Research map / 3D vision / Face and shape
NeuRBF: A Neural Fields Representation with Adaptive Radial Basis Functions
Zhang Chen, Zhong Li, Liangchen Song, Lele Chen, Jingyi Yu, Junsong Yuan, Yi Xu
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 adaptive RBFs (ablation) | NeuRBF · DIV2K image fitting | PSNR: 42.37 -> 51.53 | Improved |
| add multi-frequency sinusoidal composition on radial bases (ablation) | NeuRBF · DIV2K image fitting | PSNR: 48.19 -> 51.53 | Improved |
| add multi-frequency sinusoidal composition on both radial bases and features (ablation) | NeuRBF · DIV2K image fitting | PSNR: 43.81 -> 51.53 | Improved |
| use multi-frequency sinusoidal composition instead of positional encoding (ablation) | NeuRBF · DIV2K image fitting | PSNR: 43.72 -> 51.53 | Improved |
| use multi-frequency sinusoidal composition instead of SIREN sinusoid activation (ablation) | NeuRBF · DIV2K image fitting | PSNR: 45.98 -> 51.53 | Improved |
| add adaptive RBFs on SDF (ablation) | NeuRBF · 3D SDF NAE | NAE: 5.70 -> 4.93 | Improved |
| add sinusoidal composition on SDF IoU (ablation) | NeuRBF · 3D SDF IoU | IoU: 0.9995 in all MSC variants | No clear effect |
| finetune RBF position and shape parameters with small learning rates | NeuRBF · image fitting | PSNR | No clear effect |
| finetune RBF parameters with large learning rates | NeuRBF · image fitting | PSNR | Hurt |
| use weighted K-Means initialization instead of grid or random initialization | NeuRBF · image fitting | fitting accuracy | Improved |
| make RBF kernel position and interpolation adaptive rather than grid linear | NeuRBF · image, SDF, NeRF | accuracy | Improved |
| add sinusoidal composition in NeRF task | NeuRBF · Synthetic NeRF | PSNR | No clear effect |
Later papers that compared with this method
5 later accepted papers report NeuRBF as a baseline in their results tables; 5 of them report a better number on most of the results they share with it.
- SASNet: Spatially-Adaptive Sinusoidal Networks for INRs (CVPR 2026 · better on 3 of 3 shared results)
- GaINeR: Geometry-Aware Implicit Neural Representation for Image Editing (ECCV 2026 · better on 2 of 2 shared results)
- Large Images Are Gaussians: High-Quality Large Image Representation with Levels of 2D Gaussian Splatting (AAAI 2025 · better on 1 of 1 shared result)
- Grounding and Enhancing Grid-based Models for Neural Fields (CVPR 2024 · better on 4 of 5 shared results)
- Spec-Gaussian: Anisotropic View-Dependent Appearance for 3D Gaussian Splatting (NeurIPS 2024 · better on 1 of 1 shared result)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 3 on most shared results.
- BACON — BACON: Band-Limited Coordinate Networks for Multiscale Scene Representation (CVPR 2022) · better on 1 of 1 shared result
- Mip-NeRF 360 — Mip-NeRF 360: Unbounded Anti-Aliased Neural Radiance Fields (CVPR 2022) · better on 0 of 1 shared result
- MINER — MINER: Multiscale Implicit Neural Representation (ECCV 2022) · better on 1 of 1 shared result
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
- ResFields: Residual Neural Fields for Spatiotemporal Signals (ICLR 2024 · similarity 0.87)
- Learning Robust Generalizable Radiance Field with Visibility and Feature Augmented Point Representation (ICLR 2024 · similarity 0.87)
- Learning Spatially Collaged Fourier Bases for Implicit Neural Representation (AAAI 2024 · similarity 0.86)
- ProbeSDF: Light Field Probes For Neural Surface Reconstruction (CVPR 2025 · similarity 0.86)
- Spatial Annealing for Efficient Few-shot Neural Rendering (AAAI 2025 · similarity 0.85)
- BANF: Band-Limited Neural Fields for Levels of Detail Reconstruction (CVPR 2024 · similarity 0.85)
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.
