papers, 2022–2024 · not reported since 2024
| Year | Best so far | Set by |
|---|---|---|
| 2022 | 26.86 | Mip-NeRF 360: Unbounded Anti-Aliased Neural Radiance Fields (CVPR 2022 · 1,901 citations) |
NeRF synthetic · PSNR (higher is better)
5 papers, 2023–2025 · still improving
| Year | Best so far | Set by |
|---|---|---|
| 2023 | 32.24 | Masked Wavelet Representation for Compact Neural Radiance Fields (CVPR 2023 · 56 citations) |
| 2025 | 34.53 | Normal-NeRF: Ambiguity-Robust Normal Estimation for Highly Reflective Scenes (AAAI 2025 · 0 citations) |
Tanks and Temples · PSNR (higher is better)
5 papers, 2023–2024 · not reported since 2024
| Year | Best so far | Set by |
|---|---|---|
| 2023 | 36.24 | Cross-Guided Optimization of Radiance Fields With Multi-View Image Super-Resolution for High-Resolution Novel View Synthesis (CVPR 2023 · 19 citations) |
What has been tried
81 papers here report 846 measured changes (a component added, removed or swapped, compared with a baseline): helped 74%, no clear effect 4%, hurt 19%, mixed 3%. Papers mostly publish changes that helped, so "helped" is over-represented.
Ablations in recent papers
| What was changed | Measured on | Result | Effect | Paper |
|---|---|---|---|---|
| use dilated patch sampling with dilation rate 4 instead of rates 1, 2, 8, or 16 | Patio-High · PSNR | 17.82, 20.70, 21.41, 18.22, 16.08 | Mixed | NeRF On-the-go: Exploiting Uncertainty for Distractor-free NeRFs in the Wild (CVPR 2024 · 48 citations) |
| disable 2D and 3D context models | Synthetic-NeRF · BD-rate | +43.7%/+78.8% | Hurt | How Far Can We Compress Instant-NGP-Based NeRF? (CVPR 2024 · 34 citations) |
| jointly optimize attention maps instead of freezing them | ScanNet · mIoU | 56.45 -> 57.19 | Helped | GP-NeRF: Generalized Perception NeRF for Context-Aware 3D Scene Understanding (CVPR 2024 · 27 citations) |
| Add event supervision Lev | Ev-DeblurCDAVIS Figures · PSNR | 27.55 -> 28.24 | Helped | Mitigating Motion Blur in Neural Radiance Fields with Events and Frames (CVPR 2024 · 27 citations) |
| add DWT frequency-domain embedding instead of DFT | LLFF + Blender, 16 bits · bit accuracy (%) | 71.06 -> 95.67 | Helped | WateRF: Robust Watermarks in Radiance Fields for Protection of Copyrights (CVPR 2024 · 27 citations) |
| add rendering-based optimization to structure-based initialization | Replica room0 and office0 · median translation/rotation error | 0.030 / 0.79 -> 0.005 / 0.29; 0.082 / 1.38 -> 0.006 / 0.56 | Helped | PNeRFLoc: Visual Localization with Point-Based Neural Radiance Fields (AAAI 2024 · 22 citations) |
| Optimize the test-view time offset | Unsynchronized Plenoptic Video Dataset test view · PSNR | 30.29 -> 30.53 | Helped | Sync-NeRF: Generalizing Dynamic NeRFs to Unsynchronized Videos (AAAI 2024 · 14 citations) |
| Increase iterations ξ from 600 to 900 | NeRFBuster · ΔR | 17.8029 -> 14.8730 | Helped | CF-NeRF: Camera Parameter Free Neural Radiance Fields with Incremental Learning (AAAI 2024 · 13 citations) |
| add cross-view volume integration (VI) | DTU 3-view · PSNR | 18.74 -> 19.21 | Helped | ColNeRF: Collaboration for Generalizable Sparse Input Neural Radiance Field (AAAI 2024 · 12 citations) |
| add pose correction network ψ | Proposed synthetic scenes, easy and hard · PSNR | 22.83 -> 24.28; 22.79 -> 24.23 | Helped | AE-NeRF: Augmenting Event-Based Neural Radiance Fields for Non-ideal Conditions and Larger Scenes (AAAI 2025 · 11 citations) |
| add edge loss | Object-pair evaluation · Appearance detail | Helped | NeRF Analogies: Example-Based Visual Attribute Transfer for NeRFs (CVPR 2024 · 11 citations) | |
| Add irradiance fields to Mip-NeRF | Evaluation scenes · D-RMSE | 1.0576 -> 0.8565 | Helped | Pano-NeRF: Synthesizing High Dynamic Range Novel Views with Geometry from Sparse Low Dynamic Range Panoramic Images (AAAI 2024 · 9 citations) |
Baseline numbers that disagree
The same baseline on the same benchmark and metric, reported more than 5% apart by different papers here. Training budgets, data splits and reimplementations differ, so state which number you compare with.
| Baseline | Benchmark | Metric | Reported | Papers |
|---|---|---|---|---|
| Mip-NeRF | Blender | LPIPS | 0.043 – 0.0675 | 2 papers, e.g. CVPR 2022 and CVPR 2023 |
| NeRF | DTU, 3-view | PSNR | 6.68 – 9.9 | 2 papers, e.g. CVPR 2024 and CVPR 2022 |
What reviewers will expect
Datasets and baselines named in the 9 papers of this approach from 2025–2026 that have a full-text summary. Leaving out the common ones is a frequent review question.
Datasets
- LLFF · 2 of 9 papers
Baselines
- NeRF · 2 of 9 papers
Recent preprints
Loading recent arXiv preprints…
Who publishes on it
Institutions on the most papers of this approach since 2023 (of 73 papers; from OpenAlex and the papers' affiliations).
- Zhejiang University · 6 papers
- Hong Kong University of Science and Technology · 4 papers
- Nanyang Technological University · 4 papers
- Meta · 4 papers
- Harbin Institute of Technology · 3 papers
- Peking University · 3 papers
- Yonsei University · 3 papers
- Korea University · 3 papers
- The Chinese University of Hong Kong · 3 papers
- Hong Kong Baptist University · 2 papers
Use it
Download BibTeX of 40 key papersOpen the topic in the interactive map
Get alerts. Describe your idea in scime atlas and save it (free scime account): you get a weekly email when new preprints land close to it.
Related approaches
Same keyword cluster
- Sparse-input NeRF regularization (106 papers)
- Single-view and streaming radiance fields (49 papers)
- NeRF for scene decomposition and understanding (30 papers)
Elsewhere in the topic
- Inverse rendering with neural fields (110 papers)
- Few-shot neural rendering architectures (53 papers)
- 3D-aware generative radiance fields (20 papers)
Benchmark numbers, ablations and baseline numbers are extracted by AI from each paper's full text and can be wrong; the paper is the authority. Data: results index of 2026-11-02, citation counts from OpenAlex (2026-10-09).